Students of the game must accept the following basic, if troubling, precepts:

• Despite the hard work and admirable precision of baseball’s official scorers and official statisticians, there are an uncounted number of mistakes in baseball’s historical record.
• Despite baseball’s long obsession with its numbers, specific details about its history are not always known; in some cases, they are not even knowable.

Baseball’s “figure filberts” (as the ink-stained wretches in the press box used to refer to SABR members and their antecedents) and historians have done yeoman work in filling in missing details and correcting erroneous beliefs about the patrimony of our National Pastime. 

IMMORTAL PLAYERS, NOT ETERNAL RECORDS

A look at the first players enshrined in baseball’s Hall of Fame in 1936 shows that every one of this immortal quintet has seen changes to the numbers presented as their “official statistics” since SABR was formed four decades ago.

• TY COBB: At-bats changed from 11,437 to 11,434; Hits from 4,191 to 4,189; Batting average from .367 to .366; RBIs from 1,933 to 1,938; Runs from 2,245 to 2,246; Stolen bases from 892  to 897.
• WALTER JOHNSON: Wins changed from 413 to 417; Losses from 277 to 279; Complete games from 532 to 531; Innings pitched from 5,923.2 to 5,914.1; Hits allowed from 4,925 to 4,913; Walks from 1,405 to 1,363; Strikeouts from 3,499 to 3,509.
• CHRISTY MATHEWSON: Wins changed from 367 to 373; Losses from 186 to 188; Shutouts from 77 to 79; Games from 634 to 636; Complete games from 434 to 435; Innings pitched from 4,777.1 to 4,788.2; Hits allowed from 4,216 to 4,219; Walks from 838 to 848; Strikeouts from 2,502 to 2,507.
• BABE RUTH: Games changed from 2,502 to 2,503; RBIs from 2,217 to 2,213; Walks from 2,056 to 2,062.
• HONUS WAGNER: Games changed from 2,787 to 2,794; At-bats from 10,430 to 10,439; Hits from 3,415 to 3,420; Batting average from .327 to .328; Doubles from 640 to 643; RBIs from 1,732 to 1,733; Runs from 1,736 to 1,739; Stolen bases from 722 to 723.

The records of several other immortals who reached Cooperstown in the next three elections have been changed as well.

• GROVER ALEXANDER: Wins changed from 374 to 373; Shutouts from 88 to 90; Games started from 598 to 600; Complete games from 439 to 437; Innings pitched from 5,189.1 to 5,190; walks from 953 to 951; strikeouts from 2,199 to 2,198.
• EDDIE COLLINS: Games changed from 2,825 to 2,826; Hits from 3,310 to 3,315; Doubles from 437 to 438; RBIs from 1,299 to 1,300; Runs from 1,817 to 1,821; Stolen bases from 743 to 741.
• LOU GEHRIG: Games changed from 2,163 to 2,164; Doubles from 535 to 534; Triples from 162 to 163; RBIs from 1,990 to 1,995.
• GEORGE SISLER: Games changed from 2,054 to 2,055; At-bats from 8,262 to 8,267; Hits from 2,811 to 2,812; Triples from 165 to 164; Home runs from 99 to 102; Runs from 1,283 to 1,284.
• TRIS SPEAKER: At-bats from 10,205 to 10,195; Batting average from.344 to .345; Doubles from 793 to 792; Triples from 224 to 222; Home runs from 115 to 117; RBIs from 1,527 to 1,529; Runs from 1,881 to 1,882.
• CY YOUNG: Wins changed from 509 to 511; Won-Lost percentage from .617 to .618; Games started from 816 to 815; Complete games from 750 to 749; Shutouts from 77 to 76.

Some other changes are worth noting, including two of much more recent vintage.

• ROGER MARIS & MICKEY MANTLE: In 1961, Roger Maris of the Yankees was mistakenly credited with a run batted in. The change dropped his season total from 142 to 141 and meant that Maris was no longer the American League’s sole RBI leader that year, but instead was tied with Baltimore’s Jim Gentile. Maris had also been thought to have been tied with teammate Mickey Mantle for the AL lead in runs scored that same year. But  Mantle had been incorrectly awarded a run,  leaving him with 131 runs after adjustment, one shy of Maris’s league-leading 132. The Mantle & Maris changes were not made by MLB until 2008 even though they had been known about for more than a decade.
• TED WILLIAMS: Two additional walks found for Ted Williams in 1941 raised his league-leading total from 145 to 147. The two 1941 walks, plus one found in 1951, raised his career total from 2,018 to 2,021.
• OLD HOSS RADBOURN: Radbourn holds the record for most wins in a season with 59 in 1884. In 1971, his win total was thought to be 60. However, because an uncredited win for Radbourn was discovered in 1889 (giving him 20 that year rather than 19), his 309 career wins remain unchanged. Radbourn’s career losses total did change from 195 to 194, though.
• HACK WILSON: Hack Wilson holds the single-season major league record for most runs batted in. Wilson was thought to have driven home 190 for  the 1930 Chicago Cubs, but an additional RBI  discovered in the second game of a July 28 doubleheader that year has upped that total to 191.

Table 1: Official Statistical Categories Added Since the Start of the SABR Era in 1971

(Click image to enlarge.)

If you think that list is impressive, there’s a lot more to come. Internet and Web access, combined with the digitization of source material and Retrosheet’s ground-breaking publication of play-by-play accounts, box scores, and game logs, has greatly accelerated the process of researching and verifying mistakes in the official records. Indefatigable SABR members like Trent McCotter and Herm Krabbenhoft are investigating many other apparent discrepancies. For the specifics of Krabbenhoft’s current campaign, which affects Lou Gehrig, Hank Greenberg, and the AL single-season RBI record, check out his amazingly scrupulous articles in this BRJ and the previous volume. (Note: Links to more of Krabbenhoft’s statistical research can be found on his SABR author page.)

While a high school student, McCotter decided to begin a game-by-game review of Ty Cobb’s career. He found mistakes in five or more games in the “Georgia Peach’s” official batting stats in important categories like at-bats, runs, walks, stolen bases, and RBIs, mostly from 1905–1917. (Some of these errors compensate for others, and other errors may well be discovered later in Cobb’s career, so the net effects are not known yet.)

McCotter believes the number of mistakes in Cobb’s RBI numbers is probably much higher, but this is much harder to verify. He also found 31 cases where Cobb’s fielding position was listed incorrectly in the official daily stats, mostly the wrong place in the outfield—they wouldn’t show up in the seasonal totals that reported only “Games in the Outfield.”

AMATEUR VS. PROFESSIONAL

McCotter and Krabbenhoft are both amateurs and SABR members—indeed, virtually all of the important work being done in this field is by amateurs. While more prolific than most, they are otherwise typical of the type of amateurs that SABR has historically helped and also produced. Professionals occasionally assist these dogged researchers, and professional baseball writers, historians, and statisticians certainly benefit from the fruit of these amateurs’ labors. Far more often, though, SABR’s corps of expert amateurs is called upon to help the pros when they’re looking to get things right.

Merriam-Webster.com gives three definitions for the word amateur: devotee, admirer; one who engages in a pursuit, study, science, or sport as a pastime rather than as a profession; one lacking in experience and competence in an art or science.

The devoted sleuths who spend their nights, weekends, holidays, and vacations pursuing these issues definitely qualify as amateurs according to sense No. 1 above. Definition No. 2 is also spot-on, as most are not getting paid to do the legwork and don’t expect their work to have any monetary value.

Definition No. 3 is sometimes unfairly used to disparage this corps of researchers, hurled to draw the focus away from the merits of their work.

PERSPICACITY PLUS PERSPIRATION

Lest anyone think that this efflorescence of records-altering historiography is a function of a new generation of statistical Maoists bent on an agate-type cultural revolution, consider the wisdom of SABR’s founders four decades ago. SABR’s Baseball Records Committee was formed by the young organization’s directors in 1975, with the mandate “to establish an accurate set of records for organized baseball.” A tall order, indeed, but one that SABR has subsequently strived mightily to fulfill.

In 1993, Lyle Spatz (then and now chair of SABR’s Baseball Records Committee) wrote:

Over the years, researchers have uncovered many errors and inconsistencies in baseball’s historical record. These errors are of two major types: (1) those errors that deal with accomplishments, positive or negative, that have been accepted as major league or individual league records: the most, the highest, the fewest, the lowest, etc., or (2) errors in the yearly and lifetime records of an individual, or a team…

For the overwhelming mass of individual, team, and league statistics, these publications are in agreement and are correct. However, baseball’s record-keeping has not always been so meticulous as it is today. This is particularly true of the game’s early years. There are numerous cases of omissions, transpositions, double-counting, addition errors, and… illegibility… that occurred at the original source (in the accounting of individual games), or in the league’s accounting at year’s end.

Other mistakes have come about as a result of the work done in the compilation of the encyclopedias and record books, some as a result of putting together the Information Concepts Corporation [sic] (ICI) data in the 1960s. ICI data were collected for those years and categories for which official sheets were not available, and together with those official sheets formed the data base for the encyclopedias. Assembling these volumes is such an enormous undertaking, that it is not surprising that errors occurred. This is in no way a criticism of those responsible for these publications; that there are so few errors is to their everlasting credit. We cannot emphasize that strongly enough. Those of us who do baseball research understand how great a debt we owe them.

The players affected by the changes that we have made range from fringe players to Hall-of-Famers. In some cases we have dealt with icons of the game, men such as Walter Johnson, Tris Speaker, and Lou Gehrig. Their accomplishments are legendary, and certain numbers associated with them are so imbedded in the memory of baseball fans that some think it sacrilegious to alter them. Those of us who attempt to do so are accused of being “fanatical crusaders,” more obsessed with statistics than with the “romance” of the game. Nothing could be further from the truth. Each person who has contributed to this study has loved the game since childhood and continues to do so. It is inconceivable that anyone who didn’t love baseball would devote as much time to studying it as we do. But as Aristotle said of his mentor, “Plato is dear to me, but dearer still is truth.” That is basically what we are, seekers of truth, not for the purpose of inflating or deflating the accomplishments of one player at the expense of another, but to ensure that each player’s accomplishments are accounted for as accurately as possible.

Thus, the purpose… is not one of revolution, but rather of evolution. We want to make the historic record a little more accurate, and a little more consistent.

The people responsible for the research… have pored over old newspapers, guides, official records, and ICI records. They have looked at miles of microfilm, from a variety of sources, searching through box scores and comparing game accounts. In all cases, they adhered to the scoring rules and customs of the day. As previously stated, only in very few cases did they find what seemed to be a demonstrable error…

We fail to see how clinging to a number that is obviously wrong adds to the “romance” of the game, or how accepting a correct one detracts from it…. Therefore we offer these corrections and reconciliations in a spirit of collegiality and cooperation. We trust that they will be received in the same way…

THE LEGACY OF SABR

SABR was founded two years after the momentous debut of the Macmillan Baseball Encyclopedia, made possible by the Herculean efforts of David Neft and Information Concepts Incorporated (ICI). Neft would become SABR member No. 74 in November 1971.

The table at right details just how much baseball’s statistical landscape has changed since that watershed year. Everyone named in it is a stalwart SABR member, and SABR provided an invaluable framework for their pioneering work. While you might scoff at the value of knowing how many sacrifice bunts individual pitchers allowed in 1912, your understanding of the careers of all-time greats like Cy Young, Christy Mathewson, Walter Johnson, and Pete Alexander is framed by SABR members’ contributions.

Moreover, it is not an overstatement to say that the addition by Pete Palmer (SABR member No. 27) of On-Base Percentage to the roster of official batting statistics would eventually alter the way the modern game is played.

The appointment of baseball historian, SABR confrere, and former Total Baseball encyclopedia co-editor John Thorn as the second official historian for Major League Baseball in March 2011 will also likely affect the landscape in the future. MLB created its official historian position in 1999 and appointed veteran Chicago sportswriter and SABR member Jerome Holtzman to fill it.

Holtzman had revamped the Grand Old Game’s canon of pitching statistics in 1969 when he persuaded baseball to officially adopt his Save statistic—which would ultimately completely rearrange pitcher usage and the composition of pitching staffs. Other than a paradoxical decision to reverse longstanding practice by counting walks as outs in 1876 and walks as hits in 1887, Holtzman’s reign until his death in 2008 did not affect the status quo ante.

OBP—along with Palmer’s later analytical development of OPS (On-Base Plus Slugging)—fundamentally rewired the way we perceive baseball offense. The analytical innovations of seminal writer and analyst Bill James (SABR member No. 407, who became famous in the early 1980s for his revolutionary and bestselling Baseball Abstract), combined with the masterworks of Messrs. Palmer and Thorn (The Hidden Game of Baseball and Total Baseball) later that decade, popularized sabermetrics—a neologism coined by James—and laid the foundation for what would become known as “Moneyball”.

During the four decades of the SABR era, the Society and many of its members—titans in the fields of baseball history, research, and analysis—have truly rewritten baseball’s legendary “book.”

GARY GILLETTE is co-chair of the SABR Ballparks Committee and a member of the SABR Board of Directors. He has written or edited numerous baseball books, including “Big League Ballparks” and the “ESPN Baseball Encyclopedia”. Along with Stuart Shea, he is currently blogging the fabulous 1961 season at 61at50.com.

LYLE SPATZ has been a member of the Society for American Baseball Research since 1973 and chairman of the Baseball Records Committee since 1991.

THE PROBLEM

The Howe News Bureau served as the official statistician for the American League from 1912 through 1972. According to the official Day-By-Day (DBD) baseball records (ledgers) compiled by Howe, after having played in his final game in The Big Show on April 30, 1939, Lou Gehrig had accumulated a total of 1,990 RBIs.1 This sum can be calculated by adding up the RBIs credited to him on the official DBD sheets over the course of his career. (See Chart 1.)

Unfortunately, as has been demonstrated innumerable times, there are errors in the official DBD records.2 So, it’s not inconceivable that Gehrig’s official RBI record could be compromised with errors—which, when discovered and corrected, could be responsible for the differing RBI totals given for Gehrig. Each of the numbers—1,990, 1,991, 1,995, and 1,996—appears as Gehrig’s lifetime RBI total in at least one edition of The Elias Book of Baseball Records, published annually since 1972 by the Elias Sports Bureau, the official statistician for Major League Baseball since 1987 (and for the National League since 1923). Significantly, every edition of The Elias Book of Baseball Records has been edited by longtime SABR member Seymour Siwoff.

Here is the chronology for Gehrig’s lifetime RBI numbers presented in the various editions of The Elias Book of Baseball Records since SABR’s founding:

• 1,991 RBIs. In the 1973 edition (page 25), Gehrig is listed as being the holder of the American League record for most lifetime RBIs with 1,991. The 1,991 lifetime RBI number for Gehrig is also listed in each annual edition of The Little Red Book of Baseball, 1940 through 1963. This publication, also put out by Elias, was the direct precursor of The Elias Book of Baseball Records and was edited by Siwoff 1953–1971.
• 1,990 RBIs. Beginning in 1974 and continuing through 1995, The Elias Book of Baseball Records lists Gehrig with 1,990 RBIs.
• 1,995 RBIs. Then from 1996 through 2006, Gehrig‘s total is listed as 1,995 RBIs.
• 1,996 RBIs. The Elias Book of Baseball Records from 2007 through 2011 gives Gehrig 1,996 RBIs.

Thus it appears that Siwoff and his colleagues at Elias have discovered some RBI errors in the official records and made appropriate changes to Gehrig’s lifetime RBI number. Now, Gehrig’s lifetime RBIs should be equal to the sum of the RBIs he earned in each season, right? This raises the following question: for which individual seasons have RBI errors been discovered in Gehrig’s official DBD records?

While The Elias Book of Baseball Records does not include season-by-season RBI statistics for individual players, it does contain a list of the players who led their league in RBIs for each season from 1920 forward. (1920 is the starting season because 1920 was the first year that RBIs were recorded on the official scoresheets; there are no official RBI records prior to 1920.) Since Gehrig led the AL in RBIs in five seasons (1927, 1928, 1930, 1931, and 1934), one can begin by checking the various editions of The Elias Book of Baseball Records to hunt for changes in Gehrig’s RBI numbers for those five years.

As it turns out, four of the five years show no change. The 1927, 1930, 1931, and 1934 seasons show 175, 174, 184, and 165 RBIs, respectively, suggesting that Elias found no net RBI errors in the official records for those years. But for 1928, Elias has twice made changes in Gehrig’s league-leading RBI figure:

• In The Little Red Book of Baseball from 1932 through 1971, and The Elias Book of Baseball Records from 1972 through 2006, Lou Gehrig and Babe Ruth are listed as the co-leaders for most RBIs in the AL with 142. (The 1932 edition was the first to present the annual leaders in RBIs.)
• In the 2007 edition of The Elias Book of Baseball Records, Gehrig becomes the sole AL leader for 1928 with 145 RBIs.
• Then, beginning with the 2008 edition, Gehrig gains one more, to be listed as the sole leader in RBIs for the AL in 1928 with 146.

While Elias has presented four different values for Gehrig’s career total RBIs and three different figures for the 1928 campaign, they provide no explanation for the changes. Such information is exceedingly important to the research community, and its availability would allow others the opportunity to independently verify or refute the changes/corrections made by Elias.

Taking into account only the information presented for Gehrig in The Elias Book of Baseball Records since 2008—1,996 career RBIs and 146 RBIs in the 1928 season—compared with his former totals of 1,990 and 142, this adds six RBIs to his career and four to his 1928 total. Therefore, Elias must have also found a net increase of two RBIs in Gehrig’s other major league seasons. The question is which?

Since Seymour Siwoff is a good friend of mine, I asked him to provide this information. However, as is his right, Seymour chose to not provide me with any details of the changes Elias authorized in Gehrig’s RBI record.3 So, I had to come up with another way to ascertain the correct numbers for Gehrig’s RBI record.

THE SEARCH

The most rigorous approach to ascertaining the correct value for Gehrig’s lifetime RBIs is to obtain the complete details for each run scored by the Yankees in all 2,164 games that Gehrig played. That’s a monumental task, but a doable task. And the task is made much more doable because over the past 40 years SABR members have already corrected some errors in the official DBD records, Gehrig’s included.

Supplemental material: Click here to read Herm Krabbenhoft’s supporting evidence for the correction of errors in the official RBI record of Lou Gehrig

Sidebar: “Lou Gehrig’s Mythical 1,991 Lifetime RBIs”, by Herm Krabbenhoft

For example, in the 1992 version of the Franklin Big League Baseball Encyclopedia (the baseball encyclopedia formatted in a “hand-held calculator”), Gehrig’s batting record shows 1,995 RBIs for his career—i.e., five RBIs more than that given in his official DBD records. Franklin shows Gehrig with 112 RBI for the 1926 campaign, five more than the official records. When one enters “PETE PALMER” in Franklin, the following text appears: “PETE PALMER PROVIDED THE DATABASE AND, WITH THE SABR, THE RESEARCH.” Pete Palmer, a longtime SABR member, credits Ray Gonzalez and Bob Davids, two of SABR’s founders, with contributing research to support the changes he made in Gehrig’s RBI record.

The revised RBI information for Gehrig’s 1926 season (112 RBIs) and his career (1,995 RBIs) is also included in the third edition (1993) of Total Baseball edited by Palmer and fellow SABR member John Thorn. The first two editions of Total Baseball (published in 1989 and 1991) present 107 RBIs for Gehrig’s 1926 season and 1,990 RBIs for his career (the same values provided by Gehrig’s official DBD records). But the fourth through the eighth editions of Total Baseball (1995–2004) each list Gehrig with 112 RBIs in 1926 and 1,995 RBIs for his career.

Many other publications have adopted Palmer’s changes to Gehrig’s RBI record. (See Chart 2.)

Several prominent baseball websites also present Gehrig’s career batting statistics. As of this writing (1 September 2011), each of these websites shows Gehrig with 1,995 RBIs for his ML career: SABR.org, Baseball-Reference.com, Retrosheet.org, Baseball-Almanac.com, MLB.com, Baseballhall.org.

Palmer’s discovery of five additional RBIs for Lou Gehrig in 1926 has clearly reverberated to numerous presentations of Gehrig’s batting record, but that still does not answer our question: which specific games in 1926 had missing RBIs in the official DBD records? Pete was pleased to provide the following dates4:

Pete also included this advisory statement: “But remember, there are a lot of RBI errors, a lot more than we thought at the time the Gehrig corrections [were] made, so that the current stats have many more mistakes.”5

Proceeding to the game accounts provided in various relevant newspapers, I determined unequivocally the complete details for each run the Yankees scored in those three games. (See the Supplemental Material on the SABR website for all the details.) For the second game on 25 June 1926, from the account in the New York Times, it could be interpreted that Gehrig batted in two runs (not one run as given in the official DBD records). However, according to the more detailed accounts given in the Boston Globe and the New York World, it is unmistakable that Gehrig had just one RBI. Thus, the official record is correct for Gehrig having one RBI in the 25 June game. For the other two games, the newspaper accounts incontrovertibly show that Gehrig had four RBIs (not one) in the second game on 19 September 1926, and one RBI (not zero) in the first game on 25 September 1926. So at this point Gehrig actually had four more RBIs rather than five in those three games.

But that’s not the complete RBI story for Gehrig’s 1926 season. Pete himself indicated in his comment that there are more errors to be found. Indeed, in 2005, SABR member Trent McCotter discovered RBI errors in the official record for Babe Ruth for 1926. Trent reported his findings in the February 2005 issue of SABR’s Baseball Records Committee Newsletter.6 He subsequently found, in collaboration with Pete Palmer and fellow SABR member Keith Carlson, that there were RBI errors affecting more than 200 player-games in the 1926 season—the common denominator being that they were all from AL games played in St. Louis between 7 August and 26 September.7 During this period, the official scorer erroneously entered the number of earned-runs-scored rather than number of runs-batted-in in the column designated for RBIs; runs-batted-in stats were not entered at all. Trent, Pete, and Keith found three more 1926 games with Gehrig RBI errors:

From the game accounts in relevant newspapers, I determined unequivocally the complete details for each run the Yankees scored in those three games, corroborating the findings of Palmer, McCotter, and Carlson. Thus, Gehrig actually had one RBI (not two) in the second game on 25 September 1926, zero RBIs (not one) in the first game on 26 September 1926, and two RBIs (not one) in the second game on 26 September 1926. So, Gehrig actually had one fewer RBI in these three games than indicated in the official records. And based upon my examination of all six questioned games discussed so far, Gehrig has a net gain of three RBIs over the official DBD records.

But are there any other 1926 games with RBI errors involving Gehrig? I contacted longtime SABR member Dave Smith (founder and President of Retrosheet) and asked him if he would be able to please provide the as-yet-unpublished and un-proofed Retrosheet Play-By-Play (PBP) Accounts for 1926. Dave graciously did, emphasizing that the PBP accounts were not yet proofed, there are some games for which Retrosheet does not yet have PBP accounts, and for some of the games the PBP accounts are incomplete (for example, only up through the top of the sixth inning).8

Armed with the Retrosheet PBP information, I assembled the exact details for each run scored by the 1926 Yankees, including three key facts about each one:

• (a) the player who scored the run;
• (b) the run-scoring event (e.g., 2-RBI single, 1-RBI groundout, 1-RBI batter safe on error, 0-RBI batter safe on error, 1-RBI bases-loaded walk, 0-RBI balk, etc.);
• (c) the player who completed his at bat during the run-scoring event.

For the 34 games for which Retrosheet does not yet have PBP accounts and for the 13 games for which the Retrosheet PBP accounts are not yet complete I relied on accounts given in relevant newspapers.9

I discovered four more games with relevant errors:

So, overall for the 1926 season, there are nine games with errors in Lou Gehrig’s official RBI record. Lou Gehrig actually had 110 RBIs, not 107, a net increase of three RBIs.

Let’s return now to the 1928 season—for which Siwoff lists Gehrig with 146 RBIs, four more than the official DBD records. The only other publication of which I’m aware that shows Gehrig with 146 RBIs in 1928 is the 2011 edition of the New York Yankees Media Guide; all of the other publications and websites mentioned have 142.

Dave Smith again graciously provided the unpublished (and unproofed) Retrosheet PBP accounts for 142 of the 154 1928 Yankees games.10 Again I examined newspaper game accounts to fill in missing details and compared the apparent RBIs in the PBP data with the official DBD records—I found that there are seven games in which there are discrepancies in the RBIs credited to Gehrig; see the Supplemental Material for the details.11 Indeed, examination of multiple independent newspaper accounts for each of these seven games reveals that in each instance the official DBD records are wrong:

This matches the official DBD records for these players. But it does not match the text account given in The (New York) Evening World. Here is the first inning:

Combs doubled. Warner threw out Meusel. Combs remained at second. Ruth tripled, scoring Combs. Tavener fumbled Gehrig’s grounder, Ruth scoring. Lazzeri hit into a double play, Gehringer to Tavener to McManus.

Likewise, The World has the following accounts to consider:

First Inning: “Combs doubled to left centre to put the team in motion and, after Meusel had expired, scored on a triple by Ruth, who scored on a boot by Tavener. 

Seventh Inning: The Yanks pulled up close in the seventh when they knocked Van Gilder out and scored three runs on a pass to Ruth, a double by Gehrig and singles by Lazzeri, Koenig, and Robertson, the last a pinch hitter for Dugan.

Meanwhile, the text account given in the Detroit Times corroborates that two of the Yankees’ 10 runs were scored when Gehrig was the batter. Here’s what the Detroit Times reporter, Bert Walker, wrote:

First Inning: Combs began by doubling off the scoreboard and after [third baseman] Warner had thrown out Meusel, Ruth’s triple to right counted Combs, and Ruth tallied while [shortstop] Tavener was fumbling Gehrig’s grounder.

Seventh Inning: Ruth drew a pass at the start and Gehrig sent him hoofing it home with a double to left. Lazzeri’s single to left counted Gehrig. Koenig singled to left, putting Lazzeri on third. Holloway then replaced Vangilder and Robertson, batting for Dugan, singled Lazzeri in.

From the above descriptions, it is unmistakable that Gehrig deserves credit for one RBI in his seventh-inning at bat—Gehrig’s double batted in Ruth who was on first base. But what about the first-inning run that Ruth scored—“while Tavener was fumbling Gehrig’s grounder”—does Gehrig deserve credit for batting in Ruth from third base?

A pretty strong case can be made that Gehrig also deserves credit for one RBI in his first-inning at bat. Had Tavener fielded the ball and tossed it to first, Gehrig would have been retired for the second out; and Ruth, who was on third base, would have scored, with Gehrig receiving credit for batting in Ruth. The fact that Tavener fumbled Gehrig’s grounder, which allowed Gehrig to reach first base on the error, had no effect on whether or not Ruth scored—Ruth was going to score on Gehrig’s grounder to short, whether Tavener fielded it cleanly or not. Accordingly, Gehrig should have received credit for batting in Ruth in the first inning—error or no error.

Indeed, the official scoring rules have clearly stated this for decades. Since 1931, the official scoring rules have included this statement: “With less than two out, if an error is made on a play on which a runner from third would ordinarily score, credit the batsman with a Run Batted In.” However, from 1920 through 1930, the only mention the rulebook gives on RBIs is Rule 86, Section 8, which states: “The Summary shall contain: The number of runs batted in by each batsman.” (See Supplemental Material for a complete recap of all RBI-related sections of the official rules.)

That’s it. From 1920 through 1930, the entire scoring rule regarding runs batted in consisted of that nine-word directive. Since no official definition of a run batted in was given, each official scorer was essentially permitted to establish his own definition— which could change.

Whether or not the Elias Sports Bureau decided that Gehrig should receive credit for batting in Ruth in the first inning of the second game of the doubleheader on 26 July 1928, is unknown (by me) at this time. But it certainly could have. I believe giving Gehrig credit for batting in Ruth in that first inning is the statistically-correct action to take. He would have received credit for that RBI according to the rules in effect explicitly since 1931, and perhaps implicitly before that.

In fact, the 1928 official DBD records do credit Gehrig with one RBI in the game on 24 June 1928 for driving in a run even though he was safe at first on a fielding error. With one out in the first inning, with Koenig on third and Ruth on second, Gehrig grounded to Regan, the Boston second baseman, who fielded the ball and made a wild throw to first base; Gehrig was safe at first and both Koenig and Ruth scored, Gehrig getting credit for one RBI (Koenig) while Ruth tallied on the error. Based on that precedent, Gehrig should be credited with two RBIs in the game in question on 26 July 1928. (See the “Discussion” section for an opposing opinion.)

At this point we know specific details for the errors in the official RBI record—and their corrections—for Lou Gehrig in 1926 and 1928—110 RBIs (not 107) and 148 RBIs (not 142), respectively. What about Columbia Lou’s other major league seasons? I again prevailed upon Retrosheet, focusing on the first eight years of his major league career (1923–1930) and used as-yet-unpublished and unproofed Retrosheet PBP accounts for Gehrig’s games for the 1923–1925, 1927, and 1929–1930 seasons.12

Here is a summary of what I found from the games I was able to verify. (Full details and description are given in the Supplemental Material on the SABR website.)

We must emphasize here that every edition of The Elias Book of Baseball Records shows Gehrig as the RBI leader for 1927 with 145 and 1930 with 175; neither figure is correct. 

SUMMARY

From 1923 through 1930, the Iron Horse played in a total of 921 games. Retrosheet presently has complete (although unproofed) PBP information for 734 of those games. From my comprehensive examination of that information, I determined that there are 20 games with RBI errors in the official records. For each of the other 187 games, I examined numerous relevant newspaper accounts and ascertained that there are at least 14 games with RBI errors in the official records. I must emphatically point out that there could be additional RBI errors, as well, because I was unable to obtain complete details for all of the runs scored and batted in for three games:

• 28 Sep 1923
• 04 Jul 1929 (second game)
• 22 May 1930 (second game)

However based on the data we do have (see Supplemental Material for details), it is appropriate to consider the RBIs credited to Gehrig in these three games in the official baseball records as being correct.

Chart 3 provides a complete list of the 34 RBI errors discovered—and corrected—in Lou Gehrig’s official baseball record for the 1923–1930 seasons. Also provided is the relevant information for Gehrig’s teammates who have RBI errors directly connected with Gehrig’s RBI errors.

DISCUSSION

There are three main topics that need discussion:

1. the reliability of the RBI information for Lou Gehrig presented in this article;
2. the consequences of effecting the corrections of the errors in Gehrig’s official RBI record;
3. the implementation of the corrections in Gehrig’s RBI record.

According to my research, as indicated in Chart 3, there are 34 games in which the official baseball records have erroneous RBI statistics for Lou Gehrig during the first eight years of his major league career (1923–1930). For five of these games (Lines 10–14), Pete Palmer, Trent McCotter, and Keith Carlson discovered—and corrected—the RBI errors. I have independently corroborated each of those errors and their corrections.

The supporting documentation for the corrections of each of the 34 RBI errors is provided in the Supplemental Material, which is available on the SABR website. Also provided for each RBI-error game are newspaper box scores RBIs for Gehrig (although it is pointed out that newspaper box scores did not include RBIs before 1928).

RELIABILITY OF THE RBI INFORMATION

The most important aspect regarding the information presented in Chart 3 concerns the reliability of the information. Are the official RBI numbers alleged to be wrong truly wrong and are the “actual” RBI figures claimed to be correct truly correct? Readers are encouraged to examine the supporting documentation and make their own conclusions for each of the 34 RBI errors described here.

In the author’s opinion, 30 of the 34 corrections are “no brainers,” and the supporting documentation is irrefutable in each of the following cases:

15 games where “two-player” mistakes like the Shawkey-Gehrig mix-up shown on Line 1 of Chart 3 were made.

• Two games (20, 29) where the number of RBIs in the official records exceed the number of runs scored.
• Two games (19, 26) in which no Yankees players were credited with any RBIs in the official records—even though the team scored 10 runs and 7 runs, respectively.
• Four games with RBI errors (11–14) resulting from the official scorer mistakenly entering “earned runs scored” instead of “runs batted in” in the RBI column of the official scorer’s report.
• Seven games (3, 4, 8, 25, 28, 31, 33) where the RBI errors—and their corrections—are clear-cut based on the supporting documentation. The RBI errors in these games include mistakes of crediting Gehrig with an RBI for stealing home and not crediting him with an RBI for a bases-loaded walk.

That leaves four games (2, 22, 23, and 30) that perhaps need some discussion beyond the evidence provided by the supporting documentation. I feel that a change is warranted in each of the four:

Line 2. With Combs on third and Meusel on first, with nobody out, as stated in the Boston Globe account, “Gehrig sent an easy bounder to [shortstop] Gross, who threw to [second baseman] Rogell in an attempt to force out Meusel. Rogell dropped the ball, Meusel being safe, while Combs scored.” Even if Rogell had not dropped the ball (and even if Rogell had then relayed to first to also retire Gehrig), Combs would have scored. Combs would have scored, error or no error—the official scorer simply made an error in not crediting Gehrig with an RBI.

Line 22. In the first inning, with Ruth on third, with one out, as stated in the Detroit Times account, “Ruth tallied while [shortstop] Tavener was fumbling Gehrig’s grounder.” Had Tavener not fumbled the ball and thrown to first to retire Gehrig, Ruth would have scored anyway, so Gehrig deserves credit for batting in Ruth. Gehrig also batted in Ruth (from first) in the seventh inning (with nobody out) with a double. For the game, Gehrig deserves credit for two RBIs.

Line 23. With Paschal on third, Combs on second, and Ruth on first, with one out, as stated in the Chicago Tribune account, “Gehrig sent a hot grounder to [first baseman] Clancy. Paschal scored when Bud threw to [shortstop] Cissell to force Ruth. Perhaps Cissell shouldn’t have tried for a double play, but he did. No one was covering first, so Bill got an error, Combs scored and Gehrig went to second.” Paschal scored on the force-out that Gehrig hit into. Had the double play been executed, Paschal’s run would not have counted. But, since the error was made by the shortstop, not the first baseman, the double play was pre-empted. Thus, Gehrig deserves credit for batting in Paschal.

Line 30. With Combs on third and Robertson on first, with nobody out, as stated in the Washington Post account, “Gehrig hit [pitcher] Marberry in his midriff with a smashing drive which the pitcher pluckily fielded. He got Robertson at second, but Combs scored.” This was a simple fielder’s choice force-out—Gehrig deserves credit for batting in Combs.

CONSEQUENCES: SINGLE SEASON

Chart 4 presents the net effect of correcting the RBI errors in Lou Gehrig’s official records for each individual season during the 1923–1930 period.

Except for 1924, errors were discovered in each season during the 1923–1930 period, though those in 1925 resulted in no net change. Six of Gehrig’s season totals are different from what is given in the official baseball records. Of particular note are three seasons where he led the American League: 1927, 1928, and 1930.

CONSEQUENCES: CAREER

Effecting the corrections of the errors in Gehrig’s official RBI record for the 1923–1930 period results in his career-to-date RBI total being 810 RBIs—i.e., four RBIs more than the 806 RBIs calculated from the official baseball records.

Pursuant to the results reported here, we must investigate errors in Gehrig’s official RBI record for the last half of his major league career (1931–1939). I have already commenced the requisite research and initial findings already show at least two games with RBI errors in Gehrig’s official 1938 records. Recently, Trent McCotter presented his research for Gehrig’s 1931 season in which he discovered and corrected five RBI errors.13

IMPLEMENTATION OF GEHRIG’S CORRECTED RBI INFORMATION

The final topic of discussion concerns the implementation of the RBI information for Lou Gehrig described in this article: who is going to use it or where else will it appear?

In addition to the official DBD records compiled by the defunct Howe News Bureau for the American League, there are four fundamental databases of baseball statistics:

1. Pete Palmer’s database
2. the Retrosheet database of game box scores and each player’s daily statistical record
3. the STATS database
4. the Elias Sports Bureau database

Since Palmer’s database of baseball statistics is contractually used by several websites, including Baseball-Reference.com, Retrosheet.org, and SABR.org (including the SABR Encyclopedia), implementation of these corrections would be broad-based and far-reaching.

Likewise, Retrosheet’s database of game box scores and each player’s daily statistical record is a vital database for baseball researchers and the changes would be far-reaching.

Tom Ruane of Retrosheet did review this paper prior to publication and concurred with all but one of the corrections recommended: No. 22, the second game of 26 July 1928. Tom wrote: “I don’t think that there is enough evidence to give Gehrig an RBI on his ground-out in the first inning. It would all depend upon the positioning of the shortstop. If he was playing in and, in the opinion of the official scorer, would have been able to throw Ruth out at home had the ball been handled cleanly, an RBI would not be granted. In general, I think we have to give the benefit of the doubt to the official scorer in this case, despite the fact that it appears he made two other clear mistakes (giving both Combs and Robertson an extra RBI) in the game.”14 It seems likely Retrosheet will adopt the other changes.

When it comes to the STATS database, the authority is Don Zminda, a longtime SABR member and Vice President and Director of Research for STATS. He explained, “We [STATS] have contractual relationships with some major media clients, and those clients expect us to match the official numbers reported by MLB.”15 Those numbers are currently compiled by the Elias Sports Bureau.

Which brings us back to Elias once again. According to longtime SABR member Steve Hirdt, who is executive vice president for the Elias Sports Bureau, “as part of its duties for Major League Baseball, Elias reviews credible evidence potentially involving book-keeping errors that affect statistics from past seasons, and makes judgments regarding whether a change is warranted.”16 With regard to making changes, Hirdt has stated, “We employ a standard of proof that lies somewhere between two of the standards common to judicial matters in this country: that is, somewhere between proof that is ‘clear and convincing’ and proof that is ‘beyond reasonable doubt.'”17

Will Steve Hirdt and Seymour Siwoff deem the evidence presented here for the errors—and corrections— in Gehrig’s official RBI record to be at least “clear and convincing?“ We’ll find out.

What is known for certain is the RBI information presented here for Gehrig’s 1927, 1928, and 1930 seasons is at odds with the most recent editions of The Elias Book of Baseball Records. Thus, it will be particularly interesting to see if Siwoff incorporates any of the corrections/changes claimed here—especially for Gehrig’s AL-leading RBI totals in the 1927, 1928, and 1930 seasons. And what will be shown for Gehrig’s lifetime RBI total—1,990 or 1,991 or 1,995 or 1,996 or …?
Finally, what will Major League Baseball do with/about the RBI information presented in this article? We’ll see; the final draft of this manuscript and Supplemental Material was provided to John Thorn, who was appointed Official Historian of Major League Baseball on 1 March 2011 by Commissioner Bud Selig.

CONCLUDING REMARKS

To wrap up this article I want to emphasize the sub-title—“Striving To Get It Right, Thanks to 40 Years of Research by SABR Members.” We’re not there yet—but we’re getting closer. We’ve discovered—and corrected— 34 errors in the official RBI record of Lou Gehrig for the first half of his career (1923–1930). And we’ve started the research on the remainder of his career (through the 1939 campaign). We plan to complete the requisite research before Opening Day 2012.

ACKNOWLEDGEMENTS

It is with tremendous pleasure that I thank the following people (most of whom are SABR members) for their outstanding assistance in helping me assemble the information presented in this article: Freddy Berowski, Cliff Blau, Steve Boren, Rich Bowering, Keith Carlson, Bill Deane, Steve Elsberry, Mike Lynch, Bob McConnell, Trent McCotter, Andy McCue, Marilyn Miller, Rod Nelson, Bill Nowlin, Pete Palmer, Tom Ruane, Dave Smith, Mitch Soivenski, Gary Stone, Jerry Wachs, Tim Wiles, and Robert Zwissig. I should also like to express my gratitude to Seymour Siwoff for continuing to put out The Elias Book of Baseball Records (especially after The Sporting News ceased publication of its Baseball Record Book). 

HERM KRABBENHOFT, a SABR member since 1981, is a retired research chemist. His baseball research has focused on ultimate grand slam home runs, leadoff batters, triple plays, the uniform numbers of Detroit Tigers, and most recently, consecutive games streaks for scoring runs and batting in runs—which requires having accurate game-by-game runs and RBI statistics—which requires correcting the runs and RBI errors in baseball’s official records.

Notes

1 The 1,990 RBI total does not include the 35 RBIs Gehrig picked up in World Series games or the five RBIs he got in All-Star games.

2 See for example: (a) Herm Krabbenhoft, “The Authorized Correction of Errors in Runs Scored in the Official Records (1945-2008) for Detroit Tigers Players,” The Baseball Research Journal 37 (2008): 115; (b) Herm Krabbenhoft, “The Authorized Correction of Errors in Runs Scored in the Official Records (1920-1944) for Detroit Tigers Players,” The Baseball Research Journal 40 (Spring 2011): 66; (c) Trent McCotter, “Record RBI Streak Discovered: Hall of Famer Tris Speaker Put Together a String of 14 Consecutive Games with an RBI in 1928 That Was Unrecognized for 79 Years,” Baseball Digest (May 2008): 62; (d) Trent McCotter, “Hitting a Home Run and Not Being Credited with an RBI,” newsletters of the SABR Baseball Records Committee (August 2008): 2 and (October 2008): 2; (e) Tom Ruane, “Missing RBI,” SABR_Records@yahoogroups.com (August 22, 2010); (f) Tom Ruane, “More on Missing RBI,” SABR_Records@yahoogroups.com (September 2, 2010); low-RBI games; (g) Herm Krabbenhoft, “The Correction of Errors in RBI in the Official Records (1931– 2011) for Detroit Tigers Players,” manuscript in preparation; (h) Cliff Kachline, “ Hack Wilson’s 191st RBI: A Persistent Itch Finally Scratched,” The Baseball Research Journal (2001):76.

3 Seymour Siwoff, personal communication via telephone conversations, 21 April 2011 and 27 April 2011, and email correspondence, 22 April 2011 and 29 April 2011.

4 Pete Palmer, personal communication via email correspondence, 4 April 2011.

5 The Supplemental Material is available on the SABR website: http://sabr.org/node/3253

6 Trent McCotter, “Tweaking The Babe’s Numbers,” newsletter of the SABR Baseball Records Committee (February 2005): 1.

7 Trent McCotter, personal communication via email correspondence, 25 January 2011 and 24 April 2011; Pete Palmer, personal communication via email correspondence, 24 April 2011.

8 Dave Smith, personal communication via email correspondence, between 1 May 2011 and 31 July 2011.

9 This task was facilitated greatly by the superbly generous cooperation of several SABR members who provided photocopies or scans of the game accounts presented in newspapers to which they had access—Cliff Blau, Steve Boren, Rich Bowering, Keith Carlson, Mike Lynch, Marilyn Miller, and Bill Nowlin.

10 Smith, op. cit.

11 The RBI error for Gehrig in the game on 16 September 1928 was discovered by Trent McCotter. See reference 2(d).

12 Smith, op. cit.

13 Herm Krabbenhoft and Trent McCotter, “Most Runs Batted In: Individual Player …Single Season…American League,” presentation at SABR 41 (July 2011, Long Beach, CA).

14 Tom Ruane, personal communication via email correspondence, 1 September 2011 and 2 September 2011.

15 Don Zminda, personal communication via email correspondence, 28 August 2011.

16 Steve Hirdt, personal communication via email correspondence, 5 July 2011.

17 Steve Hirdt, quoted by R. Chamberlain in “SABR Nine Questions,” The SABR Bulletin, July-August 2006: 6.

The 1934 All-Star Game started ominously for Carl Hubbell. The first two American League batters reached base in the top of the first at New York’s Polo Grounds. The National League’s All-Star infielders gathered in a protective circle around their pitcher. They could offer plenty of advice: Hubbell’s manager with the Giants, Bill Terry, was the first baseman; Cardinals manager Frank Frisch played second; the Pirates’ new manager, Pie Traynor, was at third; catcher Gabby Hartnett was a future manager. Hartnett told Hubbell, “We’ll waste everything except the screwball. Get that over, but keep your fastball and hook inside. We can’t let ’em hit it in the air.”[fn]The Sporting News, 6 July 1944, 15.[/fn]

Babe Ruth was up. Four pitches later the Babe was down, on a called strike three. Lou Gehrig worked the count full, then fanned. As legend has it, Gehrig muttered to the next batter, Jimmie Foxx, “You might as well cut. It won’t get any higher.”[fn]Ibid.[/fn] Foxx cut, and struck out on five pitches to end the inning. After Frankie Frisch’s leadoff homer gave the National League a 1–0 lead, Hubbell faced Al Simmons to start the top of the second. Four pitches, another strikeout. Then Joe Cronin: four pitches, another strikeout. Five Hall of Famers up, five down. The Washington Post’s Shirley Povich wrote, “Here was this frail Hubbell man, his left arm whipping down from his shoulder in a monotonous tempo, reducing the greatest batting array every mustered to a helpless, hapless, hitless horde of sandlotters.”[fn]Washington Post, 11 July 1934, 1. The pitch-by-pitch account is drawn from multiple contemporary sources.[/fn]

Thirty-five years later a sportswriter asked Hubbell whether he thought Nolan Ryan could equal his feat of striking out the AL’s five best hitters in the All-Star Game. “Well, it would be kinda hard to answer that,” the old man drawled, “because Nolan Ryan won’t be pitching against Ruth, Gehrig, Foxx, Simmons, and Cronin.”[fn]Louis Rubin, compiler, The Quotable Baseball Fanatic. Globe Pequot, 2004, 248.[/fn]

KING CARL

Those 23 pitches did not make Hubbell a star. By July 1934 he had already staked his claim as the National League’s best pitcher. Just before he took the mound for the All-Star Game, he was presented with the 1933 National League Most Valuable Player trophy. He had posted a 1.66 ERA, the lowest since 1919, when Ruth started hitting home runs in large numbers. He pitched an 18-inning shutout and a National League record 45 consecutive scoreless innings. With a 23–12 record and 10 shutouts, while allowing fewer than one base runner per inning, he led the Giants to the World Series championship. Sportswriters began calling him “King Carl.”

Starting in 1933, Hubbell rang up five straight 20-victory seasons, including two more ERA titles, two more pennants for the Giants, and another MVP award—this time by unanimous vote in 1936. During those five years Hubbell and Dizzy Dean each won 115 games, but Hubbell’s winning percentage was better thanks to a stronger team. He recorded a lower ERA and WHIP while pitching more shutouts and more innings than Dean.

Diz got more publicity. “It ain’t braggin’ if you can do it,” he boasted. Dean could throw smoke as well as blow it. The quiet, skinny Hubbell was the better pitcher.

THE SCREWBALL

Dean was a screwball. Hubbell threw one. He learned it in his second minor league season, watching an old pitcher named Claude “Lefty” Thomas, who threw a sinker and “made pitching look easy.” Hubbell began working on a sinker: “I found out that the more I turned it over, the more I come up and over [overhand] I could get a much better break on it, you see. Of course, the more spin you get on the ball, the more break, and it slows it up. When I threw the screwball I came right over the top, and I turned my arm clear over and let the ball come out of the back of my hand.”[fn]Carl Hubbell, oral history interview by Walter Langford, 15 June 1982. In the SABR Oral History Committee collection.[/fn]

The screwball is the pitcher’s equalizer against the platoon advantage, because a left-hander’s screwball breaks away from right-handed batters. But Hubbell said, “The real effectiveness of the screwball was not the break at all. It’s the speed of the ball.”[fn]Ibid.[/fn] He perfected the art of changing speeds as he mixed his screwball with a decent fastball and curve, plus excellent control. The late SABR member Fred Stein, who saw Hubbell pitch at the Polo Grounds, likened him to Greg Maddux: “methodical, artful.”[fn]Fred Stein, email, 13 July 2009.[/fn]

Few pitchers throw a screwball for a simple reason: it hurts. It requires rotating the arm inward toward the body, a clockwise motion for a left-hander, the opposite of a curve. One of the rare practitioners, Christy Mathewson, called his version a “fadeaway.” Mathewson said he didn’t throw it often because of the strain.

From the start, managers and coaches warned Hubbell that the pitch would ruin his arm. Ty Cobb, manager of the Tigers, ordered him to stop using it.[fn]Hubbell interview.[/fn] Contrary to legend, Cobb did not let him go; Cobb had already left Detroit when the Tigers released Hubbell to Beaumont of the Texas League, where he pulled the screwball out of the trash. Giants scout Dick Kinsella bought him for John McGraw in 1928.[fn]The Sporting News, 6 July 1944, 15.[/fn]

THE MEAL TICKET

Hubbell set his most enduring record during the Giants’ pennant seasons of 1936 and 1937. He won 24 consecutive decisions—his last 16 in ’36 and his first eight the next spring—still the most by any pitcher.[fn]Hubbell won 24 straight decisions, not 24 straight games. The streak included five relief appearances, two of which resulted in saves instead of victories.[/fn] During the streak, which included two shutouts and 19 complete games, his ERA was 1.78. (A few purists argued that the record was bogus, since he lost one of his two starts in the ’36 World Series.)[fn]See, for example, Daniel M. Daniel, “Carl Hubbell’s Streak No Mark, Scribes Say,” The Sporting News, 20 May 1937, 1. New York writers Daniel, Damon Runyon, and Rud Rennie were among those who believed the streak ended with Hubbell’s Series defeat.[/fn]

During 1937 spring training Hubbell got his enduring nickname. Cincinnati manager Charlie Dressen was quoted as saying, “Hubbell is Terry’s meal ticket, and when Hubbell is through so is Terry.” Dressen protested that he didn’t say it—it was Brooklyn manager Burleigh Grimes.[fn]The Sporting News, 13 May 1937, 8.[/fn] Whoever the author, the name stuck.

By the end of the streak, at the apex of his career, the 34-year-old Hubbell was pitching on borrowed time. He later confessed, “My arm started crooking up in 1934.”[fn]Hubbell interview.[/fn] He had at least three X-rays. Even so, he continued winning more than 20 games each year through 1937.

HUBBELL’S ELBOW

Hubbell started strong in 1938. He pitched a one-hitter in May and won his two-hundredth career game on June 26. On August 13 he posted his thirteenth victory of the season against the Cubs’ Claude Passeau, who wore number 13. But time had run out on his aching elbow. In his next start, the Dodgers knocked him out with a four-run fifth inning and Bill Terry acknowledged that the pitcher was hurting. Hubbell said he could not straighten his arm.[fn]The Sporting News, 25 August 1938, 1.[/fn]

Terry sent Hubbell to Memphis to see Dr. J. Spencer Speed, who had operated on Terry’s and teammate Travis Jackson’s knees. It might seem odd that Terry entrusted his “Meal Ticket” to his hometown Tennessee surgeon rather than the best specialists New York had to offer, but Speed was no country doctor. A graduate of the Johns Hopkins University Medical School, he served as president of the American Orthopedic Association and was co-editor of a standard textbook, Campbell’s Operative Orthopaedics.[fn]Patricia LaPointe McFarland and Mary Ellen Pitts, Memphis Medicine, A History of Science and Service. Memphis Medical Society, 2011; Memphis and Shelby County Medical Society, Memoirs of James Spencer Speed, M.D. 1970.[/fn] The slightly built, 48-year-old Speed cut open Hubbell’s left arm (today the surgery would be arthroscopic) and removed what he described as “a loose body”—commonly called a bone chip. Speed wired Terry, “Everything satisfactory.”[fn]New York Times, 23 August 1938, 22.[/fn]

Hubbell’s arm was in a cast for two weeks. He didn’t throw a baseball all winter. Nobody knew anything about rehabilitation from injury. When he arrived for spring training in 1939, he said he was pain-free and optimistic, but cautioned, “Naturally there isn’t a chance in the world to say how things are likely to go once I start cutting loose with the ball.”[fn]Ibid., 22 February 1939, 22.[/fn] He didn’t make his first spring start for a month, and reported pain in his shoulder afterward. Terry sent him back to Dr. Speed, saying, “Hubbell just hasn’t got it and he simply can’t pitch the way he is.”[fn]Ibid., 1 April 1939, 12.[/fn] Speed diagnosed muscle soreness, “nothing serious,” and prescribed rest.[fn]Ibid., 6 April 1939, 36.[/fn]

Hubbell took another month off before making his first appearance of the season in relief on May 8. Six days later Terry left him on the mound for 10 innings in his first start. In June he went to the bullpen for six weeks. When he returned to the rotation Terry pitched him regularly on three or four days rest. The Giants manager had no clue how to handle a fragile arm; Hubbell started one July game on just two days’ rest—and endured a 13-inning complete game. In only 18 starts, he recorded a 2.75 ERA, second best in the league under 1939 rules, and an 11–9 record. Hubbell hung on for four more seasons, until he was 40, pitching on a reduced workload with inconsistent results and seldom using the screwball. His days as an elite pitcher were over. After the surgery, he said, “I was just half a pitcher.”[fn]Hubbell interview.[/fn]

Hubbell and everybody else blamed his injury on the screwball. Stories abound about his arm, so crooked that his left hand faced outward when he held it at his side, so crooked that he had to have his jackets tailored to make the left sleeve shorter than the right. All because of the screwball. In the spirit of Bill James, I asked, “Is that true?” I consulted two of the leading experts on pitching arms, men who disagree about practically everything.

Glenn S. Fleisig holds a doctorate in biomedical engineering from the University of Alabama at Birmingham, He is co-founder and research director of the American Sports Medicine Institute in Birmingham, which is affiliated with the renowned arm surgeon James Andrews. The institute has mapped the biomechanics of hundreds of pitchers from Little League age to college and professional, using eight cameras hooked to computers to perform a motion analysis of their deliveries and the forces acting on their bodies.

Michael G. Marshall earned his Ph.D. in exercise physiology from Michigan State. Right-hander Mike Marshall threw a screwball for fourteen seasons in the majors (Hubbell said, “Marshall had a pretty good one.”[fn]Ibid.[/fn]) and holds the record with 106 games pitched and 2081?3 innings in relief in a single season. Marshall and Fleisig have aired their differences publicly; they once carried on a flaming long-distance feud via the Internet, arguing their respective theories about how to keep pitching arms healthy. But when it comes to the screwball, they agree.

Although the American Sports Medicine Institute has no data on that rare bird, the screwballer, Fleisig told me, “The screwball is harder [to throw] but I don’t think it’s more stressful.” He acknowledges it may hurt more than throwing a curveball—that’s why most people think it’s more damaging—but pain does not necessarily equal injury.

Marshall says, “Throwing screwballs is safer than throwing pitches that require baseball pitchers to supinate their pitching forearm through release.” Supinating the forearm means turning your left hand counter-clockwise away from your body with the thumb up, the way a left-handed pitcher throws a curve; pronating the forearm is the opposite motion, the screwball delivery. Marshall threw his screwball more than one-third of the time, far more often than Hubbell. He has never had arm surgery.

Both men said pronating the forearm does not increase stress on the elbow, compared with a fastball or curve. In fact, Fleisig’s groundbreaking motion analyses show that a curve produces no more force and torque on elbow and shoulder than a fastball, annihilating another of baseball’s conventional wisdoms. As for bone chips, Fleisig said, “Even a pitcher with good mechanics, if you pitch a lot, you can eventually get bone chips.”[fn]Glenn Fleisig interview, 20 October 2009; Mike Marshall, email exchange with the author, 18 October 2009.[/fn]

A KILLER WORKLOAD

Hubbell pitched a lot. In his first nine full seasons, 1929–1937, he averaged 281 innings per year, more than anyone else during that period, and exceeded 300 for four straight years. He completed almost two-thirds of his starts, including 24 extra-inning games, and started 81 times on only one or two days’ rest. The explanation for Hubbell’s breakdown is clear: His arm wore out.

Still, Hubbell enjoyed an unusually long career for his time; he pitched until he was 40. Among his Giants teammates only knuckleballer Fred Fitzsimmons was effective after age 35. Before World War II extended the careers of some older players, only four starting pitchers in the majors were over 40: Lefty Grove, Charlie Root, and knuckleballers Fitzsimmons and Ted Lyons.

Hubbell continued to believe that the screwball ruined his arm. In three decades as the Giants’ farm director, he worked with hundreds of young pitchers. He never taught the screwball. (His most famous pupil, Juan Marichal, learned the pitch on his own.) Hubbell said he taught it to only one pitcher, his teammate Cliff Melton, and felt responsible when Melton came down with a sore arm.

The prejudice against the pitch remains strong, despite lack of evidence. Besides Juan Marichal and Mike Marshall, the only prominent screwballers since Hubbell were Warren Spahn, who pitched until he was 44, and Fernando Valenzuela, whose career was shortened by a bad shoulder, not an elbow injury. Today pitchers get a similar action—an opposite break from the curveball—with two-seam fastballs and circle change-ups. The screwball’s time has probably passed, though it never really came.

WARREN CORBETT is a contributor to SABR’s Biography Project and the author of “The Wizard of Waxahachie: Paul Richards and the End of Baseball As We Knew It”. He lives in Bethesda, Maryland.

The 1969 Mets’ season ROT is the mean average of the team’s 158 rotation numbers. Until a team’s rotation is established, do not count the first few games of a season: the Mets’ first four games are not counted. The Mets played their games using five different pitching rotation cycles and these span from two-man use to seven-man cycles. The Mets’ quantity of starts in each rotation category distributes as follows across that spectrum: 3, 10, 44, 76, 21, and 4. The season ROT for the 1969 New York Mets is therefore 4.72. Now take a look at the ROT spectrum for the Tigers: 1, 9, 43, 74, 20, and 5. Detroit’s ROT is higher: 4.78. If you think I’m splitting hairs to debunk Mr. Hernandez’ claim, consider this: the numbers are from the 1925 Tigers.

(Click image to enlarge.)

There is a lot more going on in the history and evolution of pitching rotations than generally acknowledged in contemporary accounts. Everything we think we know about days of rest and pitching rotations had already been attempted by major league managers by 1925. I can’t suggest that everything we know about pitching rotations was accepted by major league managers in 1925, but it had been attempted. Nineteen- twenty-five happens to be a watershed year: the five-man cycle, used in 38 percent of all games, finally eclipsed the four-man cycle, used in 34 percent of all games. From that year until Wilbur Wood’s knee injury, the unnoticed sea changes in pitching rotation usage concerned the dangers of the three-man and the impossibility of the six-man. Since 1970 we’ve seen the painful, slow demise of the four-man cycle—even when an off-day provides an extra day of rest. In 2010, 450 games were played with a four-man ROT: the lowest major league total since 1895, so the sea change continues.

If you want to know the origins of the pitching rotation—of the two-man and the three-man and of each successive strategy—you have to root around the first sixty-five years of professional baseball history. Consider the graph of each pitching cycle as a percentage of all games, 1871–1935. (See Figure 1.)

Strikingly delineated are the different pitching rotation eras. Upticks in two-man usage in 1879 and 1880 led to six years of prevalent use. From 1887 to 1897 the three-man won the day, and from 1898 to 1924, a period of twenty-seven years, the four-man won the plurality despite never exceeding forty-six percent of all starts. In 2010, the five-man was used in over 80 percent of all starts for the first time in history. Let’s take a look at each pitching rotation and its own history.

THE ONE-MAN

For any non-SABR members who might be reading this, it should be pointed out that from its origins, baseball was played by nine men on a team. A tenth man could fill in at any of the nine positions—including pitcher—should one of the starters have a hangover, a funeral to attend, or a court date. Yet in this single-pitcher era you’ll be hard pressed to find any sore arms. The general exertion for a pitcher in this era was compatible to a modern, medium paced, softball pitcher. In 1926, Sporting News columnist Jack Sheridan might have been harsh when he called it “quoit pitching—underhand tossing with the straight arm.”

ROT figures stay constant and don’t exceed 1.14 in the National Association years, but days-of-rest changes are dramatic. Days-of-rest has never been satisfyingly calculated because virtually all pitchers, at one time or another, sit out a few starts. Upon their return, player and team days-of-rest are skewed off chart. Buffalo in 1885 has the lowest ROT figures in the majors yet ranks high in days-of-rest because pitcher Pete Conway sat 79 days before his August 10 start.

Conway’s long hiatus had nothing to do with the pitching rotation, but the well-intentioned statistician is pressured to include it. Another 1885 pitcher—The Only Nolan—was rested over 50 days; dozens more were “rested” over 10 days. Brilliant standing work on this subject is by John Rickert, Associate Professor of Mathematics at the Rose-Hulman Institute of Technology. Posted January 2008 on SABR-L, Rickert’s data— calculated by ignoring actual days-of-rest of 11 or more—shows MLB pitchers resting an average of four days per start as early as 1900.

(Click image to enlarge.)

ROT data provides a unique back-door solution to the days of rest problem. A five-man rotation means pitchers sit out four games between each start. If we know the average calendar days necessary to play four games, we have our days-of-rest formula. For teams use: (ROT-1) + (team off days/games) where “games,” as mentioned above, doesn’t include those few that start a season. Since this method does not credit Conway with 79 days-of-rest in his aforementioned start, and since it can be argued that those 79 days must be counted, ROT generated days-of-rest should be referred to as “Manager’s Effective,” or meDOR. Here’s how ROT and meDOR graph for the first 65 years of professional baseball history. (See Figure 2.)

Note how meDOR’s track about 0.89 behind ROT’s after 1905. That’s because 1905 was the first year off-days accounted for nearly 11 percent of the schedule, the figure used today by MLB. As it turns out, National Association days-of-rest was more reasonable than many would think. In 1871 the average team played a schedule of 28 championship games in 147 calendar days. That’s 4.2 off-days for every game and an ROT of 1.03, numbers that produce a first season meDOR of 4.28—higher than any season since. In 1880, for the first time, there were more games played during a season than there were off-days, and this trend, combined with an apparent rise of fast ball pitching in 1877, were likely the driving forces in the origin of pitching rotations. 

THE TWO-MAN
Origin: 1879 Springfield (MA) NA, Robert Ferguson, manager and captain.
First Team: 1881 Chicago NL, Adrian “Cap” Anson, manager and captain.

Cherokee Fisher is spelled as a starter with both Baltimore in 1872 and Hartford in 1874, but these incidents are spotty and probably reflect an overall unreliability on Fisher’s part. In 1875 Robert Ferguson took over the Hartfords and, with Lipman Pike as the captain, sometimes alternated veteran Candy Cummings with rookie Tommy Bond in the second half; Boston’s lead was too big to overcome.

But it was the advent of the International Association in 1877 that brought real change. This league condoned higher-arm fastball pitching and saw a number of teams hand 20 percent of their starts to secondary pitchers. Utica’s 1878 “Joyful Nine” under James Gifford finished in third place by alternating two outfielders: Blondie Purcell and Nick Alcott, for a 1.47 season ROT, the highest to that point by a team that finished over .500. On August 20 of that year, Springfield’s Larry Corcoran walked out of the pitching box in a game against Worcester with a sore elbow. Fred Goldsmith was signed from the London, Ontario “Tecumsehs” to fill the gap. In 1879 the two made Springfield the first team to score a 1.50 ROT. Manager Robert Ferguson became ill in June but his replacement, Lipman Pike, kept the two pitchers alternating.

In the NL, George Wright managed and captained Providence to the 1879 pennant with an interesting 1.33 second-half ROT, with Bobby Mathews being used to give John Ward extra days of rest. In the National Association, Springfield disbanded September 6, ten days after their last regular season game. Corcoran and Goldsmith were signed by the Chicago Nationals within a month. Chicago’s leader—Adrian “Cap” Anson—would win the pennant in 1880 but Chicago’s ROT score was 1.41—marginally better than the Buffalos’ 1.35. More regular Corcoran and Goldsmith alternation in 1881 gave Anson the second of three straight pennants and a never-before-seen ROT of 1.89. Identifying the 1881 Chicagos as the first team to use the pitching rotation is probably not news to SABR membership.

Jim Mutrie assembled the strictest two-man pitching rotations. His pennant winning 1884 AA Metropolitans used it in 87 percent of their games, a mark topped only by his 1886 New York Nationals, who used it 93 percent of the time. This included one stretch where Tim Keefe and Mickey Welch alternated for a record 44 consecutive games.

THE THREE-MAN
Origin: 1881 Mets, Eastern Championship Assn., Jim Mutrie, manager.
First Team: 1887 St Louis AA, Charles Comiskey, manager and captain.

The three-man came into use when season schedules incrementally expanded, putting pressure on two-man staffs. It pops up in 1881 when Jim Mutrie’s semi-independent Metropolitans became the first team to schedule 150 games. Jack Neagle worked as the third pitcher to two mutton-chopped Irishmen, the one-armed Hugh Daily and John Doyle. Major league schedules at that time called for only 84 games.

In 1883 the three-man entered the NL as an option for second division teams with untested pitching prospects, and to this day, out-of-contention clubs will often add a pitcher to their rotations at the end of the season. In 2010, both Baltimore and Toronto used six-man September rotations. As noted by Rickert, six-man use actually spiked in 1998 as more teams were out of contention than in any other season. This had the effect of bringing meDOR for both leagues to over four days of rest per pitcher for the first time in history.

The American Association upped its schedule to 140 games in 1886, removing 17 off-days from each team’s schedule. Comparing 1885 to 1886, Association days-of-rest dropped from 440 to 309 while three-man cycles doubled from 223 to 447. Three AA teams come close to its first committed use this season. St. Louis and Brooklyn almost pull it off, but both use the two-man for extended periods. Harry Stovey’s Athletics have, due to injuries, five pitchers in and out of a staggering amount of three-man cycles: 58 percent. Stovey, however, spent chunks of the season trying to work an archaic 2323 rotation. This meant Al Atkinson worked the “2’s” on one day of rest while two pitchers took turns working the “3’s” with three days of rest.

In 1887, Charles Comiskey’s flag-winning St. Louis Browns, Harry Wright’s contending Phillies, and Horace Phillips’s “Pittsburgs” all played over 50 percent of their schedules with three man rotations, and the contending champion Browns stand out. They had the earliest Spring commitment to the three-man and the necessary depth to make it work. Teams of that era with strong pitchers like Bob Caruthers and Dave Foutz universally forced them into two-man tandems. Bizarre circumstances in mid-August led to three weeks of one- and two-man use: Dave Foutz dislocated his thumb, Nat Hudson’s mother died, and Bob Caruthers got food poisoning. That’s when Silver King won three consecutive starts in three consecutive days, something no pitcher has done since. But the kind of consensus that swept the two-man into acceptance didn’t happen with the three-man: both 1888 World Series participants, the Giants and the Browns, were last in three-man use in their respective leagues.

Frank Selee’s 1890 Boston Nationals were the most committed team to the three-man agenda, a decision no doubt spurred by Boston’s failed use of the one-man rotation in September of 1889. The 1890 Bostons featured John Clarkson, Kid Nichols, and Charlie Getzein, and nearly overtook the Brooklyns with three weeks to play. Brooklyn, led by smart Bill McGunnigle, won the pennant when rainouts helped the team get away with a two-man rotation during the season’s final weeks.

The 1892 Louisvilles’ stretch of 39 consecutive games using a strict three-man seems to be a record that will not be broken. The three man continued to be a tool of the major league manager until 1969. Leo Durocher’s Cubs used it 36 times that year—twice the frequency of the 1964 Phillies—and Chicago came down harder and sooner. Nineteen-sixty-nine didn’t usher in the five-man rotation as many have come to believe. It proved, once and for all, the fallacy of winning with a three-man.

THE FOUR-MAN
Origin: 1885 Detroit Nationals, Bill Watkins, manager.
First Team: 1901 Boston Nationals, Frank Selee, manager.

On June 14, 1885, Bill Watkins, manager of first-place Indianapolis of the Western League, made a deal with last-place Detroit of the National League. Watkins would abandon Indy and the Western League and sign to manage Detroit, bringing along with him a gaggle of Western League players. Among those were left-handed pitchers Dan Casey and Charles “Lady” Baldwin. Watkins combined Casey and Baldwin with Detroit’s existing pitchers, George Wiedman and Charles Getzien, to create the first four-man staff to rotate ten consecutive games. Detroit’s record was 11–34: they went on to win 9 of 10. Alas, Watkins discarded the rotation and used hunches to pick from his pool of pitchers the rest of the season.

On May 17, 1886, Robert Ferguson debuted as manager of the Metropolitan Association team. He turned a two-man staff into a three-man and, in the final games of a September home stand, leaned primarily on a four-man rotation. The team won 7 of 10 forcing Baltimore into the cellar. The following year, in 1887, Ferguson had to open the season without staff lynchpin Jack Lynch. Ferguson used a 2323 in the early going: Al Mays was the strong arm while pikers Ed Cushman and John Shaffer alternated on three games of rest. When Lynch finally joined the team on May 20, Ferguson switched to the four-man, but it was too late. The Metropolitans were in last place and Ferguson was fired ten days later.

(Click image to enlarge.)

The problem with implementing a four- or five-man rotation in the nineteenth century was insufficient pitching depth. Many teams, like Ferguson’s Metropolitans, just presented awful pitchers as the fourth or fifth starters. To put it in a modern context, nineteenth century managers who dabbled with rotation use actually believed Duane Kuiper, on ten days of rest, could beat Steve Carlton on three days of rest. Yet early teams not only discarded pitchers after failure—they discarded the entire rotation idea. A graph of league adjusted ROT and team winning percentage for 1871–1935, the first 1,016 teams of major league history, displays no reward of success for teams who pushed the envelope of longer rotation use. (See Figure 3.)

Captain Ed Swartwood’s 1887 Brooklyn Association team, owner-managed by Charlie Byrne, enacted the first real effort to honor three days of rest for each pitcher. This goal was met in 76 percent of their schedule by tacking on fourth starters for fifty games. Brooklyn became the first team to crack the 3.00 ROT; their 3.26 ROT was actually higher than the 1908 White Sox. A poor season finish led to an apparent abandonment of the policy the following year. If only they had checked their own 1887 game logs: a 1–2 (.333) record in 1 ROT starts; an 8–13 (.381) record in 2 ROT starts; a 22–32 (.407) record in 3 ROT starts; and a 20–23 (.465) record in 4 ROT starts.

Over the next thirteen years, thirteen teams incrementally advanced four-man use (See Figure 4). These trail-blazers included the 1888 Indianapolis Nationals, the 1889 Chicagos, the 1892 Phillies, and the 1896 Baltimore Orioles. The Orioles, managed by Ned Hanlon and captained by Wilbert Robinson, were the first team to surpass a 4.00 ROT score (4.04); the first team to avoid using the same starting pitcher twice in any three-game stretch; and the first pennant winner to use a plurality of four-man starts (42 percent).

(Click image to enlarge.)

Expanded roster sizes during the 1901 AL-NL war provided the impetus managers needed to give the four-man due consideration. That year three teams made serious bids to be first in this category. The Philadelphia Phillies, captained by Ed Delahanty and managed by team director Bill Shettsline, used the four-man 78 percent of the time from April 29 to August 20: Red Donohue, Al Orth, Bill Duggleby, and Doc White. Hugh Jennings took over the captaincy mid-year and used the five-man 29 times late in the season in what appears to be “double-header patching”—a temporary rotation increase by one starter so that days-of-rest for regular rotation members remains constant. The Phillies were a surprise second-place finisher.

The Baltimore Orioles, from June 20 to August 20, ran a four-man 84 percent of the time. Manager John McGraw likely deferred the handling of his pitchers to team captain and friend, Wilbert Robinson. The four- man was adopted in June when ace Joe McGinnity lost a few starts and the team dipped under .500. Jerry Nops joined McGinnity, Harry Howell, and Frank Foreman until injuries and controversy destabilized the team and the rotation in a September free fall.

You’ll notice I referred to McGinnity as the “ace” of Baltimore. ROT calculated for individual pitchers provides a quantitative definition for an often frivolous appellation. Managers bump up aces by a game or push back fifth and sixth starters. You’ll recall some aces pitch the game before the All-Star break, and then the game after—Dan Haren and Justin Verlander did that this year. These managerial moves accumulate over a full season and impact individual pitcher ROTs. McGinnity had the lowest ROT of Baltimore’s 1901 regular pitchers: 2.86. That’s 0.36 below the team ROT and indicative of a pitcher bumped up one game about 20 times. Interestingly, in a four-man rotation, every four times an ace bumps up, he adds a start to his season.

On the Fourth of July, 1901, manager Frank Selee began a strict four-man rotation to help save the Boston Nationals in what would be the last of his 12 seasons at the team’s helm. Boston had used a fuzzy four-man in the early going. From the nation’s birthday on, Selee enacted a 90 percent four-man with Vic Willis, Kid Nichols, Bill Dineen, and Togie Pittinger. Selee patched doubleheaders in June with fifth starter Bob Lawson, but got lucky at the end of the season: off-days helped keep the four-man going through most twin bills. Boston is the first team to use the four-man in 20 consecutive games and at one point used it in 43 of 45 games—a pace unapproached by the 1971 Orioles. Boston finished the season with four-man use in a record 101 games, or 73 percent of their season. Selee deserves a medal for finishing at .500 but he was dispatched out of town without regret.

Brooklyn put together a four-man staff worthy of mention early in 1902, but subsequently the form went into hibernation for 16 years. During this time managers, apparently, used tea leaves and magic dice to pick a pitcher for a day’s game. For long stretches, Rube Marquard couldn’t pitch on Fridays; Doc White couldn’t pitch in Washington, and Herb Pennock couldn’t pitch against Cleveland. Ed Barrow’s Red Sox resurrected the rotation idea in 1918 with a solid four-man that took the pennant. Their Fall Classic opponents, the Cubs, had the best four-man usage in the senior loop, making the 1918 Series the first between the top two four-man rotations from each league. It seems major league managers took notice. Seven teams during 1918 switched to four-man use mid-season. 

THE FIVE-MAN
Origin: 1889 Chicago Nationals, Adrian “Cap” Anson, manager and captain.
First Team: 1928 Chicago Cubs, Joe McCarthy, manager.

On July 9, 1882, at Louisville, the St. Louis Browns presented recently signed Morrie Critchley in the box. It was the first time in big league history that five different starting pitchers were used in five games. But this was no pitching staff strategy; captain Ned Cuthbert’s top pitchers were injured and the team was in a panic.

Similar stories mark the next five scattered incidents of fifth-pitcher use. In 1886, the Cincinnati Red Stockings accounted for eight five-man turns in one season as they grasped for the right three-man staff. These fumbling mid-season periods of pitcher tryouts just don’t qualify as candidate dawns of the five-man rotation.

The first team to stock its bullpen with five quality arms and use them in an effort to contend was Cap Anson’s 1889 Chicago Nationals. Anson’s 1889 team peppered its schedule with 15 sparse occasions of five-man use, 11 percent of their schedule, a record mark nearly doubling Cincy‘s 1886 use. This included four consecutive five-man turns as the team jostled in the early May pennant race pack. The team set a new ROT record with a 3.33 mark, even though Anson gave up five-man cycles after July 20. He adopted three-man with Bill Hutchison making 14 starts as the “2” of the 2323 rotation.

Between 1889 and 1930, 71 teams incrementally advanced five-man use across a swath of rotation sensitive statistics. For a few of these teams, five-man use is noticeably more than incremental. The 1896 Baltimore Orioles, already a noted participant in four-man development, were the first to play over 20 percent of their schedule with five-man cycles. They presented a record setting five-man/three-man ratio of 0.8 (31 to 38). This is a key statistic as noted by Figure 1: most league-wide, post-1905, advances in the five-man are accompanied by a similar retreat of the three-man. Prior to 1896, the best five-man/three-man ratio was 0.2 by the 1895 Louisvilles (13 to 46).

Buck Ewing’s 1898 Cincinnati Reds dropped the three-man and relied on the five-man in August to help them hold first place. But given more regular work, Bill Damman and Bill Hill won only three of 10 games, and Boston, using a four-man, breezed into first place for good. Ewing oversaw the team’s record-setting use of 40 five-man cycles (26 percent), and five-to-three ratio 1.1 (40 to 37). They were the first team to have more five-man cycles than three-man cycles. Then the 1899 Cleveland Spiders and 1900 Brooklyns reached a new mark in consecutive game five-man use: eight. Cleveland strung them up after Labor Day en route to a 20–134 record; Brooklyn did it in July en route to the pennant. Baseball was still years away from the first five-man rotation.

Interestingly, in 1900 Pittsburgh hosted Boston in a key four-game series ending the Fourth of July in which both teams produced five-man rotations. Pittsburgh swept, stayed above .500, and contended to September when they switched again to the five-man. On September 22, Fred Clarke broke this pattern and pitched Jack Chesbro on two days rest against sixth-place St. Louis. Chesbro was hammered and Pittsburgh lost a chance to pull to a half game behind the Dodgers. Six teams, including the minor league AL Clevelands, tried their hand with the five-man in 1900. But it was a fad; a way to get through the short term after which lower rotations would resume. The 1901 Phillies became the first team to use the five-man for 10 consecutive games and Ned Hanlon’s 1902 Brooklyns—the “Superbas”—upped that record by one and disdained the three-man enough to produce a five-to-three ratio of 3.7 (48 to 13).

Patsy Donovan’s 1903 Cardinals were the first team to have the five-man as their most common rotation number, even though they only used it in one-third of their games. Ex-Baltimore Oriole Joe Kelley’s 1904 Reds were the first team with two different five-man streaks at least 10 games long. This appears to be double-header patching. In 1905 the New York Giants opened the season with a five-man. That’s when John McGraw built the big lead which carried them to the pennant. But June 10 he reduced Hooks Wiltse‘s role and finished the year with what you might call a fuzzy four-man.

Fifteen years after the 1889 Chicagos, fuzziness was the word of the day. Strict rotation patterns for any rotation numbers vanish, as mentioned, until 1918. It was the dark ages of pitching. Pattern testing teams for rotation use during these years is like looking at 272 Rorschach tests: The 1911 Pirates… four-man? The 1909 Yankees… positively medieval.

Rotation use in broad spans of seasons can be identified and measured by using Purity Scores. Purity Scores measure the tendency of managers to stick to a rotation over consecutive games. Take a look at the daily ROT numbers for the 1909 Yankees in August: 5645674545456745675654567. Now take a look at the 2010 San Francisco Giants, also in August: 5555555555555555555555555555. Subtract day-to-day changes in ROT, count the resulting zeros, and average the resulting positive numbers. Divide the quantity of zeros by games, and divide again by the average of the positive numbers to get the Purity Score. The season Purity Score for the 1909 Yankees is 13 percent. The season Purity Score for the 2010 San Francisco Giants’ is 94 percent. These two marks are, respectively, the lowest and highest in major league history.

The 1902 to 1917 era has the lowest purity scores in baseball history. Helter-skelter patterns describe each team’s rotation use. Clouds of five-man use increase from 12 percent to nearly 30 percent in this period—all managers gravitating to five-man use as if pushing on a Ouija board. Tendencies for ROT pop up here and there: Connie Mack, with a few notable exceptions, had the league lowest ROT. Fitness-conscious Hugo Bezdek, in a short career, often ranked high. The most interesting use of the five man during this period was by Patsy Donovan’s 1907 Brooklyns and Bill Carrigan’s champion Boston Red Sox of 1915. That year both Boston and Detroit dabbled in late-summer five-man use. Boston made Detroit the second one-hundred game winner not to go to the post-season.

The re-adoption of rotation use, and the abandonment of voodoo pitcher selection in 1918, put the evolution of the five-man rotation back on track. Wilbert Robinson’s 1920 Brooklyns were the first pennant winners to lead in ROT and five-man use. They also set a record with a five-to-three ratio of 7.9 (63 to 8). When the 1922 World Series became the first between two league leaders in five-man use, the stage was set for the development of the five-man pitching rotation.

Between 1923 and 1933, seven teams stand out as possibly having the first five-man rotation. Six of those teams show higher five-man use than the 1969 Mets, the seventh switched to six-man use when bad weather created 31 doubleheaders. (See Table 1.)

From 1922 to 1927, the New York Yankees led the AL in five-man use every year, sometimes by large margins. That’s a testament to the depth that characterizes the Yankee franchise to this day. Their reliance on four-man use and reluctance to patch doubleheaders with sixth pitchers makes them just a little archaic. Their advanced use of ROT over consecutive pennants gives them something in common with the Chicagos of the 1880s, and the Baltimores of the 1890s.

Joe McCarthy’s 1928 Cubs display a staggering 5-to-3 ratio of 38.0, the greatest leap in any five-man measure for any major league team. This shattered the 8.6 ratio of the 1927 Tigers. McCarthy’s seemingly complete rejection of the three-man means his team, easily, is the first whose pitcher use could have been accomplished in our twenty-first century without raising eyebrows—significantly set apart from the Yankee teams of the day. Using ROT values to identify aces, Sheriff Blake leads the 1928 Cubs with a 4.59 ROT over Pat Malone’s 4.80 ROT. But Malone is saddled with 17 relief stints, so many that a calculated meDOR for him is virtually meaningless. This is a common plight for top pitchers in the 1928–1931 period. It is better explained in a history of relief pitcher use, and better explains the high batting averages of the 1930s. These Cubs, by the way, are the first to play a majority of their games with a five-man cycle.

Heinie Wagner’s 1930 Red Sox opened the year with a near perfect five-man “rotation scheme” for the first 20 percent of the season. This countered baseball commissioner Kenesaw Mountain Landis’s decree that starting pitchers not be announced until game time, a measure adopted to thwart gamblers. The Boston Herald didn’t seem to notice the rotation until four weeks into the season. On May 12 they wrote: “The Big Five of the Sox… a pitching oligarchy of power and guile.” The Red Sox dropped the idea June 3 when they pitched Jack Russell on two days of rest. Boston had a 13–29 record and slouched deep in last place. They are the only team in this collection unable to contend.

Gabby Street’s 1931 St. Louis Cardinal world champions had fewer five-man turns than the 1969 Mets, but, as mentioned, had more doubleheaders: 31 to 22. They countered this with massive, record-setting six-man use, a pattern also used by the 1932 Yankees. Eighty percent of the 1931 Cards’ schedule was played with a ROT of 5 or more, a huge increase over the 71 percent attained by McCarthy’s 1929 Cubs, and compatible with an average team from 1997. The 1932 and 1933 Cubs—a tribute to Grimm and Rogers Hornsby—advance the cause of the five-man and are subsequently followed by Joe McCarthy’s Yankees of the 1930s. Unperceived by all, especially the 1969 Cubs, the 1928 Cubs had let the horse out of the barn.

Pitching rotation strategy is still unreported and tides of rotation strategy ebb to this day. The unreported strategy—or lack of strategy—of the new millennium is the mega5. Created by Bobby Cox in 1995, somewhat displayed by Joe Torre’s 1998 Yankees, and attempted by both 2002 World Series participants, Anaheim and San Francisco, the mega5 is a complete elimination of the four-man option for an ace—even after an off-day. With advanced grounds-keeping to minimize doubleheaders, teams can string together upwards of 50 consecutive games with 100 percent pure five-man work. Since then, 28 teams have put together mega5’s including three in the first half of 2011.

On August 9, 2011, Milwaukee’s Chris Narveson cut his pitching thumb with scissors while lacing up his glove, ending the Brewers’ mega5 at 84 games. This left the team 12 shy of the record held by the 1995 Braves. It went unreported.

FRANK VACCARO is a longtime SABR member and Teamsters Local 812 shop steward for Pepsi-Cola (KBI) in Northern Queens, New York. He lives in Long Island City with his wife Maria and their cat Furgood.

Notes

1. All information on game starting pitchers and ROT spreadsheets are sourced from Retrosheet. Accessed 12 August 2011.
2. Mark Liptak, “Interview with Wilbur Wood,” Baseball-Almanac.com, accessed 30 August 2011, www.baseball-almanac.com/players/wilbur_wood_interview.shtml.
3. The Sporting News, 10 June 1926, 4.
4. John Rickert, email to SABR-L Listserv. Archives, post 072962, 12 January 2008.
5. The Sporting Life, 22 October 1884, 3.
6. Springfield (MA) Daily Republican, 21 August 1878.
7. Chicago Inter Ocean, 4 October 1879.
8. Cleveland Plain Dealer, 17 August 1887
9. Dennis Purdy, Kiss ’Em Goodbye, (Bristol, CT: ESPN Publishing, 2010), 117.
10. Jack Kavanagh and Norman Macht, Uncle Robbie, (Cleveland: SABR, 1999) 51.
11. Sporting Life, 17 April 27, 1901, 6.
12. Chicago Tribune, 23 September 1901.
13. Stan McNeal, “Scissors mishap will shelve Brewers’ Chris Narveson,” Sporting News Online, accessed 1 September 2011, http://aol.sportingnews.com/mlb/story/2011-08-10/scissors-mishap-will-shelve-brewers-chris-narveson.

But the position, which has been home to 14 Most Valuable Player awards and over 20 Hall of Famers, witnessed a steep decline in production in 1971—a drought that would last for five seasons—before slowly rebounding, and eventually reaching the height of production posted by the shortstops in the years directly following the end of World War II.[fn]According to the Baseball Hall of Fame website, www.baseballhall.org, there have been 22 players inducted that are recognized as shortstops. Luis Aparicio, Luke Appling, Dave Bancroft, Lou Boudreau, Joe Cronin, George Davis, Travis Jackson, Hughie Jennings, John Henry “Pop” Lloyd, Rabbit Maranville, Pee Wee Reese, Cal Ripken Jr., Phil Rizzuto, Joe Sewell, Ozzie Smith, Joe Tinker, Arky Vaughan, Honus Wagner, Bobby Wallace, John Montgomery Ward, Willie Wells, and Robin Yount. Two notable exclusions are Ernie Banks and George Wright. Banks, who is often associated with the position, played 1,125 games at shortstop, but also 1,259 games at first base, so the Hall of Fame recognizes him as a first baseman. George Wright, who was elected as an executive, played 12 seasons as a shortstop beginning in 1871, and hit .301/.318/.398.[/fn]

POST-WORLD WAR II ERA

In 1946, the first year following the end of World War II, 25 percent—or four of the league’s 16 teams—started a Hall of Fame shortstop; one other team started a former MVP, and two other teams also opened with very good offensive shortstops.[fn]The four future Hall of Fame shortstops are Luke Appling, Lou Boudreau, Pee Wee Reese, and Phil Rizzuto. The former MVP winner, Marty Marion, captured the award two seasons prior, in 1944. And the two other offensively minded shortstops that season were Boston’s Johnny Pesky and St. Louis Browns’ Vern Stephens, who hit .335/.401/.427 and .307/.357/.460 respectively. The Washington Senators also opened the season with Cecil Travis at shortstop, but he was never able to match his pre-war career hitting line of .327/.381/.436; he retired following the 1947 season.[/fn] Clearly, it was a position that was not without its stars, and the overall production proved it.

Using Weighted Runs Created Plus (wRC+), a park and league adjusted statistic scaled so 100 is league average and one point above or below that is one percent better or worse than the league average, Figure 1 illustrates the median value for everyday shortstops.[fn]For further reading on wRC+, please see the glossary section on www.FanGraphs.com.[/fn] For this purpose, an everyday shortstop will be defined as any player who’s played 61.7 percent or more of the season’s games at the position (i.e. 100 games in the current 162 game schedule, 95 games in a 154 game season, etc…)[fn]The median was used in this study, as well as most of the following analysis, in hopes of eliminating any outliers, both positive and negative, that could heavily and unjustly influence the data set, and in hopes of better capturing how the overall position performed as a group. This explanation is probably served best in the following example: 21 players appeared in 100 or more games at shortstop in 1969. The home run totals for those 21 players follow: 40, 14, 12, 10, 10, 7, 6, 5, 5, 4, 4, 4, 4, 3, 2, 2, 2, 2, 2, 0, 0. Combined, those players hit 138 home runs; an average of 6.57 home runs per player and a median of 4. But there was an extreme outlier: Rico Petrocelli. Of the 138 home runs slugged, he was responsible for 40 of them, or approximately 29 percent. Clearly, his output heavily influenced the average, which, in the bigger picture, is a bad representation of the group’s production; especially considering that more than 70% of the group falls below the average.[/fn], [fn]The 61.7 percent cutoff also creates a sufficient sample size. Of the 65 years studied throughout the duration of this paper, there were only two seasons in which at least 70 percent of the teams’ shortstops did not qualify. For example, in 1992, only 18 shortstops matched or exceeded the 61.7 percent cutoff; there were 26 teams in the league. So, 69.23% of the teams were represented that year.[/fn]

Keeping in mind that wRC+ is a park and league adjusted statistic and 100 is league average production; Figure 1 illustrates that six of the first eight seasons following World War II—1946 to 1953—half of the everyday shortstops performed offensively at 90 percent of the league average or better. To put it in a better perspective, this feat would only occur three additional times—in 1962, 1964, and 1966—before the establishment of SABR in 1971.

Also, despite the relatively high marks between 1961 and 1966, the 1961 and 1962 expansion clubs, for the most part, were hindered by the lack of production at shortstop, and that, ultimately, lowered the overall median production. The Washington Senators and Los Angeles Angels, who joined the American League in 1961, were on opposite sides of the spectrum in terms of shortstop production. Neither team had a shortstop start 100 or more games in 1961, but the Angels got decent production from Joe Koppe the following season—wRC+ of 89—and superb years from Jim Fregosi, who never posted a wRC+ less than 100 from 1963 to 1970. On the other hand, the Senators started Ed Brinkman, who was skilled with the leather, for the majority of the 1960s; he had two seasons in which he posted a total of 80 or better. The Colt .45’s and Mets, the National League expansion teams in 1962, both suffered below-average production from the position in the first five years following their inception, too; though Sonny Jackson did have his best year in 1966 with Houston when he posted a 101 wRC+. So, one team out of the four had a consistently above-average offensive shortstop. [fn]Ed Brinkman, whose top two offensive seasons were in 1969 (87 wRC+) and 1970 (80 wRC+), had three seasons—1967, 1969, and 1970—in which he saved 20 or more runs above the league average, according to Total Zone.[/fn], [fn]The 1967 season appears to be an anomaly in terms of shortstop production for the following reasons:
a. Several shortstops suffered a noticeable decline in production in 1967 and either rebounded the following season or didn’t qualify at all. Those players, along with their wRC+ totals for all three seasons beginning with 1966, include: Luis Aparicio (97, 77, 90), Ed Brinkman (64, 47, N/A), Dal Maxvill (71, 64, 89), Sonny Jackson (101, 71, N/A), Don Kessinger (64, 50, 64), and Denis Menke (109, 91, N/A). Zoilo Versalles also suffered a severe drop off—82 wRC+ to a 54 wRC+—but he held steady the following year, in 1968.
b. The 1966 season featured three players—Dick Groat (78 wRC+), Dick McAuliffe (149 wRC+), and Maury Wills (84 wRC+)—who did not qualify in 1967, and two of which (McAuliffe and Wills) performed well.
c. The 1968 season also featured several well-below-average performers who did not qualify the previous season. For this purpose, we’ll define “well-below-average” as 21 percent or greater below average. Those players, along with their wRC+ total for the 1967 season, are as follows: Ruben Amaro (70), Hal Lanier (41), Ray Oyler (60), and Bobby Wine (45).
d. While the league ERA dropped from 3.52 to 3.30 to 2.98, the position’s median value rebounded during the third year, the “Year of the Pitcher.”
e. Finally, after the initial decline in 1967, the position’s median value rebounded for two straight seasons before another slight decline—one which was less drastic than the decline in 1967.[/fn]

The 1967 season is another point worth noting. It was the lowest total in 21 years and would become the third lowest total in the post-WWII era. A strong case can be made for it being an anomaly.[fn]Based on team averages, runs per game have increased in all but one (1962) of the expansion seasons beginning in 1961. However, expansion teams, especially those prior to 1993 (both the Marlins in 1993 and Diamondbacks in 1998 had good offensive shortstops), have been hindered by poor offensive shortstops, which, in turn lowers the position’s median production. For example, including the qualifying expansion shortstops for 1969, the median wRC+ is 87, and without the expansion shortstops it is 88.[/fn]

THE FIRST HALF OF THE SABR ERA

The 1971 season, the first year of SABR, was the beginning of a lull in shortstop production, coincidentally. Whereas the median wRC+ total for everyday shortstops sunk below 80 two times from 1946 to 1970, the total remained below that point eight of the first 11 seasons during SABR’s existence and 11 of the first 20.

More than half of SABR’s first 20 seasons in existence were marked by well-below-average offensive production, which would be acceptable if, and only if, there was a positive shift in defensive play. There wasn’t.

From 1966 to 1970, the 10 teams—excluding the 1969 expansion clubs—to receive the least amount of production from both their starting and backup shortstops were (starting with the worst) the Giants, Cardinals, Cubs, Phillies, Mets, Reds, Indians, Senators, Twins, and Tigers. Using Total Zone (TZ), an advanced defensive metric based on play-by-play data, those clubs, which posted a wRC+ of 74 or fewer, combined for a +93 TZ, or 93 runs saved.[fn]While there are more advanced metrics, Total Zone—developed by Sean Smith—is considered the most reliable advanced metric for seasons predating 2002. In an effort to maintain consistency and accuracy, it was used for all seasons in this paper. For further reading on Total Zone, please refer to www.Baseball-Reference.com or www.FanGraphs.com.[/fn], [fn]During the 1966 to 1970 study, the 1969 expansion teams— Royals, Pilots, Expos, and Padres—were ignored because they only played two of the five seasons.[/fn]

In contrast, the 10 teams to receive the worst shortstop production between 1971 and 1975 (including the 1969 expansion teams)—Braves, Expos, Pirates, Indians, Tigers, Brewers, Cardinals, Padres, White Sox, and Yankees—posted a -174 TZ, or cost their teams 174 runs because of sub-par defensive play. And the difference may seem significant, but it isn’t. Remember, this is over a 10-year span for 20 teams. The first time period saved, on average, 1.86 runs per year. The second period, 1971–75, lost 3.48 runs per year.

So, despite the decline in offensive production from the late ’60s to the early ’70s, there appears to have been no shift towards increased defensive play.

But delving deeper and looking at the qualifying everyday shortstops who were well-below-average offensive contributors during the same two time periods, the differences remain insignificant as well, 61 and 53 runs saved (or an average of 1.5 and 1.2 runs saved per player).[fn]Similar to Note 7c, “well-below-average” is defined as 21 percent or greater below the league average (i.e. a wRC+ of 79 or less).[/fn], [fn]A change in philosophy would require managers relying more on defense, not offense. Even though TZ is scaled to the league average, the defensive skill for poor-hitting shortstops would have to increase relative to the league average to prove a change in philosophy; it didn’t. In fact, it remained in close proximity in both time frames (1966–70 and 1971–75). So, even if defense at shortstop improved league-wide, the worst offensive shortstops showed no drastic improvement relative to the league. Thus, managers/teams did not appear to willingly sacrifice offense for defense.[/fn]

Similar to the previous expansion years, the 1969 expansion season in which four teams—Kansas City Royals, Seattle Pilots, San Diego Padres, and Montreal Expos—were added seemed to have a negative effect too. During those first five seasons, only Freddie Patek’s 1971 was above-average. The same can be said about the 1977 expansion teams. Both the Blue Jays and Mariners suffered during their first five seasons, except this time it was Craig Reynolds’s 1978 season that hovered near the league average.

One other season of note is the 1981 strike-shortened season, which will be discussed later.

THE SECOND HALF OF THE SABR ERA

After witnessing a decline and minor rebound during the first half of the SABR era, the second half blossomed, beginning in 1991 and lasting for the next 19 seasons.

Unlike the 1971 to 1990 period, the median production topped 90 or better eight times, and four of the last five seasons. And there was only one instance, 1994, when the median production dipped below 80 percent of the league average, similar to strike-shortened 1981.[fn]The 1995 season, while technically a strike-shortened season, lost only 17 games, unlike the 1981 and 1994 seasons in which 51 and 45 games were lost, a significant decrease in sample size. Also, the cutoff point, 61.7 percent, resulted in 90 games for everyday shortstops in 1995, and only 69 in 1981 and 72 in 1994, also significant differences. Because the 1995 season played 90% of a 162-game regular season, it was treated as one. [/fn]

While both strike-shortened years played fewer than 120 games, it is much more difficult to come to a concrete conclusion on why significant downturns were experienced in both seasons.

Both years were within five seasons of an expansion—1977 saw two teams added to the AL and two teams were added to the NL in 1993. But, again, both strike seasons followed and were followed by significantly better seasons. So, it’s not likely that the expansions were a main cause, though they may have contributed somewhat.

Both years did have several players post good numbers the year before and the year after, but witnessed a sharp decline, which helped explain the 1967 season as an anomaly. So, this, too, could have been a contributing factor, at least to some degree. But perhaps the biggest influence was sample size. In both cases, qualifying shortstops only needed to play 69 and 72 games, respectively; which is roughly 30 games fewer than the surrounding years and a significant difference. Overall, it’s difficult to know definitively if that’s the main cause for the sudden declines, though the author believes it is the main influence.

Finally, within the last decade or so, the median shortstop production rebounded quite nicely. In fact, it’s on par with the level of production that the first post-war years witnessed.

The author would be remiss not to mention the use of steroids during the 1990s and 2000s. Did they affect the game? Yes, undoubtedly. But the feeling of this author is that the “Steroid Era” will inevitably be considered just another era in baseball history, much like the way the Deadball Era is viewed today. In both cases the statistics were altered by something; in the case of the Deadball Era, the ball and the overall strategy of the game, and in the case of the Steroid Era, performance enhancement. Basically, it’s difficult to separate the non-steroid users and the steroid users, which is unfortunate, but it is a fact. Knowing so, it’s best to consider the game’s stats during this recent era as a whole, accepting them at face value and for what they are.

CONCLUSION

Figure 4 illustrates the median wRC+ for all qualifying shortstops from 1946 to 2010, and it clearly shows that the years closely following the end of World War II are followed by a downturn, which eventually leads to a lull beginning in 1971. This was followed by a slowly upward trend in production which ultimately peaks during the last five or so seasons.

While the majority of the median production on both ends of the graph resides above the 85 mark, the individual performances parallel each other as well. From 1946 to 1951, there were 28 instances that a player hit league average or better, or approximately 5.6 players per year. Since there were 16 teams in the league during that period, it averages out that 35 percent of the teams during that stretch benefited from league average production or better. And applying the same logic to the 44 times a shortstop topped 100 or more from 2006-10; 29.3% of teams within the last five years received league average or better offense from shortstops. Similar numbers.

So, relative to league average production, are shortstops better hitters today? Yes, generally speaking, they are better hitters than they were at the time SABR was established; they are better hitters than they were 30 years ago, and 20 years ago, too. But they are producing relative to the league near the same level that shortstops did during the first 15 or so years following World War II. And how many people would have suspected that? Those who tout the “common wisdom” about shortstops of generations past being defensive specialists who were not great hitters are suffering from either selective memory, a lack of knowledge,
or both.

JOSEPH WERNER seemed to always have a baseball in his left hand while growing up in suburban Cleveland. He played the game throughout his childhood, and continues to do so. He embraces every facet of America’s pastime. He has worked with Baseball Info Solutions as a video scout and analyst. Werner, an Ohio State grad with a degree in electrical engineering, is now co-chair for the Minor Leagues Committee of SABR. For further baseball analysis, visit his website: www.ReleasePoints.com. To contact Werner, email him at JosephMWerner@yahoo.com, and Werner welcomes feedback and suggestions.

The Bucs came close to doing so a few years earlier, as Willie Stargell recalls. “In 1967, in Philadelphia, Harry Walker [then manager], started eight of us, but the pitcher, Denny Ribant, was white.”1

Stargell was a Pittsburgh mainstay in 1971, going into his tenth season in the big leagues, in the lineup with Roberto Clemente, Manny Sanguillen, and many other players who had been Pirates for some years. Al Oliver felt it was not manager Danny Murtaugh’s intention to make history that day. It’s doubtful Murtaugh even was aware of how unusual the team on the field was. His goal was to put the best players available on the field.2

Oliver “really wasn’t surprised, because if you look at the Pirates back then, we started five [black players] anyway. So what you’re basically talking about is adding four more. Our whole outfield was black and then of course you had Dave Cash in the infield and Jackie Hernandez at shortstop.”3

In fact, Al Oliver didn’t notice the uniqueness of the configuration until the third or fourth inning. “I mentioned to someone. I said, ‘You know what? We have all brothers out here.’ To me it wasn’t a big thing because that’s the way the Pittsburgh Pirates organization was. Even when I was coming through the minor leagues I would go to spring training and I would see black and Latin players. And the Pirates were known for that way before the other teams were and so I wasn’t really surprised with the lineup that Murtaugh set up.”4

Oliver does mention one other important element: “But the key to the whole lineup was that Woody Fryman was pitching against us that night with Philadelphia and as a result of that they would platoon. Gene Clines would play and [usually] Bob Robertson would play first base. But that night Bob didn’t play first base; I played first base.”5 Oliver was told by Dave Cash that “Murtaugh was kind of disappointed with Bob for some reason, I don’t know what the exact reason was, but he was disappointed in Bob, so he played me at first base.”6

The lineup on the clubhouse wall was:

• Stennett, 2B
• Clines, CF
• Clemente, RF
• Stargell, LF
• Sanguillen, C
• Cash, 3B
• Oliver, 1B
• Hernandez, SS
• Ellis, P

“Oh, yeah, we were aware,” Steve Blass recalls. “The thing that I remember about it, when he was interviewed Murtaugh said, ‘I put the best athletes out there. The best nine I put out tonight happen to be black. No big deal. Next question.'”7

In 1971 the Pirates represented baseball’s most heavily integrated team, with black and Latin players accounting for nearly 50 percent of the team’s roster. Richie Hebner, who was injured, says the pre-game reaction to the lineup from the players was an example of the friendly racial humor that was prevalent with the Pirates. “The humor was doled out purely for fun, and not intended to be taken personally.”8

When starting pitcher Dock Ellis saw the lineup he said, “Aw, man, what’s going on here? Who made up this lineup? Murtaugh didn’t know what he had done. He really didn’t. Somebody was probably hurt and needed a day off, and that’s what he came up with. And it didn’t even dawn on him. I would bet my life he didn’t know what he had done. Somebody probably had to tell him after the game.”9

An air of speculation surrounds the issue of why the event wasn’t recognized more prominently at the time. Some local sports writers felt perhaps race was an issue to the extent that fans weren’t coming to the games because of the racial and ethnic makeup of the team. As Phil Musick, a Pittsburgh sportswriter, wrote, “Baseball at that time, in my opinion, had a whole lot of racial division and I think it went around inside baseball and angered some people … There was also some hostility in the city. Pittsburgh is a conservative city and there were a lot of snide remarks made privately. I’m sure there wasn’t a major reaction in the media other than to observe that it had taken place and it was a first.”10

Another writer, Bill Nunn, had a blunter viewpoint. “It’s always been a problem of management. How many blacks will the fans take?”11

“I went down to Joe Brown’s office not long after that game, and he had a stack of mail, and he said, ‘You can take out any letter you want and it will be negative.’ Some of the most derogatory letters, and he had stacks of them, were from white fans.”12

General Manager Joe L. Brown, though, was proud of the roster he built. “[I] was always proud of the fact that we never paid any attention to color in our organization. I don’t think any club in the history of baseball had as many blacks on their roster at one time like we did at one time, and consistently over a period of years. One time 13 out of 25 on the list were black. That had never happened before, and to my knowledge has never happened since.”13

According to Richie Hebner, “The ’71 Pirates were a very harmonious team. Friendships and social gatherings often crossed racial lines.” The tensions in the city that might have kept white fans away from the ballpark were not reflected in the clubhouse. “We had a loose group all laughing and hollering and teasing each other. I thought it was a great reaction.”14

Oliver felt there was a good mix. “It was well-blended. We had no problems at all. None at all. It was a great team to play on. When we got black players, and Latin players, and white players you think over the course of 162 games that something would go wrong. You know, someone might say something to someone. But it never happened on the ’71 team. We had it together. Good people. The best way to describe our team was characters with character. We had our personalities, but we got along well.”15

Dave Cash agreed. “I heard a lot of remarks for the late Danny Murtaugh. I remember him saying he didn’t realize who was out there, he just wanted to put the best team on the field and with the Pirate family it didn’t matter what color you were. We were about winning. That was the most important thing. In 1970 when we got in the playoffs and lost, we tasted that defeat and didn’t want it to happen in the next year. So in ’71, we took care of business!”16

They did indeed “take care of business” by defeating the heavily favored Orioles in seven games in the 1971 World Series.

GEORGE SKORNICKEL has been a die-hard Pirates fan for over 50 years. A retired teacher, he has written “Beat ’em Bucs: the 1960 Pittsburgh Pirates”, and he and his daughter collect 1960 memorabilia. A member of the Forbes Field Chapter of SABR, he lives with his wife, Kathy, and his labs, Bailey and Maz.

Notes

Forty years ago, Forman’s column could not have been written. At the time, the analytic techniques did not exist, the data were inaccessible, and the computing power unavailable.

In 1971, the year the Society for American Baseball Research was founded, the analysis of baseball statistics was still in its infancy, and computers were in the hands of few. Computers were expensive to buy, difficult to use, and most were used primarily as scientific instruments and business machines dedicated to single tasks. In that year, Intel introduced the 4004 microprocessor and went into volume production with the first viable semiconductor memory chip, the 1103, and IBM developed the 8-inch floppy drive. All these were precursors to the personal computers that would appear during the next decade. In 1971, the Internet did not exist; its predecessor, ARPANET, linked 23 computers together at universities, private research centers, and government agencies.

In 1971, those interested in doing statistical analysis to study the game of baseball already existed, but since SABR had just been created, the term sabermetrician hadn’t yet been coined. They often worked where computers were available as part of their work or research. Pete Palmer accessed an IBM mainframe in the early ’70s and kept his database on punch cards. In even older times, researchers like George Lindsey and Earnshaw Cook used “tabulators”—fancy adding machines to perform the necessary calculations to draw conclusions—and kept large paper-based accounts of baseball games to help them use these machines to do their work.[fn]According to Alan Schwarz, Lindsey spent hundreds (it could have been thousands) of hours to answer simple questions about baseball strategies. As Schwarz says, “…obviously, he had no computer to handle the computation.” Schwarz also reported that Cook used a sliderule and a Frieden calculator. The Numbers Game, 72 and 77.[/fn]

Since 1971, progress in both fields has accelerated greatly, with information technology making possible analysis that wasn’t feasible 40 years ago. Sabermetrics developed alongside the information age, with personal computers enabling those who did not work where computers were easily available to develop their algorithms and analyze data at home. The Internet spawned websites and blogs, connecting people and enabling them to gather and store vast amounts of data. Sabermetrics has gone from something a passionate few studied to something millions access and understand. These days articles in Sports Illustrated talk about zone ratings; Murray Chass and Alan Schwarz have sliced and diced data for columns in the New York Times; Rob Neyer uses statistics and stats jargon extensively in his work for SB Nation. Millions of readers were able to enjoy and understand Sean Forman’s piece on Ryan Howard in the Times.

INCEPTION: THE 1970S

Dick Cramer, an early pioneer in baseball statistical analysis, was fortunate to have access to an IBM computer as a graduate student. He got hooked on programming when he worked with a baseball simulation program. He later wrote his own simulation program for the PDP-1 at the Harvard Research Center. It was during this time that he developed the concept of base-run average, which he later published with Pete Palmer. By 1971, Cramer had left Harvard and no longer had access to a “real computer,” so he used the HP-67 desktop calculator to continue his work. At roughly the same time, Pete Palmer was using an IBM 1800 available to him at his workplace. He amassed a large database, stored on punchcards which took up thirty drawers in a card file cabinet. He later moved to a CDC 6600 and moved his database to tape storage. It was on these machines that he began to develop his models for run scoring. Palmer met Cramer through SABR, and throughout the early seventies, they collaborated on seminal articles in the Baseball Research Journal, starting in 1974.[fn]Palmer introduced such fundamental concepts as On-base average (1973), home park advantage (1978), and with Cramer base-run percentage (1974).[/fn] According to Cramer, they exchanged typewritten letters through the postal service in the process of developing their manuscripts. Later, their collaboration would lead to Cramer’s involvement with STATS, Inc. Cramer and Palmer were not alone in their work for long. Bill James published an article in the 1977 Baseball Research Journal, the same year he published his first Abstract.[fn]The Bill James Baseball Abstract, 1977.[/fn]

Table 1. Timeline of SABR/Baseball and IT Events

(Click image to enlarge.)

Cramer and Palmer relied on access to workplace computers, as would many of their successors. In 1971, the idea of a “personal” computer was too far-fetched, except for a few computer researchers, to even consider. Two months after SABR was founded, one of the signal events in the history of computing occurred, the introduction of the Intel 4004. The 4004 was the first single-chip CPU available as a commercial product.[fn]The 4004 was developed by a team of Intel engineers. Ted Hoff, the first Intel fellow, was responsible for the concept, Stan Mazor for designing the instruction set. Federico Faggin did the logic and circuit design for the chip, with the assistance of Masahito Shima, who was on loan from the first customer for the chip, Busicom.[/fn] The 4004 proved the feasibility of developing a small computer system based entirely on a technology that could be mass-produced.[fn]The MOSFET (metal-oxide semiconductor/field-effect transistor) was the technology Intel was created to commercialize. It would take over the market from other semiconductor companies like Fairchild and Texas Instruments. The 4004 as well as the surrounding memory chips were MOS integrated circuits.[/fn] Intel followed the 4004 quickly with the 8008, an 8-bit microprocessor that enabled a variety of new applications. These processors led to the first commercially available[fn]Xerox is often attributed with the creation of the first true personal computer, the Alto, at their Palo Alto Research Center (PARC). These were not the first desktop computers; HP and Wang had both developed programmable desktop calculators that executed stored programs. Dick Cramer, as well as the author, used such machines.[/fn] personal computers, primitive as they were. One of these was a kit, the MITS Altair 8800,[fn]The Altair was the inspiration for Bill Gates and Paul Allen to form Microsoft, to develop software for the new generation of computer.[/fn] first available in 1975, and used principally by hobbyists, the same sort of people who in an earlier era built Heathkit stereos. The Altair is long gone, but it led to the creation of Microsoft; founders Bill Gates and Paul Allen built the first BASIC[fn]BASIC was widely used as a programming language for various forms of personal computer.[/fn] interpreter for the machine.

Computers were not essential for conducting all research. For simple analysis, a hand-calculator and statistics available from Major League Baseball were sufficient. As Alan Schwarz details in his history of baseball statistics, The Numbers Game, people were counting up stats and working with them on paper from the game’s earliest days. However, for some deeper analysis the ability to process large amounts of data quickly was needed. Statistical analysis such as linear regression required significant computing power (at least in the day). The 1976 Baseball Research Journal published an article by George T. Wiley[fn]Wiley was an outstanding undergraduate athlete at Oberlin College. He went on to get a PhD and taught history at the University of Pennsylvania, where he was when he wrote his paper for The Baseball Research Journal. He was inducted into the Oberlin Athletics Hall of Fame in 1998.[/fn] entitled “Computers in Baseball Analysis.” Wiley starts his article with this emphatic statement:

The most significant development in the use of statistics over the past 25 years has been with computers. Mathematical computations that formerly took hours to do by hand are completed by the computer in seconds. Masses of statistical information are now being analyzed in ways never before thought possible

Wiley’s research concerned a linear regression model in multiple dimensions that related numerous baseball statistics to a team’s ability to win. He analyzed 640 different teams, and correlated 17 different pieces of information to the final standings of those 640 teams. Although today that analysis would be a relatively trivial undertaking, the limits of computing resources forced Wiley to limit his study to the years 1920 through 1959, less than half the games played at the time the article was written.

When Wiley wrote his article, even the machines in use at research institutions and corporations were relatively limited by today’s standards. For example, the IBM 370 model 158 computed at 1 MIP, had 87 MB of disk capacity—or about 2 percent of a cell-phone’s computing power and storage. When it was introduced in 1971, its primary data input was through punch-cards—the IBM 3270 data entry terminal (the most popular data entry terminal in its day) was still one year away.[fn]The IBM 2260 terminal, which was a punch-card-emulation system, was introduced in 1964. It had the ability to transmit punch-card images electronically without producing the actual cards. In some respects, the 3270 was a major leap forward.[/fn], [fn]Mark Andreesen joked in the early days of the Internet browser that the earliest browsers were a return to the 3270. Only with the introduction of multimedia capabilities to the browser was this analogy broken, something that Tim Berners-Lee did not anticipate when he invented the World Wide Web. (Private conversation.)[/fn]

Another event in 1976 would help change the state of computing: Intel started development of the 8086 16-bit microprocessor,[fn]For an early history of the Intel microprocessors, see Intel Microprocessors: 8008 to 8086 by Stephen P. Morse/Bruce W. Ravenel/Stanley Mazor/William B. Pohlman. Dr. Morse was the architect of the 8086 and is still active on the Internet. See stevemore.org for an online copy of this article.[/fn] whose “little brother,” the 8088[fn]The 8088 used the same instruction set, but was able to directly interface to 8-bit peripherals widely available in the market. This fact made the IBM PC cheaper to build and hence more economically viable than a fully 16-bit computer.[/fn] was the heart of the first IBM PC. It would take another five years before the IBM PC was introduced, but during this time, other widely usable personal computers entered the market and were in use by many of the early sabermetricians. Other significant developments occurred during the 1970s that would benefit analysts in later years. The SQL database access method (and database language) was invented at IBM in 1974. It would be the progenitor for practical databases developed in the 1980s and 1990s.

THE PC ERA: THE 1980S

By 1980, Dick Cramer had switched to using an Apple II, an early personal computer useful in schools and homes. In 1981, Cramer helped form STATS, Inc., and did the bulk of the programming for that venture on his Apple II in UCSD Pascal. By that time, Bill James had published numerous versions of his Baseball Abstract, and Pete Palmer had met John Thorn, with whom he would develop Total Baseball using the database that he had created on cards years earlier.

The personal computer had already become a useful tool by 1980, but the introduction of the IBM PC in 1981 changed the nature of the computer business. It also changed sabermetrics. As Gary Gillette explained to the author,[fn]Private conversation, May 5, 2011.[/fn] “What the PC allowed was a democratization of baseball statistical analysis.” This didn’t happen immediately, as most uses of the IBM PC were in business; home users still used Apple IIs and Commodore 64s. However, as PC clones appeared, prices came down.

The PC enabled non-computer professionals to develop programs using programming languages like BASIC (which Pete Palmer used to develop BAC-Ball for the Philadelphia Phillies and Atlanta Braves), and desktop database tools such as dBASE III and Paradox. Gillette used both when he was running Project Scoresheet. Tom Tippett, David Nichols, and David Smith wrote the input and analysis code for the spreadsheet using Borland’s Turbo C.[fn]InfoWorld, April 2, 1990, 43.[/fn] They later wrote code for Baseball Advanced Media using Turbo C and C++.[fn]Turbo C, Turbo C++, and Paradox were all products developed by Borland International. The author led the development team that produced several versions of Turbo C++.[/fn] (Tippett went on to become director of baseball information services for the Boston Red Sox.) According to Gillette, Bill James, who started Project Scoresheet, opposed computerizing the data, perhaps for reasons similar to those that motivated his opposition to using PC’s in the dugout.

Computing power during the PC Era progressed from tens of thousands of instructions per second to many millions. As a point of reference, in the Infoworld article,[fn]Infoworld, April 2, 1990[/fn] Pete Palmer provided an example of searching 10,000 data in 15 seconds on a PC/AT. The AT processor was capable of 0.9 MIPS in the then current release. By the end of the ’90s, Intel Pentium III-based processors were 2,000 times as fast.[fn]Today’s extreme Intel microprocessors are theoretically capable of attaining nearly 150,000 MIPS, for applications tuned to their particular architectures.[/fn] (The author’s laptop computer uses an Intel microprocessor that exceeds 2,000 MIPS.) That kind of processing power enables sophisticated modeling, such as the Markov models that Mark Pankin, Tom Tango, and others use extensively in their analyses.[fn]Markov models rely on the manipulation of matrices, special kinds of tables of data. Matrix multiplication, inversion, and other operations are facilitated by high-speed computers.[/fn], [fn]Mark Pankin says that prior to 2000, there were limitations to what he could easily do on a PC, but that since then, he has been able to create his models with relative ease, usually only needing Microsoft Excel to do his work. The author has had the same experience.[/fn]

Desktop publishing enabled researchers to combine their numerical data and the written word and to easily cast it into a form that could be made into articles and books. All it took was a word processor and access to a spreadsheet and/or a database. Nonetheless, sharing information generally required sharing diskettes (or the printed word). Versions of the Total Baseball database were made available on diskettes in the late 1980s and early 1990s, for example. Into the early nineties, many corporations lacked local area networks. Communication among computers, even for small groups of people working together, was difficult at best.[fn]The STATS team of Palmer, Cramer, and Jay Bennett attempted to connect an Apple II with a PDP-10 to facilitate their work. With low speed connections and interface incompatibilities, this proved frustrating, according to Cramer.[/fn] Bulletin board systems (BBS) were just coming into popularity, and AOL was one of the few online services available. That was to change shortly, and the changes would impact tremendously the way information and ideas would be shared.

THE EMERGENCE OF THE WORLD WIDE WEB: 1991-2000

The next significant event in computing, and one that has had the most impact on not only baseball research, but our entire way of life, was the emergence of the Internet and World Wide Web. It has led to the creation of significant resources for research and analysis in every field—and it allows us all to conduct most everyday aspects of our lives online, too.

The Internet started as a loose collection of smaller networks (hence the name) in the mid-to-late 1980s.[fn]The proliferation of networks started with NSFNET, a sort of clone of the second generation ARPANET used by NSF researchers, opening up connections to small commercial networks. The widespread adoption of the TCP/IP protocol (see the next note) was critical to this occurring.[/fn] It remained primarily a research tool, except for a few technology companies and government contractors who connected to it to exchange information. Among the earliest methods of information exchange was email. We can’t take email for granted. Before the ability to swamp one another with daily missives became commonplace, exchanges of statistical information were usually done with typewritten (occasionally printed) pages, sent via snail mail. Bill James remarked in an interview for volume 14 of The Baseball Research Journal that he had received a letter with important information that had led him in a different direction, and then commented that because people are in communication with one another, it is possible to make adjustments.[fn]Jay Feldman, in A Conversation with Bill James, The Baseball Research Journal 14, 26.[/fn] That is very true today, and email was just the first important step forward in increasing the frequency and detail of those communications. Even SABR uses email extensively in communicating with members, both at the national and regional levels, as a way to effectively communicate in a timely fashion—something SABR’s paper-printed newsletters often did not do.

Groups of users exchanged information on Usenet, a kind of global bulletin board system, which was a precursor to a number of social media phenomena that have occurred in the last decade. Usenet was home to a few hardcore analysts, such as Voros McCracken. McCracken published his seminal, and sometimes controversial, work on pitching stats on Usenet, but became widely known when Bill James later highlighted his work. Usenet facilitated exchanges of information, much like modern blogs, but it was still inaccessible to nearly everyone, as only 1 in 10 Americans had access to the Internet in the early 1990s, and only 1 in 10 of them accessed Usenet.

The next step forward in the Internet was the creation of the World Wide Web[fn]The Web, as well as email, newsgroups, streaming media, and other things available through the Internet, are applications. The Internet protocols provide a platform for those. Many people conflate the terms Web and Internet, which is technically incorrect.[/fn] in 1991. The Web was conceived of as a way to interchange not just messages, but documents. Its creation was accompanied by the creation of a new application, the web browser. The Web browser was made popular among universities and research institutions by the NCSA Mosaic browser.[fn]Tim Berners-Lee at CERN wrote the first browser, WorldWideWeb, for the NeXT computer in 1990.[/fn] Developers of that browser left the University of Illinois and independently developed the first commercial Web browser at Netscape Communications; it was released in beta late in 1994. The following year, accompanying the initial commercial shipments and the integration of the Internet’s key communication methods (called protocols[fn]The backbone of the Internet and WWW are the twin protocols, TCP and IP, generally abbreviated to TCP/IP. TCP/IP replaced an earlier ARPANET protocol in 1983. Prior to 1985, it was very difficult, if not impossible, for average users to use TCP/IP on their computers; even computer-savvy individuals had to exert considerable effort to get TCP/IP working on a PC. Windows 95 made it easy, helping usher in the Web.[/fn]) into Microsoft’s Windows 95, the Web blossomed into a true consumer phenomenon.

Table 2. Internet Users in the World.

(Click image to enlarge.)

It took but a few years before the Web started to heavily influence baseball research. In 1995, Sean Lahman made his baseball database available over the Internet. Tom Tango credits the Lahman database with a major increase in his productivity.[fn]Email communication.[/fn] Another significant event was the creation of the Retrosheet website in 1996. Prior to 1996, Retrosheet used a small collection of individuals as well as the resources of teams and sportswriters to gather data, very slowly at first. The introduction of the website started a slow gathering of momentum that increased with each subsequent year. In 1996, the Web was still fairly nascent, even though tens of millions of people were using web browsers. Few people had access to more than slow dial-up connections, which made all but the simplest forms of media frustrating to download. Students in universities and researchers were, again, the earliest adopters of the new technology. And like their predecessors who used workplace mainframes and university computing resources for their baseball analysis, many university students weren’t always using the Web for class research.

By the end of the decade, Internet usage had roughly tripled during the period between 1995 and 2000, and would do so again during the following decade (see the Internet growth graph in Figure 1 at right). Along with the accelerated adoption came both problems and new applications. The latter would benefit baseball analysis greatly.

THE NEXT GENERATION WEB: 2001-2011

In 2000, Sean Forman launched the first of several websites that made up the Sports Reference collection, Baseball-Reference.com. Forman was a professor of mathematics and computer science at St. Joseph’s University at the time. Within six years, Baseball-Reference and its companion reference sites had become big business and Forman left his university job to run Sports Reference LLC full-time. At the same time, Retrosheet used its online outreach to gather and process box scores for nearly every year from 1911 on, and to provide additional detail wherever it is available. (Retrosheet also provides a venue for researchers to publish articles, although this is a secondary contribution for the website and project.)

In 2000, the so-called dot-com bubble burst. In the aftermath of that, however, the Web became even more widely available than it had been before. Cable service and DSL made fast Internet connections available to a large population. New applications, better tuned to users’ needs than failed online retail sites trying to sell pet food, sprang up. First was the search application, driven by Yahoo and then Google. Social network sites emerged, first slowly, and then like a firestorm. The Internet had moved to mobile devices by the end of the decade. Communications technology changed rapidly during this decade, progressing from 9600 baud modems, to 56K, to multi-megabit DSP and cable transmission, and now to wireless mobile phone networks. These advances in the available data rate to all users made access to rich media and massive amounts of data easy.

Widespread use and quick access, along with online databases spurred the development of such applications as UZR, Ultimate Zone Rating, the invention of Mitchel Lichtman. UZR uses data regarding the zone in which a batted ball lands, data that were previously not recorded by scorekeepers or Major League Baseball. Volunteers recorded such data and uploaded them. Without such data, Mitchel Lichtman would not have been able to develop his innovative approach. Today, game-time information can be kept on a tablet device and immediately uploaded to the appropriate database. However, with online applications such as MLB.com’s Gameday, and the development of MLBAM’s Fieldf/x system (in which the flight of every batted ball and the positioning of every fielder is digitized and recorded) even this approach will become obsolete.

If the PC era democratized analyzing baseball data, the Internet blog[fn]Short for weblog, originally a primitive way to keep an online diary.[/fn] democratized reporting and editor-ializing about nearly every aspect of baseball. It is difficult to determine the exact number of baseball blogs—they certainly run into the millions—but a large number have become popular with avid fans (and statheads like the author). Some are regional, some are devoted to teams. Some websites, such as FanGraphs, are dedicated to the analysis and reporting of the most esoteric of baseball stats like UZR and WAR. FanGraphs also publishes its own Power Ratings. Baseball Prospectus was founded in 1996 and grew out of a group of statistically minded friends who met through Usenet. Baseball Think Factory, Diamond Mind, Bleacher Report, and many other websites feed information to a number of amateur baseball historians and stats fanatics.

The other phenomenon that is likely to have a powerful effect on communicating among the community is social networking, with both social sites such as Facebook and professional sites like LinkedIn enabling people to connect one-to-one and in groups. This kind of online communication replaces the old mailing list and telephone tree with immediate contact and spread of information, allowing for easier harnessing of the group mind and faster propagation of new ideas and research. 

THE IMPACT ON RESEARCHERS AND THE REST OF US

In 1971, there were few computer professionals. Most researchers into baseball stats used computers as an adjunct. Few universities offered degrees in information technology and a handful of companies employed the majority of IT professionals. That has all changed. There are more of us now who are trained in information technology—Sean Forman and Sean Lahman are good examples. But more than that, high school kids are now taught to program as part of their regular curriculum and even those who don’t program can easily create blogs and websites. Desktop applications like Microsoft Excel and Access can turn non-programmers into data processing mavens. Social sites and email allow us to stay in constant communication and exchange ideas. People may be unaware, or may have forgotten, that the World Wide Web was invented as a tool for sharing research, not selling ads or pet food. The web continues to be effective in that vein today.

The ability to analyze massive volumes of statistics has had a significant impact on ballclubs, as well. Many employ consultants to do statistical analysis as a way to better deploy players, to determine best trades, and to help in salary negotiations. Some, like the Boston Red Sox, have departments dedicated to information services. The Red Sox have employed James and Tippett as well as other SABR members, the Mariners Tom Tango, the Orioles Eddie Epstein, the Phillies and Braves Pete Palmer. Tango has also done work for the St. Louis Cardinals, along with Lichtman.

It goes beyond the researcher. Tablets and home computers provide a great platform for watching or listening to games, providing relevant stats and information live during streaming of game media. This author has observed his fellow fans Googling baseball data on their phones to settle arguments, or merely to keep up to date. We can all look forward to more leverage of our current and future computing environments for research, as well as to enhancing the experience of the ordinary fan.

When SABR was founded, computing was on the cusp of moving from an expensive tool for research and big business to a part of our everyday lives, a process that has paralleled the evolution of sabermetrics and of SABR as an organization. We can only expect more interesting perspectives of the game to emerge from advances in information technology, something to look forward to.

ACKNOWLEDGEMENTS

The author is indebted to the assistance of Dick Cramer, Gary Gillette, Pete Palmer, Mark Pankin, Tom Tango, as well as Sean Lahman and Sean Forman. They have helped recreate the history of sabermetrics, supplied the record of who was contributing when and what, and provided important data. The author would also like to thank the Computer Museum in Palo Alto for their work in maintaining the historical record of computing. Also, without such companies as Intel, Netscape, Microsoft, Google, and others, we would not have the rich computing environments that allow us to do the work we do today.

RICHARD SCHELL has been involved with computing for decades. At the time SABR was founded, he was finishing his BA in Mathematics and Computer Science, while working as a computer programmer affiliated with the PLATO Project at the University of Illinois. Since then, he has held numerous technical and senior management positions in the industry with companies such as Intel Corporation, Sun Microsystems, and Netscape Communications. Dr. Schell is currently a venture partner with ONSET Ventures. He is an avid baseball fan and a San Francisco Giants season-ticket holder. Dr. Schell has been a SABR member since 2007.

Washington was born on November 16, 1951, in Belzonia, Mississippi, but moved to Flint, Michigan when he was less than one year old. Washington and his family moved so his parents could work in the auto plants.[fn]“Herb Washington: World Record Sprinter & Business Success, ” MSU Spartans Alumni website, 19 February 2007, accessed 1 August 2011, www.msuspartans.com/genrel/021907aac.html.[/fn] He started high school at Flint Northern, but was ruled ineligible to compete in track and field because of a school district boundary dispute. After 10th grade, he transferred to Flint Central High School so that he could compete. At Flint Central he met the track coach, Carl Krieger, who helped Washington become one of the greatest runners in the world. “Coach Krieger had a vision and got me to believe,” Washington would later say. “He said the only thing holding me back was me.”[fn]Ibid.[/fn]

After receiving many scholarship offers in both track and football, “Hurricane Herb” chose Michigan State, where he ran track and field for the Spartans  1969–1972. In one of his races as a freshman, after leading most of the race, he lost to John Carlos, a member of the 1968 United States Olympic team infamous for their Black Power Salute. As a sophomore he beat Carlos at the 47th Michigan State Relays. After that, he was known as a better short distance runner than Carlos. By the time Washington graduated he had won one NCAA title, seven Big Ten titles, was a four time NCAA All-American, and had set world records in the 50-yard and 60-yard dashes.[fn]Ibid.[/fn]

He was drafted by the NFL Baltimore Colts but it wasn’t the right job for Washington. Instead he took a job broadcasting sports for Channel 6 in Lansing, Michigan. It was there, in January 1974, when he got a message he did not expect—from Charlie Finley.[fn]Ibid.[/fn]

Charles Oscar Finley bought the Kansas City Athletics in 1960 and moved the team to Oakland in 1968. He was one of the most colorful baseball owners in history. His legendary stunts included Charlie O (a mule mascot) and green and gold uniforms that were visible from a mile away. He was an outspoken supporter of the newfangled DH rule. The year was 1974 and he was about to make another memorable move.[fn]Nick Acocella, “Finley entertained and engaged,” ESPN.com, Accessed 1 August 2011, http://espn.go.com/classic/biography/s/Finley_Charles.html.[/fn]

Washington received a message from Finley and the Athletics wondering if he wanted to play for them. He thought it was a joke. He hadn’t played organized baseball since junior high. Then he got paged by the A’s. It was no joke. Finley told Washington, “Herbie, I want you to play and be a pinch runner.”[fn]MSUSpartans.com[/fn]

The four time All-American decided the idea was worthwhile. The contract negotiations were rough. Washington wanted a no-cut contract. Finley was in shock. He told Washington that a no-cut contract was for players like Catfish Hunter, Vida Blue, and Reggie Jackson. Washington replied by saying, “They can’t out-run me!” After holding out for a while, Washington got the contract he wanted. A one year, no-cut contract for $45,000 with a signing bonus of $20,000. Included in the contract was a clause that required him to grow a mustache before Opening Day. Washington, however, couldn’t grow one little hair. To get the $2,500 bonus that came with growing it, he used an eyebrow pencil to draw a believable mustache and got the bonus.[fn]MSUSpartans.com[/fn] He also got a base-running tutor in former base-stealing champ, Maury Wills.[fn]Franz Lidz, “Whatever Happened To… Herb Washington,” Sports Illustrated, Volume 79 Issue 3 (1993), 86.[/fn]

Not everyone was as sure as Finley that Washington would be able to make a contribution to the team. “He’s a great athlete, but he’s not a baseball player,” said future Hall of Famer Reggie Jackson at the time.[fn]Eric Prewitt, “Herb Washington On Waivers,” The Portsmouth Journal, (1975), 12.[/fn] Washington’s career as “designated runner” got off to a shaky start. Not only was he unsuccessful in four of his first five attempts, he was also blamed for Oakland’s poor start. A’s pitcher Catfish Hunter set the critics straight: “There has been a lot of talk about Herbie, but he don’t hit and he don’t pitch, so it ain’t him.”[fn]Lidz, Sports Illustrated.[/fn]

For the rest of the season, both Washington and Oakland rolled. He ended up playing in 92 games, stole 29 bases, and scored 29 runs.[fn]“Herb Washington”, Baseball-reference.com[/fn] During one stretch he went 16 for 19.[fn]Lidz, Sports Illustrated.[/fn] Oakland, led by Cy Young winner Catfish Hunter (25–12), and by Vida Blue (17–15), Ken Holtzman (19–17), Rollie Fingers (9–5, 18 saves), and Reggie Jackson (.289, 29 HRs, 93 RBIs), ended the season with a record of 90–72.[fn]“1974 Oakland Athletics,” Baseball-Reference.com[/fn]

Herb Washington had many unusual moments. Even though he hadn’t played since junior high, he knew baseball. Once, Milwaukee Brewers first basemen George Scott told him he needed him to move so he could clean the base. Washington, having seen the ball in Scott’s glove, called time. Manager Alvin Dark once asked if he wanted to pinch-hit against Nolan Ryan. Washington said no, not because he was afraid to face the fireballing Ryan, but because he didn’t want to lose his uniqueness: a non-pitcher who had spent a whole season on a major league roster and yet had never once been to the plate.[fn]Lidz, Sports Illustrated.[/fn]

After winning the 1974 American League Championship Series three games to one against the Baltimore Orioles, the Athletics faced the Dodgers in the World Series. Finley set the bar high for Washington, saying, “He is going to rise and shine in the ’74 World Series.”[fn]“1974 Oakland Athletics,” Baseball-Reference.com[/fn] Finley was partially correct, Washington did become known.

The A’s won the first game. Then they were down by one run in the ninth inning of Game Two when Washington entered the game. Representing the tying run, Washington took his place at first base next to Dodgers NL MVP first baseman Steve Garvey, also a Michigan State alumnus. On the mound was another former Spartan, NL Cy Young Award winner Mike Marshall. Marshall actually taught both players at Michigan State.[fn]Lidz, Sports Illustrated.[/fn]

Everyone in Dodger Stadium knew what Washington was supposed to do: steal second base. The right-handed Marshall stepped off and threw over, and Washington got back easily. Then Marshall attempted to pick him off again and this time he stunned Washington. Washington dove back head first but he was too late; Garvey placed the tag. The NL MVP recalled the play: “He was frozen. We wanted to catch him flat-footed and we did.” Washington pounded the ground, saying “Damn!” The misstep gave the Oakland A’s little chance to win the game. But Washington did get his ring; Game Two was the only game the A’s lost as they won the Series 4–1. Washington appeared in three of the five games.[fn]Kevin Lamb, “Marshall’s Pickoff Ends Oakland Rally,” Milwaukee Journal (1974), 5.[/fn]

The next season came. The A’s were rolling when Finley cut the struggling Herb Washington on May 5, 1975.[fn]Ibid.[/fn] He had played in 13 games with only two steals that season. Said A’s team captain Sal Bando, “I’d feel sorry for him if he were a player.”[fn]Prewitt, The Portsmouth Journal.[/fn] [fn]Sarah Pileggi, “They Said It,” Sports Illustrated, Volume 42, Issue 20 (1975), 18.[/fn] His cumulative stats were 105 games, 31 steals, and 33 runs scored.

He was caught stealing 17 times (not a very good percentage for a base stealer). He had no plate appearances and never took a position in the field. Herb Washington’s baseball career was over.

Instead, he kept up his success in his business career. Washington is the owner of many McDonald’s restaurants in Pennsylvania and Ohio. He has held positions in the Federal Reserve Bank of New York and he is active in many community organizations. Washington is married and is the father of two children.[fn]Eric Prewitt, “Herb Washington On Waivers,” The Portsmouth Journal, (1975), 12.[/fn], [fn]MSU alumni donations newsletter, Summer 2010, Accessed 1 August 2011, www.egr.msu.edu/files_egr/content/Developments_Summer2010-Consumers4-5.pdf.[/fn]

PETER WARREN is a 12-year-old, seventh-grade student from New Jersey. He has always been a fan of baseball. His interest in baseball history and statistics began when he was given a Baseball Statistics book at age six. He plays travel baseball and basketball in his hometown.

• In 1976, when the DH was used for the first time in the World Series, DHs for the Yankees hit .063 (1-for-16) while the Cincinnati Reds’ DHs hit .357 (5-for-14). The Yankees used three players in the DH role, while the Reds used only one: Dan Driessen. The Reds swept all four games.
• During all World Series in the 1970s, the National League DHs outhit the American League DHs .281 (9-for-32) to .256 (10-for-39).
• Average runs scored per World Series game in the 1970s were 4.45 when the DH was used, only 3.73 when not.
• In 1982, the NL St. Louis Cardinals’ DHs outhit the AL Milwaukee Brewers’ DHs .429 (12-for-28) to .125 (3-for-24). The Cards beat the Brewers in 7 games.
• In 1984 the NL San Diego Padres’ DHs hit .412 (7-for-17, including 2 home runs). The AL Detroit Tigers’ DHs hit .059 (1-for-17). However, the Tigers won the Series, 4­–1.
• In the 1981 World Series, the pitchers on both teams (Los Angeles Dodgers and New York Yankees) went a combined 0-for-26.
• In the 1983 World Series, the pitchers on both teams (Baltimore Orioles and Philadelphia Phillies) went a combined 1-for-23.
• Pitchers “hit” the height of futility in the 1985 World Series when all the pitchers batted .000, going a combined 0-for-30 in the all-Missouri World Series (Kansas City Royals versus the St. Louis Cardinals).
• During the 1980s, the National League DHs outhit the American League DHs .303 to .216. During that time, American League pitchers had 1 hit in 69 at-bats for an average of .014. The average number of runs per game was 4.25 when the DH was utilized and 3.84 when the pitchers hit.
• During the 1988 World Series, the Los Angeles Dodgers’ pitchers were perfect hitting 1.000 (3-for-3) while the AL Oakland A’s’ pitchers were the opposite, hitting .000 (0-for-4). The Dodgers won the Series, 4–1.
• During the 1990s, the average number of runs per game was 4.70 when the DH was utilized and 4.04 when the pitchers hit.
• In 1997 the NL Florida Marlins were in their first Fall Classic in their brief history. Their DHs hit .500 (6-for-12) while the Cleveland Indians’ DHs hit .182 (2-for-11). The Marlins won the Series, 4–3.
• In 2003, the Marlins returned to the World Series against the Yankees. New York’s DHs hit .125 (1-for-8) while Florida’s hit .111 (1-for-9). The Marlins won the Series, 4–2.
• In 2005, DHs for the AL Chicago White Sox hit .429 (3-for-7) while the DHs for the Houston Astros hit .167 (1-for-6). The White Sox swept the Series.
• It is natural to assume that when the DH is utilized, starting pitchers will go further in the game than when they hit. However, this was not the case during the 1980s, when starters averaged only 5.65 innings per start in DH games, but 5.84 innings when they had to hit.
• There have been only three World Series where both teams used the same player as DH throughout the series:
    1989: Oakland Athletics, Ernest Riles; San Francisco Giants, Dave Parker
    1992: Toronto Blue Jays, Dave Winfield; Atlanta Braves, Lonnie Smith
    2005: Chicago White Sox, Carl Everett; Houston Astros, Jeff Bagwell
• The Detroit Tigers used the most DHs in a single World Series game when they used 4 in Game Five of the 1984 World Series.
• The most impact a DH had on a World Series game came in the Game Six of the 2009 World Series played between the New York Yankees and the Philadelphia Phillies. The Yankees DH, Hideki Matsui, contributed to six of the seven runs his team had in the game. Matsui batted .600, the highest average for a DH in a World Series, and was also the first DH to win the World Series Most Valuable Player Award.

JOHN CRONIN has been a SABR member since 1985 and currently serves on the Minor League Committee. Cronin is a lifelong Yankees fan and his area of special interest is Yankees minor league farm teams over the years. He is a C.P.A. and a retired bank executive, who has a B.A. in History from Wagner College and an M.B.A. in Accounting from St. John’s University.

Sources

• Baseball-Reference.com
• Enders, Eric. 100 Years of the World Series, Barnes & Noble Publishing, Inc., 2003.
• Okkonen, Marc. The Federal League of 1914-1915 Baseball’s Third Major League, The Society for American Baseball Research, 1989.
• Vescey, George. Baseball: A History of America’s Favorite Game, New York: Random House, 2006.
• www.Baseball-Almanac.com

Until Weaver called attention to it, the concept of only twenty-seven outs was—and perhaps still too often is—lost in the actual playing of the games. Managers may resort to various stratagems to try to seize an advantage, particularly in the early and middle innings, at the expense of thinking about the long arc of a game. Weaver’s constant awareness of the potential importance of each out was at the very core of the offensive strategies that characterize his legacy.

WEAVER’S GRAND STRATEGY

Earl Weaver as a grand strategist is remembered today for his philosophy on the merits of the three-run home run (he subtitled the chapter on offense in Weaver on Strategy “Praised Be the Three-Run Homer”), disdain for run-creation strategies (they cost outs and are inefficient), and deliberately building his bench with pinch hitting and substitution options for position players in anticipation of batter-pitcher matchups. This also allowed him to platoon in his daily starting line-up with nearly wild abandon. “You need someone for each job that needs to be done when the time arises,” he wrote.[fn]Weaver on Strategy, 25.[/fn] Weaver was a manager with a plan for every individual player on his team. The Earl of Baltimore did not want mere emergency back-up players on the bench, who rarely played but were there when needed. He didn’t just want “utility” players who could play a variety of positions, or a power bat from each side of the plate. He had in mind a specific set of roles—that’s plural—for each player sitting on the Baltimore Orioles’ bench. He knew what each could do in particular situations against particular pitchers, and used them often and accordingly. Everybody played, and he expected them all to contribute.[fn]The Bill James Guide to Baseball Managers From 1870 to Today (Scribner, 1997), 243–245.[/fn]

The philosophy now identified with Weaver reflects where he was at the end of his brilliant career, but the core elements were in place at least a half-decade before he retired following the 1982 season. His philosophy began to evolve the day he took over the Orioles in mid-season 1968. While the basic principles of Weaver’s grand strategy were perhaps always there for him, they developed into a coherent narrative through his experiences in the first half-dozen years or so that he managed the Orioles. At least, that’s what the data indicate from the games Earl Weaver managed.

Nineteen seventy-seven appears to have been the year where Weaver’s application of his strategic principles first became consistently discernible. The following analysis is based on data compiled through the extraordinary efforts of Retrosheet and available on the website Baseball-Reference.com for Earl Weaver’s 14 full seasons as Baltimore Orioles manager from 1969 to 1982.[fn]Weaver returned for an encore at the request of Baltimore’s owners in mid-season 1985 and managed the Orioles for all of 1986, but this was an interruption in his retirement to try to right a suddenly-failing franchise. (Weaver on Strategy, 189.)[/fn] 

THE HOME RUN IS ABOUT SCORING BIG

The fundamental premise of Weaver’s offensive strategy was: “In my mind, the home run is paramount, because it means instant runs.”[fn]Weaver on Strategy, 34.[/fn] Since the end of the Deadball Era, any manager who had sluggers on his team loved the three-run home run. Miller Huggins certainly did with Ruth and Gehrig, Joe McCarthy with Gehrig and DiMaggio, Casey Stengel with Mantle and Berra. Even McGraw—who was a loud-mouthed evangelist for “scientific” small ball as the best way to score runs[fn]Lyle Spatz and Steve Steinberg, “1921: The Yankees, the Giants, and the Battle for Baseball Supremacy in New York,” The Baseball Research Journal, Fall 2009 (Volume 38, number 2), 12–13.[/fn]—surely appreciated first baseman George Kelly as one of the National League’s premier home run hitters when the Giants won four straight pennants from 1921 to 1924, not to mention young slugger Mel Ott who hit 42 home runs for him in 1929. Small ball has its place, but as Weaver said, there is no more sure or faster way to score runs than with a home run, especially with runners already on base.

None of Earl Weaver’s teams, however, ever led the American League in homers, which of course belies his faith and commitment to the home run. In his 14 full seasons as manager, Weaver had only two of his players finish in the top three in home runs—Eddie Murray, who tied with three others for the league lead in the strike-disrupted 1981 season, and Reggie Jackson who finished second in 1976 (his only year in Baltimore). But though the Orioles under Weaver never had more than the third-highest home run total in the league, when they did peak at third, it coincided with their pennant-winning years of 1969, 1970, and 1979, then again in 1976, 1978, and 1982. Weaver nonetheless usually did have a pair of dangerous power hitters who delivered in the middle of his line-up—Frank Robinson and Boog Powell when he won three straight pennants from 1969 to 1971; Lee May and Don Baylor in 1975; May and Reggie Jackson in 1976; May and Eddie Murray in 1977 and 1978; Murray, Ken Singleton, and his left field platoon of Gary Roenicke and John Lowenstein from 1979 to 1982. These guys could make the three-run home run happen with some measure of frequency. The only years Weaver’s Orioles lacked consistent power were 1972 to 1974, after Frank Robinson had been traded away and as Powell got older, and before the arrivals of May, Singleton, and Murray.

THE ORIOLES’ SACRIFICES

It is well known that Earl Weaver hated the sacrifice bunt as wasting a precious out: “There are only three outs an inning, and they should be treasured. Give one away and you’re making everything harder for yourself.”[fn]Weaver on Strategy, 38.[/fn] It is not that Weaver thought that bunting was never appropriate—“I’ve got nothing against the bunt—in its place”[fn]Ibid., 33.[/fn]—but that its place was limited to where it gave his team a chance to win a game in the late innings.

None of his teams ever led the league in sacrifices, but in Weaver’s first seven full seasons as manager, the Orioles were in the top three of the 12 American League teams in sacrifice bunt attempts every year. In four of those years, Baltimore had more sacrifice attempts than any other team in the league, and twice the second most. We must allow, however, for the fact that until 1973, when the designated hitter rule went into effect, American League pitchers were still required to hit for themselves. Exactly half of the 288 sacrifice bunts laid down by the Orioles from 1969 to 1972 were by the pitcher, according to the data, and Weaver’s pitchers had more opportunities to attempt a sacrifice both because the on-base-percentage (OBP) of Baltimore batters (not including pitchers) in the bottom third of the order was generally comparable to the team’s OBP as a whole, and because Baltimore’s superior starting rotation often pitched late into games allowing them more plate appearances. 

Even after the DH came into being, however, the Orioles still led the league in sacrifice attempts the first two years, and were third the year after that (1975). But these years coincided with Baltimore’s (temporary) drop in home run power, previously mentioned. It wasn’t until 1977 that the number of sacrifice attempts ordered by Weaver dropped dramatically. In 1978 and 1979, the Orioles were last and next to last in the league in attempted sacrifices, in 1980 they attempted the third fewest, and in 1981 the fifth fewest. After averaging 99 sacrifice attempts per year from 1973 (the first year of the DH, so no pitchers batting) to 1976, Weaver’s Orioles averaged 65 per year from 1977 to 1982, not including the 1981 strike-shortened season, a 34-percent drop.

There are two other indicators of Weaver disdaining the sacrifice even more after 1977 than before. From 1969 to 1976, according to the data, 64 percent of Baltimore’s sacrifice bunts came in victories and 22 percent in the seventh inning or later with the score tied, the Orioles ahead by one, or behind but with the potential tying run on base, at the plate, or on deck. From 1977 to 1982, by contrast, 75 percent of the Orioles’ sacrifices came in victories, and only 15 percent in the seventh inning or later in close games. This means that Weaver became more reluctant to bunt both when his team was losing and when his team needed a run in a close game in the late innings. In either situation, it is apparent Weaver was more inclined after 1977 to let his hitters hit—looking for that home run to instantly make up ground, or for hits to get more runners on base and drive in runs.

Weaver was consistent throughout his managerial career in rarely bunting a runner over when there was already one out in the inning, except when American League pitchers had to bat. It is worth noting, however, that with a runner on second base and first base open, including with nobody out, Weaver’s Orioles laid down a total of only 11 sacrifice bunts in the six years from 1977 to 1982, compared to 32 in the four years from 1973 (the beginning of the DH) to 1976. In this circumstance, after 1977, Weaver was most clearly not of a mind to sacrifice an out to get a runner already on second over to third; better to try to drive him home with a hit.

Weaver’s disdain of the sacrifice bunt was further validated by the Orioles’ routinely poor percentage in successfully laying them down. The Orioles had a worse success rate in sacrifice attempts than the league average every one of Weaver’s 14 full seasons at the helm, generally far below average. Weaver’s Orioles actually attempted 112 more sacrifice bunts than the average American League team, but were successful only 63 percent of the time, while the AL average was 76 percent. Weaver himself wrote: “Since a manager should only have those hitters bunt who have the skill, about 80 percent of your sacrifices should be successful.”[fn]Ibid., 41.[/fn] Mastering the bunt may not have been part of the famed “Orioles Way,” but the Orioles’ frequent inability to bunt successfully certainly made going for the base hit—or better yet, the home run—a more compelling option.

ORIOLES NOT SO PRODUCTIVE IN MAKING PRODUCTIVE OUTS

The 1970s and early-1980s Orioles generally were not very good in advancing base runners, perhaps because this was not important to Weaver even at the beginning of his managerial tenure. Besides disdaining the bunt, Weaver did not want to sacrifice outs just to move a runner over, whether by hitting behind the runner or even the more dynamic hit-and-run. He considered the hit-and-run “the worst play in baseball”[fn]Ibid., 47.[/fn] In the statistical category of making productive outs—defined by Baseball-Reference.com as a batter advancing any runner with nobody out or driving home a runner with the second out of an inning—the Orioles were below the league average every year Weaver was manager except for 1969, 1973, and 1974.

While Weaver’s Orioles remained relatively consistent in percentage of productive outs,  1977 represents a break point. From 1969 to 1971, while winning three straight pennants, Baltimore batters made productive outs in 27 percent of their opportunities with runners on base, compared to the American League average of 30 percent. And from 1972 to 1976, most years in which Weaver did not have an imposing pair of power hitters in the heart of his batting order, Orioles’ hitters made productive outs in 32 percent of their chances to move up or score a runner, when the league average was only slightly higher at 33 percent. But from 1977 to 1982, Baltimore’s 32 percent success rate in productive outs was substantially below the league average of 37 percent.

The frequency of Baltimore batters grounding into double plays relative to the rest of the league is yet another metric that suggests Weaver’s disdain for small ball and his desire for  his batters to hit with runners on base really took off in 1977. Since Weaver did not put on plays to advance his base runners, the potential for double plays increased. Indeed, wrote Weaver, “About the only thing you can say for the hit-and-run is that it prevents the double play grounder.”[fn]Ibid., 47.[/fn] In Weaver’s first eight full seasons as manager—1969 to 1976—his Orioles hit into fewer double plays than the league average every year except for 1969. But from 1977 till Weaver retired in 1982, Baltimore not only hit into more double plays than the league average, but substantially more; the 820 double plays Weaver had to watch his team hit into from the dugout those six years was 11 percent more than the league average of 741 per team.

STOLEN BASES OK

While the stolen base was one small ball strategy that Weaver did have use for, he also warned that “the failed stolen base attempt can be destructive, particularly at the top of the order, because it takes the runner off the base paths ahead of your home run hitters.”[fn]Ibid., 44.[/fn] Unlike sacrifice bunts and hitting behind the runner, at least a caught-stealing did not give up the batter as an out. With talented base stealers like Don Baylor in the mid-1970s and Al Bumbry after that, Weaver was willing to regard the stolen base attempt as an acceptable risk. In his 14 years as Baltimore manager from 1969 to 1982, Weaver had at least one player steal 20 or more bases sixteen times, including three players in 1974 and  1976.

Until his last two years as manager, when Baltimore was fourth and third in the league in fewest steals, Weaver appears to have been comfortable with the stolen base as a tactic. The Orioles led the league in steals in 1973, were third in 1974, and were typically well above the league average in successful stolen base attempts. There were only three years—including the last two before he retired (the first time)—that Weaver’s Orioles were below the league average in stolen base percentage. Nonetheless, his Orioles were successful 66 percent of the time during his tenure, substantially below the 75 percent Weaver believed was necessary for the steal “to be worthwhile.”[fn]Ibid., 44.[/fn] 

WEAVER AND THE PLATOON ADVANTAGE AT BAT

Earl Weaver was also acclaimed for seeking to gain the advantage in batter-pitcher match-ups during games when it mattered most. Although other managers long before him were aware of how their batters matched up against certain pitchers, Weaver was probably the first to be so systematic about it. In the late 1940s, the Brooklyn Dodgers’ statistician, Allan Roth, pioneered the keeping of these kinds of statistics, but outside of Burt Shotton, Dodgers managers chose not to rely on them.[fn]Joshua Prager, The Echoing Green: The Untold Story of Bobby Thomson, Ralph Branca and the Shot Heard Round the World (Vintage Books, 2008), 183[/fn] For the most part, when managers went to their bench to play the percentages with a pinch hitter, all that basically mattered was who was available who batted from the opposite side the pitcher threw. There was, of course, strategizing as to who to use when—managers would choose when to pinch hit a slap-hitting speed guy to get a rally started and when to send in a big bopper to drive in runners or to make it a game again with one mighty blow. But it was Weaver who was ahead of his time in using statistical-based analysis to determine the make-up of his bench and decide at potential turning points in the game which of his hitters had the best chance of coming through in the clutch.[fn]Weaver on Strategy, 25–26, 52–53[/fn] These were informed decisions by Weaver, based on players’ past performance.

Earl Weaver almost always had a regular platoon at one position or another—most famously the left-handed batting John Lowenstein with the right-handed batting Gary Roenicke in left field from 1979 to 1982—and sometimes he platooned players at two or three positions at any one time. But for most of the first half of his career as Baltimore manager, Weaver’s seeking the platoon advantage in any given at bat was not out of the ordinary compared to other managers. From 1969 to 1976, there was not one single year in which the Orioles exceeded the league average in percentage of plate appearances where they had the platoon advantage—defined as a batter facing a pitcher throwing from the opposite side. Not one. American League teams on average during those eight years had a platoon advantage against the pitcher in 54 percent of their plate appearances; for Weaver’s Orioles, the annual average was 51 percent.

Aside from 1975, when the Orioles had the platoon advantage in 45 percent of their plate appearances, compared to 55 percent for the league as a whole, it is not surprising that the biggest gap between the Orioles and the rest of the league—9 percent fewer plate appearances with the platoon advantage—was when Baltimore dominated the American League by winning three consecutive blowout division titles and pennants from 1969 to 1971, during which they also led the league in scoring twice and were second in 1969. These were the years when Weaver had his most stable daily line-up, with Boog Powell at first, Davey Johnson at second, Mark Belanger at short, Brooks Robinson at third, and Don Buford, Paul Blair, and Frank Robinson in the outfield. Powell and Buford (who was a switch hitter) were the only left-handed bats among his core regulars. That’s every regular position on the field except catcher, where Weaver platooned the left-handed Elrod Hendricks with the right-handed Andy Etchebarren.

Although Weaver gave Frank Robinson and Buford more days of rest in 1970 and 1971, his seven core players at the seven positions behind the pitcher each met the major league minimum of 502 plate appearances to qualify for the batting title in every one of those three years. After 1971, Weaver’s teams never had more than five players in any given year play 130 or more games in the field and have enough plate appearances to qualify for the batting title, except in 1975 (he had six), when the Orioles had their biggest deficit in the platoon advantage relative to the rest of the league. In five of Weaver’s ten seasons as manager after the DH came into effect, the Orioles’ designated hitter also had enough plate appearances to meet the major league minimum.

Once again, 1977 was the year Weaver changed his dynamic. From 1977 through 1982, Weaver managed his offense to ensure that the Orioles had the platoon advantage in any given plate appearance 64 percentage of the time—substantially higher than the 57 percent averaged by the now-fourteen teams in the American League. In Weaver’s last three years as manager before he decided to retire, the Orioles were second in 1980 and first in the league in 1981 and third in 1982 in platoon advantage at the plate. Having a pair of dangerous switch hitters from either side of the plate all six of those years in Eddie Murray and Ken Singleton, who could always bat from the opposite hand of the pitcher, certainly helped Weaver attain such a high platoon advantage.

In the DH era, AL managers wanting a platoon advantage at the plate had to pinch hit for position players. With his carefully constructed bench, and his copious notes on Orioles batters versus particular pitchers, Weaver gained a deserved reputation as the mastermind of pinch hitting. In 1977 this Earl Weaver trait became clearly apparent. In his first three full seasons as manager, Weaver sent the fewest pinch hitters to the plate of any AL manager twice—in 1969 and 1971—and only Boston’s manager Eddie Kasko made fewer in 1970. This statistic is misleading, however, because pitchers still had to bat in the American League and Weaver had the best starting rotation on the most dominant team in the league. His staff had the lowest starters’ ERA in the league all three years, had the highest percentage of quality starts, and averaged the most innings pitched per start in the last two years, and were second the first year.

With pitchers the caliber of Jim Palmer, Mike Cuellar, and Dave McNally anchoring his rotation all three years—joined by Pat Dobson in 1971 to become only the second major league team in history (after the 1920 Chicago White Sox) to have four 20-game winners—and the Orioles also prolific in scoring runs, Weaver had little occasion to pinch hit for his pitchers. Moreover, Weaver was willing let his starters go the distance if they had the lead and were pitching well. Sparky Anderson, Weaver was not.

Once the DH came along, Weaver was generally in the mainstream when it came to pinch hitting for position players through 1976. In the first four years of the DH, the Oriels averaged 107 pinch-hit at bats per year, five percent fewer than the league average per team. Then from 1977 to 1982, Weaver sent a pinch hitter to the plate an average of 150 times per season, 29 percent more often than the league average of 137.

WEAVER’S POSITION PLAYER SUBSTITUTIONS

All of this is consistent with the total number of in-game position substitutions Earl Weaver made as manager of the Baltimore Orioles, with 1977 once again a clear delineation. In his first four full seasons as manager, Weaver was not at the high end of managers substituting for position players. But of course, the quality of Baltimore’s core regulars those years—at least from 1969 to 1971—was such that there was likely little to be gained for Weaver to replace players during games beyond resting his starting position players every now and again. Weaver made an average of 159 position player substitutions per year while winning three straight pennants, allowing his position players to complete 88 percent of the games they started, according to data in Retrosheet.[fn]Position player substitutions are the difference between the total number of player games at each position (other than pitcher) and the total number of position games, which is calculated as eight positions (not including pitcher) times the number of games played by the team that year.[/fn] In 1972, the last year American League pitchers had to hit for themselves, Weaver made 198 position player substitutions, allowing his starters to stay in the game 84 percent of the time.

In the first four years of the DH, Weaver began rising to the top tier of American League managers making position player substitutions, averaging 215 per season. The percentage of games completed by the Orioles’ starting position players dropped from 87 percent in 1974, to 83 percent in 1975, and 81 percent in 1976. The major reason for this was that power-hitting first baseman Lee May, who was not especially gifted defensively, completed only 66 percent of the 142 games he started at first in 1975 and 52 percent of the 93 games he started there in 1976. Weaver also started May at DH in 51 games in 1976.

Beginning in 1977 until he retired after the 1982 season, and not counting the strike-shortened year of 1981, Weaver never again made fewer than 300 in-game position substitutions, usually precipitated by pinch hitting for an advantage. And in 1981, the 239 substitutions he made projected to equal 369. Weaver made 387 position player substitutions in 1977—52 percent more than in 1976—and the percentage of position players who completed the games they started was down to 71 percent. May and rookie Eddie Murray—both of whom also started in the DH role— completed only 55 percent of their starts at first base. Over the next five years, Weaver’s starting line-up of position players (not including the DH) completed only 76 percent of their games, meaning that on average in any given game, two of his eight starters who took the field in the first inning were out of the game before it ended.

The jump from the annual average of 192 position substitutions in Weaver’s first eight seasons to 312 in his last six years beginning in 1977 was dramatic and indicative of Earl Weaver consolidating into enduring practice two of his most fundamental tenets. One, he prized fielding to back up his pitching staff. Two, he prioritized batter match-ups at key moments of the game. Particularly in the last half of his tenure as Baltimore manager—when he no longer had seven position players who should indisputably take the field every day—Weaver put together teams that had redundancy at virtually every position, giving him options for both pinch hitting and defensive replacements.

The only position players who were assured the likelihood of finishing the games they started in Weaver’s final six years as Baltimore manager were first baseman Eddie Murray, third baseman Doug DeCinces (until he was traded to the Angels before the 1982 season), center fielder Al Bumbry (most years), and shortstop Cal Ripken in 1982. Even Ken Singleton, one of the best outfielders in the American League, was in the game at the end of only 52 percent of the 643 games he started in right field from 1977 to 1981. Despite Singleton’s prodigious offensive production—averaging 23 home runs, 89 RBIs, and .301 at the plate—Weaver felt his defense left much to be desired. By 1982, Singleton was the Orioles’ DH. When he had his two best years in 1977 (24 home runs, 99 RBI, a .328 batting average, and a .438 on-base percentage) and 1979 (35 home runs, 111 RBI, a .295 batting average, and a .405 on-base percentage), Singleton still completed only 61 percent of the games he started in right field both seasons.

The outfield was where Earl Weaver made the preponderance of his in-game substitutions from 1977 to 1982. For those six years, outfielders in his starting line-up were still in the game when the final out was made only 65 percent of the time. In some respects, Weaver’s frequent in-game substitutions of outfielders harkened back to George Stallings’ model with the 1914 “miracle” Boston Braves, the year Stallings made platooning famous. Weaver had Singleton and Bumbry starting most often in right and center and used Roenicke and Lowenstein as a platoon in left (whereas Stallings had, maybe, only Joe Connolly of that caliber among his outfielders). But Weaver also always had some combination of the left-handed Pat Kelly, Tom Shopay, Larry Harlow, and Jim Dwyer and the right-handed Andres Mora, Carlos Lopez, and Benny Ayala on his bench to pinch hit and play the outfield as needed, which was often.

The Orioles’ infield was also not immune to Weaver’s substitutions. From 1978 to 1980, Weaver started Kiko Garcia at shortstop in 237 games—almost exactly half of the total games Baltimore played, with Mark Belanger starting most of the rest—but kept him in the game in only 130 of them. Belanger, one of the best defensive shortstops ever, usually finished up for Garcia. Meanwhile, Lenn Sakata was a versatile infielder who often took over second base for Rich Dauer.

MORE WINS THAN EXPECTED AFTER 1977

Whatever one thinks of the merits of Weaver’s grand strategy, it is likely not a coincidence that his Baltimore Orioles exceeded their “Pythagorean” wins expectancy every season beginning in 1977. From 1977 to 1982, the Orioles won 35 games more than projected based on their runs scored versus runs allowed, an average of 5.8 extra wins per year. The only manager in history besides Weaver to exceed his team’s Pythagorean expectancy to such an extent for as many consecutive seasons was Joe Torre with the 2001–06 Yankees.[fn]Pete Palmer and Gary Gillette, ESPN?Baseball Encyclopedia (Sterling Publishers, 2008).[/fn] That the Orioles won only one division title (and one pennant) in that time and still won nearly six games a year more than they should have suggests Weaver’s philosophy was responsible for the Baltimore Orioles being as good and competitive as they were. Weaver’s approach to building his teams so that he had maximum flexibility for making moves during games, of seeking the platoon advantage at bat in key situations, of not sacrificing outs just to move base runners along, and of trying to score runs by batting them in—preferably in bunches with three-run home runs—made his Baltimore Orioles one of the toughest teams to beat.

BRYAN SODERHOLM-DIFATTE lives and works in the Washington, D.C., area and is devoted to the study of baseball history. His website, www.thebestbaseballteams.com, identifies the best teams of the twentieth century in each league using a structured methodological approach for analysis.