Arthur: What free agent signings can teach us about player evaluation

From SABR member Rob Arthur at Baseball Prospectus on January 3, 2020:

Sabermetrics currently straddles a huge and growing chasm between the public consensus and what front offices know. While 20 years ago the staff at Baseball Prospectus was arguably further ahead than most teams, now the most backward front office employs at least a handful of dedicated analysts, some culled from the very ranks of the best public investigators, others with doctorates in quantitative fields. The advent of Statcast broadened the public/private gap even further, because while Baseball Savant surfaces a handful of data points per play (exit velocity, launch angle, etc.), team researchers get the unfiltered path of every person and object on the field. Some are able to measure fielding, baserunning, and hitter or pitcher mechanics in ways that would make public analysts drool.

So in an era when teams are lapping the public’s understanding of baseball, what are we to do? If we can’t get the same data and do the same exploration as them, perhaps we can work backward: Reverse-engineer what teams have found using the decisions they’ve made. We’ll never get to the level of detail front office personnel have at their fingertips, but perhaps we can extract the broad outlines of what player characteristics teams pay for and what they ignore. (I originally presented a version of this research at this year’s Saberseminar.)

Read the full article here:

This page was last updated January 3, 2020 at 11:48 am MST.