My Gift to You
Since it’s the holidays, I thought I’d give you all a bit of a present. So here is a spreadsheet containing complete linear weights data for the 2009 season.
A breakdown of the spread:
“LWTS_RC” are linear weights on an absolute scale; these figures differ from FanGraphs’ “wRC” in that they include intentional walks, sacrifice hits, reaching base on error, catcher’s interference, strikeouts, ground into double plays, and pickoffs. These values are derived from Retrosheet’s play-by-play data (via Terps) rather than the “quick-n’-dirty” method that is implemented at FanGraphs. I’d like to think that means these values are more accurate, and they should be. The figures with asterisks implement BP’s park factors. To be perfectly honest, I’m not sure how they develop their PF. Since the PF varies from player to player on the same team, I imagine it’s based on a per-plate appearance basis or they’re using an “individual” park factor. I’ve included the non-adjusted page so you can apply the PF of your choice.
“R/G” stands for Runs per Game. Based on the rate of runs and outs the player creates, we can estimate the amount of runs a lineup of that player’s production would score per game. It’s simply (LWTS_RC/Outs *26.25).
“RAA” and “RAR” are Runs Above Average and Runs Above Replacement, respectively. Personally, I prefer to stick with a baseline of average since it’s universally recognized and easy to work with (+0 is exactly average), but RAR seems to be the more current popular hitting valuation method out there. Replacement level in the spreadsheet is defined as +20 runs per 600 plate appearances for the National League and +25 runs per 600 PA for the American League.
“VORP” is the player’s Value Over a Replacement Player (I’m just using the same name as BP’s statistic). This is simply RAR plus the positional adjustment.
“rAVG” and “rOBA” stand for “relative average” and “relative on-base average.” I cleaned up the methodology for “wAVG” and thought re-naming it might be better, since it’s not set up the same way wOBA is. The same goes for rOBA. Both rates are set so that the league average rAVG is always .260 and rOBA is .330. These are just meant to provide some rate statistics for people to play around with. I’m not claiming these to be new statistics or anything like that.
EDIT (1/1/10): I was using the wrong set of data- the spreadsheet now includes the proper league adjustments in both the linear weights used and the replacement levels.
Here are the leaders for 2009:
1. Albert Pujols, +84.7
2. Joe Mauer, +83.0
3. Hanley Ramirez, +73.9
4. Prince Fielder, +70.6
5. Derek Jeter, +68.5
6. Chase Utley, +66.8
7. Ryan Braun, +65.6
8. Pablo Sandoval, +62.5
9. Ben Zobrist, +58.4
10. Adrian Gonzalez, +57.6