Player Game Percentages, World Series (October 8, 2003)
This is similar to Mills' Brothers Player Win Averages and Doug Drinen's Win Probability Added.
--posted by TangoTiger at 05:43 PM EDT
Posted 3:27 p.m.,
October 9, 2003
(#1) -
Rally Monkey
I'm shocked that Spiezio's G6 homer didn't make the top 10 plays of the series. Interesting link, though.
Posted 3:46 p.m.,
October 9, 2003
(#2) -
reno dakota
Does anyone keep this stat on a regular-season basis? (I imagine it would take a lot of time, so my suspicion is no.) I can't think of a more useful stat for settling MVP/Cy Young debates than this offhand.
Posted 3:47 p.m.,
October 9, 2003
(#3) -
reno dakota
Taken in concert with defensive stats and a small fudge factor, of course.
Posted 5:03 p.m.,
October 9, 2003
(#4) -
Pete
Is the win probability table adjusted for different runs per game environments? A run in the 3rd inning of a 0-0 game should be worth far more in Dodger Stadium in the 60s than in Coors Field in the 90s.
Also, is it available online?
Posted 5:11 p.m.,
October 9, 2003
(#5) -
tangotiger
Reno, I'm thinking about doing this, though I'm not sure if I've got the time.
Pete, you are right that you need different win probability matrix for different run environments. You also have a different run environment, from the hitter's perspective, if you have Pedro at Dodger Stadium, than Joe Schlub at Coors. It gets to be a very very messy thing to try to do right.
I'll post a link to Doug Drinen's article, as it was top-notch as well.
Posted 5:14 p.m.,
October 9, 2003
(#6) -
tangotiger
(homepage)
This is Doug Drinen's Win Probability Added (WPA).
Posted 6:31 p.m.,
October 9, 2003
(#7) -
studes
(homepage)
I just finished reading Curveball, and I enjoyed it immensely (which just proves that I'm a total geek, I guess). Many of the topics that are covered here are described in detail in the book. I'd highly recommend it if you're interested in this approach.
Posted 8:04 p.m.,
October 9, 2003
(#8) -
FJM
If you have the Win Probability Matrix for a given situation (say, Coors Field, 2000-2002)in an Excel spreadsheet and you are able to read in the PBP data for Coors in 2003, it shouldn't be too difficult to calculate the WPA for each player, game by game. But what if the 2003 version of Coors is significantly different than 2000-2002, as it seems to have been? Then the old WPM may no longer apply. But if you try to create a new matrix based solely on 2003 data, you'll get all sorts of odd results. Even 3 years data won't be enough to prevent that entirely. The point is, doing the calculation isn't hard; creating a reliable situation-specific matrix is.
Posted 8:22 p.m.,
October 9, 2003
(#9) -
tangotiger
Having 1 million games won't give you the reliable data you need at the inning/score/base/out level. You need to create a math model based on various event/base-out-state/transition rates.
For those who don't know, the author of the article cited is the author of Curve Ball.
Posted 9:39 p.m.,
October 11, 2003
(#10) -
reno dakota(e-mail)
Tango-- You're totally right that this thing could get really messy. I think it would be really useful, though. Let me know if you need any help running numbers.