[ View Thread ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

NFL Forum

Pythagorean Formula for wins

Posted By: docriver
Date: 16 Aug 01, 12:02 pm

A long time ago I read a Bill James Baseball Statistics book. He's the guy that first started thinking of baseball stats in mathematical ways and published books on them - annually called the Baseball Abstract. In one of these annuals, he devised a way to estimate wins using Runs Scored and Runs Allowed. He called it the Pythagorean Forumla, as the estimated winning percentage was simply : RS^2/(RS^2+RA^2) where RS = Runs Scored and RA = Runs Allowed. This formula very closely matched the actual wins for the year, and better yet, it was a better predictor of how a team would do in the following year than the actual wins. It has grown hold in the baseball stats circles, and is evenly currently used in the ESPN MLB Expanded Standings.

So, I tried using this for football, and with the stats I got from ESPN, I had 1999 Wins/PF/PA and 2000 Wins. These are the two correlations I got :

1999 Actual Wins versus 2000 Wins : .36
1999 Pythag Expected Wins versus 2000 Wins : .40

This is not that big of a jump, but nevertheless, it is slightly higher. Which leads me to believe that when using stats like strength of schedule during the middle of the season, it may be better to use the Pythagorean estimated wins than actual wins.

But of course, one years correlations means little...if anyone has data on previous years win/loss records and Points for/against records, I'd love to have them to do these correlations to see if the Pythag formula is useful. My email is docriver2002@yahoo.com

Any comments, insights, problems, etc. would be highly appreciated!

Messages In This Thread

Pythagorean Formula for wins -- docriver -- 16 Aug 01, 12:02 pm
Pythagorean -- Editor -- 16 Aug 01, 1:33 pm
tails? -- docriver -- 16 Aug 01, 1:57 pm
Tails -- Editor -- 16 Aug 01, 3:50 pm
Starting point -- Colin Caster -- 16 Aug 01, 9:40 pm
1988 thru 2000 data -- docriver -- 17 Aug 01, 2:51 pm
also.. -- docriver -- 17 Aug 01, 3:01 pm
Nice work! Testing significance of correlation differences -- Colin Caster -- 19 Aug 01, 1:01 pm
email stats? -- docriver -- 19 Aug 01, 7:18 pm
Sure -- Colin Caster -- 19 Aug 01, 10:13 pm
Results (and computation) -- Colin Caster -- 20 Aug 01, 1:02 pm
Re: Bill James -- T. Hopper -- 19 Aug 01, 2:01 pm
I am told that the exponent for college football is 1.68 (nt) -- Colin Caster -- 19 Aug 01, 10:16 pm

[ View Thread ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

NFL Forum is maintained by Pi Yee Press