Friday, May 24, 2013
Reliever pitcher correlations
Bill gives us the chart.
Please, please, please, if you report the correlation, you MUST report the number of opportunities. You can achieve r approaching 1 if you have an infinite n (and no systematic bias), and you can achieve r approaching 0 if you have n close to 0. Even though the underlying participants are identical!
It's why we get bullsh!t claims about payroll and wins not correlating. If you look at the number of games over say just one month, the correlation will be pretty low. And if you look at it over 162 games, the correlation will go up. And if you look at it over 324 games, the correlation will go higher still. The more games, the higher the correlation. (At some point though, it tops off, simply because the team in year 1 is not the same as the team in year 6. And naturally we have the systematic bias of the Rays and A's not going after free agents, while the Yanks do. The same applies to players, so that a player in 2005 is not really the same as the player in 2012.)
But, anyway, that's the general idea. Report the number of opportunities.
Recent comments
Older comments
Page 1 of 150 pages 1 2 3 > Last ›Complete Archive – By Category
Complete Archive – By Date
FORUM TOPICS
Jul 12 15:22 MarcelsApr 16 14:31 Pitch Count Estimators
Mar 12 16:30 Appendix to THE BOOK - THE GORY DETAILS
Jan 29 09:41 NFL Overtime Idea
Jan 22 14:48 Weighting Years for NFL Player Projections
Jan 21 09:18 positional runs in pythagenpat
Oct 20 15:57 DRS: FG vs. BB-Ref
Apr 12 09:43 What if baseball was like survivor? You are eliminated ...
Nov 24 09:57 Win Attribution to offense, pitching, and fielding at the game level (prototype method)
Jul 13 10:20 How to watch great past games without spoilers