tag:blogger.com,1999:blog-38600807.post6906691239996251525..comments2011-11-04T13:11:10.419-04:00Brian Burkenoreply@blogger.comBlogger7125tag:blogger.com,1999:blog-38600807.post-51809625778853404082011-11-04T13:11:10.419-04:002011-11-04T13:11:10.419-04:00&quot;Each variable is listed here with its correlation with season wins.&quot; There is a broken link to your list.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-30280230312238128112011-10-05T17:49:20.360-04:002011-10-05T17:49:20.360-04:00Brian can you put on the link again to this page its not working thanksMattnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-61726335214231853902011-08-05T04:42:51.222-04:002011-08-05T04:42:51.222-04:00Need this link to get working!Johnnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-26557630256507198772010-12-17T11:26:34.205-05:002010-12-17T11:26:34.205-05:00link doesn&#39;t work!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-48587977271277394662009-11-19T12:19:12.912-05:002009-11-19T12:19:12.912-05:00It would be interesting to run a stepwise regression and find 3-4 variables which jointly predict wins effectively. Neat correlations; I knew turnovers were important, but wow.Roberthttp://www.blogger.com/profile/04622889259843346014noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-19693611093921475532009-11-07T03:17:30.144-05:002009-11-07T03:17:30.144-05:00Brian,<br /><br />Have you done much research on what causes teams to deviate from expectation? For example, say you use the strongest correlations to formulate your weekly probabilities. Now suppose that over a period of time (season?) a given team loses three games they were expected to win and wins two games they were expected to lose.<br /><br />Let&#39;s say the three games they lost were games where they faced short stocky running backs while the two games they won were against tall lanky running backs. I would be interested in studying what causes deviations from the simple model. These delta functions could then be used as refinements.<br /><br />Things such as height to weight ratio of primary running back, total weight of offensive and defensive lines, height of #1 and #2 receivers, etc. Obviously, it couldn&#39;t be too complex and it would be hard to find statistically distinct relationships based on sample size (unless they could be extended across all teams).<br /><br />Just a thought. Curious if you&#39;ve studied it. You&#39;ve obviously done a lot more work on this than I have. Great site.<br /><br />HappyHappyhttp://www.bestofblog.net/nfl_picks_2009/2009_YTD_nfl_analysis.htmlnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-3551791601753439912009-09-22T12:49:53.658-04:002009-09-22T12:49:53.658-04:00Interesting, and very useful.<br /><br />However, I would add one VERY CRUCIAL stat. NET POINTS.David B.http://www.thumper300zx.comnoreply@blogger.com