tag:blogger.com,1999:blog-38600807.post3052992792181499809..comments2018-06-02T14:19:34.554-04:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): Game Probabilities - Week 11Unknownnoreply@blogger.comBlogger4125tag:blogger.com,1999:blog-38600807.post-45311036805737247342013-11-14T16:39:55.462-05:002013-11-14T16:39:55.462-05:00Dont forget there are 32 teams, and we only get ex...Dont forget there are 32 teams, and we only get excited about the best ones and worst ones. So yeah, its unlikely that KC should have gone 9-0, but each year we have 32 shots to see how extreme somebody's record could be. Andyhttps://www.blogger.com/profile/13033784825091918451noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-36430922260505461262013-11-14T15:30:39.590-05:002013-11-14T15:30:39.590-05:00and just for fun,
If we simply assumed N.F.L. ga...and just for fun, <br /><br />If we simply assumed N.F.L. games were 50/50 independent coin flips, KC chances of of being 9-0 is a somewhat insanely improbable 0.195 percent. But they did it. <br /><br />Their chances of actually going to 16-0 now would be 0.78%Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-28850611757224995062013-11-14T15:12:04.103-05:002013-11-14T15:12:04.103-05:00aren't all possible combinations of wins and l...aren't all possible combinations of wins and losses equally rare?<br /><br />for instance WLWLWLWLWLWLWLWL<br /><br />one could look at it as there is only one way to go 16-0, but many ways to go 8-8.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-72217801267351264932013-11-14T13:41:43.219-05:002013-11-14T13:41:43.219-05:00I would guess that the Chiefs' chances of goin...I would guess that the Chiefs' chances of going undefeated are higher than the less than 1 percent you listed for the simple fact that each game isn't independent. For example, if the Chiefs win in Denver (and assuming it doesn't happen because of 3 Broncos fumbles returned for touchdowns), their efficiency ratings will likely go up, meaning they would likely be a better team than we estimate now and their chances in their future games likely go up as well. I'm not really nitpicking given that I have no idea how I would incorporate Bayes theory into a model like this, but I thought I would just point this out.Anonymousnoreply@blogger.com