tag:blogger.com,1999:blog-38600807.post2397354697396665063..comments2018-06-02T14:19:34.554-04:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): Playoff Probabilities: Week 15Unknownnoreply@blogger.comBlogger11125tag:blogger.com,1999:blog-38600807.post-71993316869527044582011-12-17T12:49:50.788-05:002011-12-17T12:49:50.788-05:00James--Getting the playoff success probabilities f...James--Getting the playoff success probabilities for an adjusted Houston would actually be a little tricky... In terms of seedings, putting their adjusted numbers in gives them a 50% probability of entering the playoffs as the #3 seed, and gives the biggest seeding boost to Pittsburgh, upping their probability of securing the #1 seed to 30%. (PIT has lost to the Texans, so they have the most to gain.)<br /><br />But I wouldn't underestimate the Texans, even without Schaub under center. Their defense and running game are still very solid, and, as demonstrated last Sunday, Yates JUST WINS.Josh Katzhttps://www.blogger.com/profile/06646400031653670129noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-4042919226027376962011-12-16T14:06:24.815-05:002011-12-16T14:06:24.815-05:00To nit-pick some more, can we see the playoff succ...To nit-pick some more, can we see the playoff success odds with an adjusted Houston? They are really messing up all the numbers.Jameshttps://www.blogger.com/profile/01838293735141324662noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-68949870153103470842011-12-16T02:26:47.263-05:002011-12-16T02:26:47.263-05:00No worries, tables now updated with the correct va...No worries, tables now updated with the correct values. And thanks to anonymous commenter #1 for catching that--though I'm sure the people of New Orleans will be quite displeased to see their team's probability of a Super Bowl win drop by more than half.Josh Katzhttps://www.blogger.com/profile/06646400031653670129noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-70672355228674246542011-12-15T23:59:17.535-05:002011-12-15T23:59:17.535-05:00I had the wrong team efficiency rating entered for...I had the wrong team efficiency rating entered for New Orleans. Both my on-line software and my blog are now updated with the correct rating for the Saints. Sorry for the confusion, especially to Josh. I really appreciate the time he takes to do these write ups.Chrishttp://nfl-forecast.comnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-8321874592783448772011-12-15T22:16:25.304-05:002011-12-15T22:16:25.304-05:00Chris can speak to the random number gen better th...Chris can speak to the random number gen better than I could, but the numbers in the table are actually drawn up using significantly more than 5000 sims--I'm fairly certain it's not a sample size issue.Josh Katzhttps://www.blogger.com/profile/06646400031653670129noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-71563223342840196962011-12-15T22:10:34.074-05:002011-12-15T22:10:34.074-05:00Hehehe This is the same Anon as before. I just to...Hehehe This is the same Anon as before. I just took my final for Statistics today, and for a sample that large, means that far off are 2*(5000)^.5 stds away. Meaning, that is certainly not the answer.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-62994538595348139732011-12-15T21:48:53.866-05:002011-12-15T21:48:53.866-05:00perhaps the simulations use the same random number...perhaps the simulations use the same random number to simulate all the games in each week so there is a correlation between weaker teams winning the NFC and AFC? or there is serial correlation in the random number generator giving a similar effect?<br /><br />also 5000 isn't that big of a sample size. For a team like GB that made 2000 SBs in the simulation the stdev on the winning percentage is ~ .01 so if two teams have a SB winning percentage of .55 and .57 there is a reasonable chance the simulations are going to have the .55 team with a higher SB win rate.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-83898637408503925812011-12-15T19:12:42.929-05:002011-12-15T19:12:42.929-05:00That is odd... I'll look into it. Perhaps ther...That is odd... I'll look into it. Perhaps there is a glitch in the matrix.Josh Katzhttps://www.blogger.com/profile/06646400031653670129noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-14677686336108082972011-12-15T18:16:47.226-05:002011-12-15T18:16:47.226-05:00Shouldn't that show up in the 'Make Super ...Shouldn't that show up in the 'Make Super Bowl' column though?<br />As far as conditional probability goes:<br />GB wins 22/43 SB's<br />NO wins 22/39 SB's.<br />It would seem to be a necessary implication that either NO is stronger by your model (its not) or that that somehow NFC and AFC matchups arent independent variables. While not completely independent(perhaps there are some factors that make NO's potential matchups better in the case it makes the SB), this does not explain the discrepancy.<br /><br />Unless there's something very basic I'm missing here, the divisional round being an easier matchup shouldn't matter in this case?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-20876234673084699972011-12-15T18:12:03.991-05:002011-12-15T18:12:03.991-05:00Don't know for sure, but I'd speculate the...Don't know for sure, but I'd speculate they might get an easier matchup in the div round.Brian Burkehttps://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-42733909965602392252011-12-15T17:48:47.859-05:002011-12-15T17:48:47.859-05:00Genuinely curious how it is possible that the Sain...Genuinely curious how it is possible that the Saints have a better chance of winning the SB should they get there than the Packers winning should they get there.Anonymousnoreply@blogger.com