tag:blogger.com,1999:blog-38600807.post5922409306483208212..comments2015-03-23T23:15:10.100-04:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): Win ProbabilityBrian Burkenoreply@blogger.comBlogger27125tag:blogger.com,1999:blog-38600807.post-67624216549350149692014-05-06T06:36:24.135-04:002014-05-06T06:36:24.135-04:00The fact that the +7 and -7 aren't reflections...The fact that the +7 and -7 aren't reflections of each other suggests something in your calculation is incorrect.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-78718449295718527922010-11-12T03:32:28.230-05:002010-11-12T03:32:28.230-05:00Interesting stuff. But shouldn't your WP of go...Interesting stuff. But shouldn't your WP of going for it also factor in the probability of missing the field goal? Shouldn't the equation be something such as .... <br /><br />WP(go for it) = 0.30 * WP(+4 point lead) + 0.70 * WP(-3 point deficit) + (probability of missing 19-yard field goal) * WP(-3 point deficit) ????Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-21398532198859373602009-11-18T02:53:49.081-05:002009-11-18T02:53:49.081-05:00Oops, that should read "love the model you ha...Oops, that should read "love the model you have created" ...Jan Suchaneknoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-30763580657934760612009-11-18T02:52:11.059-05:002009-11-18T02:52:11.059-05:00I smell some sample issues with this study. On top...I smell some sample issues with this study. On top of the points Western Spartans made (which I don't think were really resolved) there are two other spots on the charts that can't be right.<br /><br />One, at the end of the game where the orange line climbs above the purple, it can't be possible that a 3-point deficit is superior to a 1-point deficit. Also, I don't think it should ever be the case that the green line drops below 50%. Having the ball must always be superior to not having the ball.<br /><br />Sorry if that comes off as too critical! Love the you are created and the logic behind it. It just seems like you need a whole bunch more data to make it accurate.Jan Suchaneknoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-14823526765438152152009-11-16T23:50:13.847-05:002009-11-16T23:50:13.847-05:00I think the formula is wrong. I think that based ...I think the formula is wrong. I think that based on your chart, the probability of winning with a 4 pt lead is about 80%, not 92%. That's because when you score the TD on 4th and 2, you're giving the ball to the opposition. If you look at the chart, -7 and the ball is roughly 15% and -3 and the ball is roughly 22%. So I'm guesstimating that -4 and the ball is 20%.<br /><br />Similarly, if you fail on 4th and 2, the win probability is roughly 10%, not 22% because again, you're giving up the ball. IOW, the opponent has the ball and a 3 pt lead so the chart says he has 90% chance of winning and thus you have 10% chance.<br /><br />If you change the equation to reflect these new values, you get:<br /><br />.3*.8 + .7*.1 = .31. Just about a wash compared to kicking the FG.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-54514899012201226062009-11-16T23:05:44.627-05:002009-11-16T23:05:44.627-05:00good question. That was for 2007 only, which had a...good question. That was for 2007 only, which had an unusually low conversion rate. The numbers in the Belichick analysis are based on a much, much larger data set. See part 3 of 'The 4th Down Study' article linked to at the top right of this page. It has the full numbers.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-75251645785417682792009-11-16T23:01:01.943-05:002009-11-16T23:01:01.943-05:00Why was it stated in the Belichek thread that ther...Why was it stated in the Belichek thread that there was a 60% chance of making 4th and 2 when this very article states that the probability is roughly 30%?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-78692700930236731712009-11-09T17:47:41.646-05:002009-11-09T17:47:41.646-05:00fascinating stuff. I'm confused though, in thi...fascinating stuff. I'm confused though, in this post you said that your WP does not factor in field position, down, distance. But your WP calculator does have those factors. So does that mean you have an updated WP chart?<br /><br />http://wp.advancednflstats.com/winprobcalc1.phpAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-44836021878201648992009-10-10T00:55:36.229-04:002009-10-10T00:55:36.229-04:00we don't like jimmy johnson on lombardi ave :(...we don't like jimmy johnson on lombardi ave :(coach tomhttp://coachtom.comnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-10117820360801516702009-10-10T00:47:35.725-04:002009-10-10T00:47:35.725-04:00they usually were overdogs :)they usually were overdogs :)coach tomhttp://coachtom.comnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-54777733752903434882009-10-10T00:41:13.324-04:002009-10-10T00:41:13.324-04:00do you believe, the outstanding coaches of recent ...do you believe, the outstanding coaches of recent times [walsh, beliczek] were less timid than the run-of-the-mill coaches in play-calling?coach tomhttp://coachtom.comnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-57238381269780855242009-08-05T12:49:22.244-04:002009-08-05T12:49:22.244-04:00"For trivia purposes, (and to show off how ea..."For trivia purposes, (and to show off how easy it is for me!) the number of cases since 2000 that featured a 1st down when down by 7 on their own 47 with 13 min left in the 4th is: two."<br /><br />Well played, and thanks for the response!81Trucolorshttp://www.blogger.com/profile/02280571817548405090noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-32493413854329830062009-08-04T00:30:56.022-04:002009-08-04T00:30:56.022-04:00Thanks. It varies greatly. I use chunks of data, s...Thanks. It varies greatly. I use chunks of data, so depending on the situation, I'll average a block of 20 yards of field position and up to 5 minutes of time. Then I'll interpolate between "chunks" for the particular win %. There's a lot of sophisticated modeling and smoothing going on underneath the raw win%. So even if the "chunk" size is very small (<100 observations), it's supported by data in adjacent chunks.<br /><br />For trivia purposes, (and to show off how easy it is for me!) the number of cases since 2000 that featured a 1st down when down by 7 on their own 47 with 13 min left in the 4th is: <i>two</i>.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-74192150180960302762009-08-04T00:15:53.175-04:002009-08-04T00:15:53.175-04:00"The win probabilities are derived empiricall..."The win probabilities are derived empirically. I simply look at the database and compare all games with the same time remaining/score difference/field position/etc. Whatever percent of the time a team in the same (or very similar) situation wins becomes the WP. "<br /><br />Brian you have an amazing site. Really not sure why you don't have a job with an NFL team. My question is, how large a sample size are we talking about for each situation? For example, how often is a team down by 7 on their own 46 yard line with 13:00 left in the 4th quarter?81Trucolorshttp://www.blogger.com/profile/02280571817548405090noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-11424843957009779762009-03-17T14:02:00.000-04:002009-03-17T14:02:00.000-04:00Yes. But that is a general average that does not a...Yes. But that is a general average that does not account for the fact one team is due to receive the kick off.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-67102993813032691832009-03-17T13:38:00.000-04:002009-03-17T13:38:00.000-04:00Is it safe to assume that at 30min remaining the s...Is it safe to assume that at 30min remaining the score differential reflects that half time score and is unaffected by field position and down?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-90957734438280932112008-12-04T00:20:00.000-05:002008-12-04T00:20:00.000-05:00Thanks. The win probabilities are derived empirica...Thanks. <BR/><BR/>The win probabilities are derived empirically. I simply look at the database and compare all games with the same time remaining/score difference/field position/etc. Whatever percent of the time a team in the same (or very similar) situation wins becomes the WP. <BR/><BR/>There's some data smoothing and interpolation too, but mostly that's it.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-9136185955152259972008-12-04T00:15:00.000-05:002008-12-04T00:15:00.000-05:00Great site. How do you generate the win probabili...Great site. How do you generate the win probabilities?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-28169506742429750552008-09-10T02:04:00.000-04:002008-09-10T02:04:00.000-04:00Hey, I just stumbled upon this here. I was the one...Hey, <BR/><BR/>I just stumbled upon this here. I was the one who created Protrade's Win Probability model. Even though our business model has steered us away from analytics a bit for the time being, I still see it as my baby :-) It took a LONG time to build.<BR/><BR/>Some comments:<BR/>* we used about 7 years of Play by Play data (about 2M plays)<BR/>* we did not group into 5 minute intervals, and did account for end of game situations<BR/>* we attempt to account for discountinuous effects (scoring comes in 3/7 pt chunks and down by 4 vs 5 late is very similar)<BR/>* our model takes many factors into account: score differential, down, distance, field position, time, timeouts, field types,...)<BR/>* the output of our WP model does agree with Roemer that coaches are too conservative on 4th down.<BR/><BR/>Keep up the good work Brian! All these articles are very interesting.<BR/><BR/>-Mark mkamal@protrade.comMark Kamalnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-2060633956190536842008-08-29T17:47:00.000-04:002008-08-29T17:47:00.000-04:00WS-Very perceptive. Regular reader JonnyMo pointed...WS-Very perceptive. Regular reader JonnyMo pointed that out to me. See the explanation in my article "The End Game." Bottom line is that teams with very small leads play too conservatively and teams with small deficits play more aggressively, which may be closer to the generally optimum level of risk/reward balance.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-16394776357594260572008-08-29T15:24:00.000-04:002008-08-29T15:24:00.000-04:00I understand how having the ball could give you a ...I understand how having the ball could give you a >50% chance of winning with -1 or 0 pt deficit, but how can being down 1 point (vs. 0 or +1) have a better WP? <BR/><BR/>In other words how can the purple line ever be above the green (let alone red). Ditto green over red for that mystery spot at the beginning of the 4th quarter.Western Spartanshttp://www.blogger.com/profile/01427596644438760253noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-26536295206282902012008-08-12T20:55:00.000-04:002008-08-12T20:55:00.000-04:00Argh. Didn't read through all the comments. The ...Argh. Didn't read through all the comments. The "has the ball" was the part I missed.milesnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-67945465021452319792008-08-12T20:53:00.000-04:002008-08-12T20:53:00.000-04:00Can you explain the +0 line? In that situation, b...Can you explain the +0 line? In that situation, both teams should have it, so, I don't grok how the line could deviate from 50%?milesnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-69459457275732759322008-08-07T11:54:00.000-04:002008-08-07T11:54:00.000-04:00Anon-No, the probabilities for +1 and -1 would not...Anon-No, the probabilities for +1 and -1 would not add up to 100% in this case. But good question and this is something I should clarify. <BR/><BR/>The probability curves in the graph consider possession. So the probability curve labeled +1 is for a team up by 1 point <I>and has the ball</I>. Conversely, the curve labeled -1 for for a team down by a point <I>and has the ball</I>.<BR/><BR/>The extra 22% you point out could be considered the value of simply possessing the ball, in terms of probability of winning at that point in the game. <BR/><BR/>Thanks for the question.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-10813476252241521122008-08-07T11:44:00.000-04:002008-08-07T11:44:00.000-04:00The way I see it, you should be able to take the +...The way I see it, you should be able to take the +'s and -'s and apply one to you and the other to your opponent. So there are 10 minutes left in the 4th quarter and you are u by one point. <BR/><BR/>Shouldn't your chance to win when up by 1 point + your opponents chance to win when down by 1 point for any given minute of the game add up to 100%.<BR/><BR/>I'd expect that the + curves the - curves would exact reverses of each other.<BR/><BR/>With 10 minutes left in the 4th quarter the team that's up by 1 has a 57% chance of winning and the team that's down by 1 has a 65% chance of winning. That adds up to 122%. Seems wrong to me but everything I know about statistics I learned from reading NFL stat sites so I could be wrong :)Anonymousnoreply@blogger.com