tag:blogger.com,1999:blog-38600807.post8312751554564343299..comments2023-11-05T04:16:44.937-05:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): Roundup 9/4/10Unknownnoreply@blogger.comBlogger9125tag:blogger.com,1999:blog-38600807.post-82291545817075939432010-09-08T17:14:03.296-04:002010-09-08T17:14:03.296-04:00"It's possible they've fallen for the..."It's possible they've fallen for the same trap I fell into years ago. The NFL calls penalty yards by an opponent "defensive penalties." But if true, it's interesting. The real question would be why don't defensive penalties matter?"<br /><br />I don't get the trap you fell into. Can you briefly elaborate?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-20913623882916655372010-09-07T20:29:20.808-04:002010-09-07T20:29:20.808-04:00I think sometimes you're a little harsh on F.O...I think sometimes you're a little harsh on F.O.<br /><br />I agree that the "Rule of 370" is really gimmicky and a bit of cherry picking numbers but I found their piece on defensive penalties really interesting. <br /><br />They did attempt to answer the "Why don't defensive penalties matter". The answer is that off penalties are typically a screw up of some kind, but defensive penalties are typically a result of playing just a little too hard. It does make a certain amount of sense.Jeff Clarkenoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-43209414585598409732010-09-06T16:00:55.956-04:002010-09-06T16:00:55.956-04:00Brian,
I think you missed a "we needed a stu...Brian,<br /><br />I think you missed a "we needed a study for this?" finding that they list: "Highly-drafted wide receivers without many college touchdowns are likely to bust."<br /><br />No, really? I wonder how large a data set this is. I can't really imagine an NFL GM drafting a WR in the 1-2 rounds who hasn't had a significant number of TD's in college.Unknownhttps://www.blogger.com/profile/07609159899149741576noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-22793511678713741852010-09-05T21:02:12.671-04:002010-09-05T21:02:12.671-04:00Hello Brian,
I've noticed that the running ga...Hello Brian,<br /><br />I've noticed that the running game (and by extension run defense) has received a lot of bad press lately and so I made a post on my blog that I hope might redress the balance just a little. I'd be very grateful if you'd check it out and give me your thoughts.<br /><br />http://keepingthechainsmoving.blogspot.com/<br /><br />Cheers,<br />Chris.Chrishttps://www.blogger.com/profile/18182426936194426623noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-980734109455976492010-09-05T14:39:44.329-04:002010-09-05T14:39:44.329-04:00Adam - I had a thought that you could use the norm...Adam - I had a thought that you could use the normal distribution c.d.f. to work out how many 'ranking points' a result gets. For instance, in the NFL the standard deviation of winning margins is 14.9 - so a 50 point win becomes normalised to a z-score of 3.36, scoring 99.96%. A 57 point win gets you 99.993% - so ever so slightly more. A 7 point win would get 68.1%, and scoring another TD gets you a 14 point win and a score of 82.6% for ranking. <br /><br />This would help provide a good ranking for college teams but whilst also removing the benefit for teams to run up the score. <br /><br />The problem with rankings like this is that it's always a compromise between a clearly understandable system giving ropey ranking and one that provides accurate ranks but is too complicated for anyone without a degree in maths to understand.Iannoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-2278253349402763302010-09-05T13:27:27.037-04:002010-09-05T13:27:27.037-04:00"Championship teams are generally defined by ..."Championship teams are generally defined by their ability to dominate inferior opponents, not their ability to win close games."<br /><br />Not only do I agree with this, but I laugh at the college football pundits who, several years ago, demanded that the BCS rankings remove any reference to margin-of-victory in their calculations. There was so much hand wringing about the supposedly evil incentive for coaches to run up the score, that they completely eliminated one of the most accurate indicators of a team's relative strength.<br /><br />IMHO, the real solution to the run-up-the-score incentive was simply to ensure that MOV considerations are done in tiers. In other words, if you beat a team by 50 points or by 57 points, you should really get the same bonus in whatever scoring system you are using, because there really is no huge distinction between a 57-point win and a 50-point win. But it's ludicrous to think that a team gets the same credit (in ranking systems) for a 50-point win that they would get for a 1-point win.Adam Davishttp://blog.insoftusa.comnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-88449366301743511302010-09-05T08:51:42.895-04:002010-09-05T08:51:42.895-04:00"A team will score more when playing a bad de..."A team will score more when playing a bad defense, and will give up more points when playing a good offense"<br /><br />That sounds more a corollary than a result of a study. If we define bad defenses as ones that give up more points (not contentious, right?), then the study is an exercise in the blindingly obvious.Iannoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-46480259200537743602010-09-04T20:29:58.055-04:002010-09-04T20:29:58.055-04:00I didn't spot this:
'Championship teams a...I didn't spot this:<br /><br />'Championship teams are generally defined by their ability to dominate inferior opponents, not their ability to win close games'<br /><br />To be honest that would seem an obvious statement. A game in which one team dominates another tends to be a game where the pre-game probability of one team winning was much greater than another, and as such the few elite teams which can be expected to have such wins more often than not will appear to be dominant in such games, yet when one elite team plays another it is much more of a coin toss as to the winner, so winning close games can be more down to luck, and so not a great indicator of a truly championship worthy team.v-zerohttps://www.blogger.com/profile/11677374172047801935noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-79987994675525616322010-09-04T20:18:45.099-04:002010-09-04T20:18:45.099-04:00Regression to the mean is such a simple, yet poorl...Regression to the mean is such a simple, yet poorly understood thing. <br />I have always liked to think about it as the result of limits, and nothing more, but that's the mathematician in me.<br /><br />The biggest error made with regression to the mean is to think that a high or low measurement is an indicator of a future sudden fall or rise in the measurement, which i simply the gambler's fallacy.v-zerohttps://www.blogger.com/profile/11677374172047801935noreply@blogger.com