tag:blogger.com,1999:blog-38600807.post5856614568882782842..comments2023-11-05T04:16:44.937-05:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): Is "Red Zone" Performance Real?Unknownnoreply@blogger.comBlogger15125tag:blogger.com,1999:blog-38600807.post-7517841034953201292011-12-20T21:06:58.498-05:002011-12-20T21:06:58.498-05:00I realize this is an old article, but I was lookin...I realize this is an old article, but I was looking for year over year red-zone performance for quarterbacks, as a way to prove this same conjecture, and I stumbled upon this article (and you're imo superior approach).<br /><br />I have been very interested recently in quantifying luck football, and this piece helps a lot to that end.Bustohttps://www.blogger.com/profile/10103696785400203704noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-27655044321240387802011-08-18T16:00:52.460-04:002011-08-18T16:00:52.460-04:00Hey Brian, I just stumbled on this site and have b...Hey Brian, I just stumbled on this site and have been rather impressed with your content. I'm slowly working my way from the beginning to current day now.... <br /><br />You mentioned the difficulty in using yards per attempt as a predictor. Did you consider using a metric that measures performance as a % of the distance to goal?<br /><br />For instance, a 5 yard pass on the 20 would be equal to a 1 yard pass on the 4 (25% of the distance to goal). Then you could take the average per attempt.<br /><br />By the taking the avg per attempt, you'd also get an idea of how long it took to get into the endzone. A perfect score would be 1 (1 TD pass per attempt).<br /><br />Obviously this analysis was from several years ago, and it'd be a bear to get the required stats. Just wanted your thoughts.<br /><br />Keep up the great work. I look forward to reading the rest!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-40457002082810749892010-10-28T20:45:17.574-04:002010-10-28T20:45:17.574-04:00I think "Chances are Buffalo fans will be dis...I think "Chances are Buffalo fans will be disappointed next year" should win the IgNobel Prize for Statistics!<br /><br />(Note: I love reading this site, so that's not meant as insult. But with the Bills, if we don't laugh...)Eric Moorehttps://www.blogger.com/profile/01496717842529415018noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-38817580566155548542009-12-10T00:37:26.921-05:002009-12-10T00:37:26.921-05:00Why not use a Bonferroni (sp?) correction? To acco...Why not use a Bonferroni (sp?) correction? To account for the number of tests your running, simply divide your alpha level by the number of tests you run and use that as your acceptable alpha level. It'll give you an extraordinarily difficult level, but that unfortunately comes with the territory when you run 30 independent t-tests.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-35781867229290184462009-10-05T08:27:43.097-04:002009-10-05T08:27:43.097-04:00That's my entire point.That's my entire point.Brian Burkehttps://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-35957810609159394632009-10-05T07:44:25.154-04:002009-10-05T07:44:25.154-04:00I think you should also keep in mind that you have...I think you should also keep in mind that you haven't adjusted for the number of tests you've performed. So even with those p values you did find with a p = 0.04, there's a 4% chance it's occured due to chance, and since you've performed so many t-tests there's actually a solid chance that one of your "significant" results is itself a result of random variation.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-78537787124722244632009-09-23T00:43:05.381-04:002009-09-23T00:43:05.381-04:00I think the best argument against red zone perform...I think the best argument against red zone performance as part of a players skill set is Jerry Rice.<br /><br />One year he sets the record for most TD catches (21 or something?), the next—in more games—he catches only 9. <br /><br />Any argument crediting Rice with skills resulting in TDs must also ask if he just forget how to catch them next year.<br /><br />It's like the oft-used "That guy is a winner." I guess, but what is he if they lose?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-27435261920525196892008-03-05T10:34:00.000-05:002008-03-05T10:34:00.000-05:00Joey, the problem with your analysis is that you o...Joey, the problem with your analysis is that you only took the 2007 stats Brian listed here. If you go back and look at McNabb's red zone stats for a few more years, you'll see that historically he's been quite good down there.<BR/><BR/>Notwithstanding the occasional ball thrown well short of the end zone with five seconds left and no timeouts...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-91494192075469955522008-02-09T10:00:00.000-05:002008-02-09T10:00:00.000-05:00My background hardly compares with yours so I am n...My background hardly compares with yours so I am not going to dispute the results of your statistical reasoning. BUT as they say, figures lie and liars figure. There is something amiss here.<BR/><BR/>Looking at team like the Eagles and Donovan McNabb, you can see they are highly productive in the large area of the rield and less so in the more confined and close Red Zone. McNabb can't get around so easily in that space and has to rely totally on his passing and reading skills which often come up short. That's why they can be6thin the league in offensive plays but 12th in points. Too many field goals.<BR/><BR/>Forget the numbers. What is needed inside the restricted area known as the Red Zone is quickness. You need to read more quickly, react more quickly and move more quickly to get free because the defenders are on you immediatley. And there is increased pressure.<BR/><BR/>Quarterbacks that can handle those circumstances do well. Those that cannot, do less well. A quarterback with great numbers outside the Zone may have okay numbers inside the zone and still not be very good because he starts out with so much. Marginal quarterbaks perhaps have less to work with and any loss of ability under the pressure of the Red Zone might just be too much to easily assimilate.<BR/><BR/>To me, numbers aside, some guys perform well in that tight space and others do not. Great QB's perform well everywhere on the field and the Red Zone is no exception. They just dial up what they need and move on. <BR/><BR/>But others don't have enough in the tank to be able to dial up more when more is needed. So they fail.<BR/><BR/>Football in the Red Zone - in my opinion - is much different football than football outside the Red Zone. <BR/><BR/>My name is Joey and I approve of this message - right or wrong lol<BR/><BR/>Go Birds.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-44231209262759686702008-01-28T10:52:00.000-05:002008-01-28T10:52:00.000-05:00PV-I think you nailed it. Height and jumping abili...PV-I think you nailed it. Height and jumping ability are probably the key attributes for red zone/goal line passing success. Speed may be the more important attribute outside the red zone. A tall QB <I>and</I> tall receivers might be the key. <BR/><BR/>I wonder if we'll come to a point where all teams will use highly specialized red zone personnel. They would just be gargantuan guys without much speed--basketball centers. If I were a GM I'd invite a few of NCAA centers to camp to see what they could in the back of the end zone against 6-ft tall safeties and 6-ft-3 linebackers. Even if they don't pull down the ball, they might draw a lot of interference calls.Brian Burkehttps://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-33215206194287224312008-01-28T09:59:00.000-05:002008-01-28T09:59:00.000-05:00I still wonder if at some positions, certain playe...I still wonder if at some positions, certain players are more suited to perform well in the red zone than others. No, not because of a "clutch" ability in their psychology. I would ask, "What is fundamentally different about football in the red zone and football outside of the red zone?" The practical answers include less space, no threat of a deep pass, and space limitations at the back of the end zone. Defenders, then, have less space to fill, and don't have to play safe for a big play, and thus everything is more difficult on all offensive players. The back of the end zone, however, allows some plays that offenses wouldn't run elsewhere on the field because there is no safety over the top.<BR/><BR/>So, are there certain WRs, TEs, and RBs that are better suited to perform well in a situation where more defenders are clogging space, but new plays are also opened up due to the back of the end zone? I'm not sure. If there are such players, it's not because of cluth ability, I think, but because they have physical attibutes suited to the different situation (size, strength, leaping ability, etc.). Of course those same skills have usefulness all over the field, so such players might not stand out as "extra" good in the red zone. Perhaps the red zone, then, brings certain limitations to certain players who are effective elsewhere on the field, more than it brings advantages to certain players who are suited to the red zone.Pacifist Vikinghttps://www.blogger.com/profile/16630996018868040440noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-51795221428400907102008-01-26T18:00:00.000-05:002008-01-26T18:00:00.000-05:00"my analysis here does not prove the non-existence..."my analysis here does not prove the non-existence of clutch red-zone performance. It just shows [if it exists] the effect is so small that it doesn't really matter."<BR/><BR/>You're right, which means it doesn't exist for any practical purposes, and you really do show that.<BR/><BR/>The "asymetrical downside" is just a theoretical nit, even in baseball. Any NFL QB who's really that bad won't be around long enough for anyone to collect a meaningful data set on him, so to speak. Sorry I mentioned it. <BR/><BR/>And yes, since James landed a spot on the Red Sox payroll and has been so impressed by Big Papi he's kind of backtracked from "there's no evidence of clutch hitting's existence" to "there's not enough evidence to fully disprove clutch hitting's existence", which may be technically true, but really, for any practical purpose...JGhttps://www.blogger.com/profile/11164150812219689611noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-26843163784306941892008-01-26T13:11:00.000-05:002008-01-26T13:11:00.000-05:00Good point that the effect may not be symmetrical....Good point that the effect may not be symmetrical. Hadn't considered that. I'd need more years of data to examine that. <BR/><BR/>Unfortunately, I'm limited to the data that's available out there. I'd love yds per attempt in the red zone, or using play action, etc. I used data from myway.com (http://sports.myway.com/nfl/stats.html) which is one of the more comprehensive sources for stats. It's where I get by YAC data for my QB rating.<BR/><BR/>And speaking of James, it's important that I point out that my analysis here does not prove the non-existence of clutch red-zone performance. It just shows that we almost certainly could never detect it by what we observe. Just like clutch hitting in baseball, it either does not exist, or the effect is so small that it doesn't really matter.Brian Burkehttps://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-66979573427842965332008-01-26T00:36:00.000-05:002008-01-26T00:36:00.000-05:00"The opposite could be said of QBs such as Kellen ..."The opposite could be said of QBs such as Kellen Clemens ... chances are he will not under-perform inside the 20 as severely as he did last season"<BR/><BR/>Very probably, yes, but this may not be quite as certain as that players who over-performed by as much won't do it again.<BR/><BR/>In this regard, FWIW, I'll just mention Bill James's observation that in baseball's stat categories where overperformance is chance or luck, underpeformance does sometimes to some degree reflect (lack of) skill. The situation need not be symmetrical on the up and down sides.<BR/><BR/>One example he used for which there is a ton of data is one-run baseball games. He found that teams that won a lot more than their share of one-run games during a given period of time had zero ability to persist in doing so during a subsequent period, indicating the winning stretch reflected just luck, not any ability to win one-run games. <BR/><BR/>However, teams that lost far more than their share of one-run games *did* show some persistance in continuing to lose them thereafter, indicating their under-performance wasn't entirely bad luck.<BR/><BR/>It seems to be a "Red Queen competition" thing. The talent level in pro sports is really very competitive and fairly equal overall. Teams and players have to compete like all get out just to stay even and not fall behind. So in situations that are very tight and turned by small things influenced by chance, it is very difficult to get a systematic advantage over the rest of the league. Even a good team will win no more in those situations than luck allows. But if your team is outside the normal distribution with *really bad* talent and/or coaching, it's a lot easier to demonstrate systematic inferiority to the rest of the league, which is visible even in close situations highly influenced by chance. <BR/><BR/>Still, James said the great majority of losing in one-run games is merely bad luck. And relating all this to "red zone" NFL QB play is a bit of a stretch. So it may not be worth mentioning here, I admit.<BR/><BR/>I do so *only* because you mentioned Kellen Clemens, and I am a decades-long suffering Jets fan. When I watched him this year he looked dreadful. The NFL rating system had him 33rd in a 32-team league, Football Outsiders ranked him 43rd, etc. <BR/><BR/>As a Jets fan my gut fully expects our QB of the future to be the *one* NFL QB to ever indisputably embody James's "losing in clutch situations can indeed be a skill".JGhttps://www.blogger.com/profile/11164150812219689611noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-80241468497774338552008-01-24T18:23:00.000-05:002008-01-24T18:23:00.000-05:00This makes for a nice companion piece to your 3rd ...This makes for a nice companion piece to your 3rd down performance analysis, although the approach is not quite as conclusive, due to this being based solely on completion percentage. It's still possible that there are some QB skills that are relatively more impiortant in the red zone, and these skills are not reflected in completion percentage.Tarrhttps://www.blogger.com/profile/14368810359650066790noreply@blogger.com