Writers of Pro Football Prospectus 2008

Most Recent FO Features

GurleyTod15.jpg

» OFI: Letdown Saturday

A week after big upsets of Stanford and Ohio State, the USC and Virginia Tech themselves fell to less-talented opponents. Georgia also fell to South Carolina after pummeling Clemson in the opener.

01 Sep 2006

Too Deep Zone: On Bended Knee

by Mike Tanier

Over the summer, the Football Outsiders research staff discovered a statistical trend so revolutionary that it will change the way you think about football.

We stumbled upon this trend by accident, like paleontologists tripping over a stone that turns out to be the skull of some new species of dinosaur. Once unearthed, this trend changed our entire football zeitgeist. It's a powerful, unarguable fact that may alter your viewing or wagering habits. It may even change the way the game is played.

Here's what we discovered: teams that execute two or more "quarterback kneel" plays win 90 percent of the time. Teams that execute one or fewer kneels only win 46 percent of the time. And while winning teams knelt 1.65 times per game from 2000-06, losing teams have averaged just 0.25 kneels per game.*

The data is clear and irrefutable: teams that kneel on the ball win far more often than teams that don't. What's more, the margin of victory correlates closely with the number of kneels: teams that kneel more win far more blowouts, while teams that kneel infrequently are involved in far more tight victories.

There's only one conclusion to be drawn from the data. If you want to win, you have call more kneel plays.

Reflections on Genuflection

Our Kneel to Win theory had the potential to turn the football world on its ear, but at first it was just a rough diamond that needed to be cut and polished. We spent the summer applying state-of-the-art scientific methods to the theory. The results were fascinating.

In games in which the quarterback did not kneel at all, teams average 18.4 points per game. When the quarterback knelt once in a game, teams averaged 20.2 points. In games with two kneels, teams averaged 22.1 points per game. By simple linear regression (a method that relates one variable to another and allows statisticians to extrapolate predictions), it follows that if a team calls 40 quarterback kneels in a game, they will score 92.4 points per game.

Clearly, offensive coordinators are missing the boat with all of their rushing and passing and West Coast Jet Smoke gobbledygook. The data indicates that the easiest way to score in the NFL is not to move the ball at all. Further research proved that the same trend is prevalent at the college and high school levels.

How can teams score more points while kneeling more often? After studying the statistics, we developed several theories. A team that kneels more often is clearly more rested than a team that doesn't, and you can't overestimate the effect of fatigue on football players. It's almost impossible to injure a kneeling quarterback, so teams with a kneel-first philosophy won't have to rely on a backup in a key game (unless the starter gets tendonitis in his kneeling knee). And constant kneeling would allow a team to dictate the tempo of the game while projecting an air of confidence and a winning attitude. Great teams always play with confidence and a winning attitude.

We approached several NFL offensive coordinators about our Kneel to Win theory. We even had some interns devise a "kneel chart," that would show coaches what game situations call for a quarterback kneel. The coordinators were polite, but skeptical. They think of their phonebook-thick playbooks as children, and they are reluctant to part with them. They just couldn't comprehend a truly scientific, visionary approach to their sport. It's a shame, because according to the data, we could be watching football games in which both teams snap the ball and genuflect for 60 minutes, with scores that rival the NBA in the 1980s.

Correlation and Causation

The Kneel to Win theory is pretty stupid, isn't it? Obviously, anyone who thinks that kneel plays lead to victories doesn't realize the difference between correlation (the relationship between two variables) and causation (one variable directly affecting the other).

In the example above, the relationship between the variables is exactly backwards. Kneeling does not cause winning. Winning causes kneeling. Every 10-year-old knows this, and anyone who thinks otherwise needs serious psychological attention.

So we can all chuckle at the Kneel to Win theory. We're all smarter than that. But then, we pick up the local newspaper and read pre-game analysis like this: "The Home Team has to run the ball more often this week. When they run the ball 30 times, they win 90 percent of the time."

Here we go again, folks.

The relationship between running the ball and winning is more complicated than the relationship between kneeling and winning. A good running game certainly contributes to the winning effort. But teams don't win because they run 30 times. Teams run the ball 30 times because they are protecting the lead on their way to a win. It's a fact every thoughtful fan appreciates, and the Run to Win myth was debunked decades before Football Outsiders came into existence. Yet sportstalk hosts and local columnists still trot out the Run to Win theory as if they think that 30 straight handoffs to start the game will guarantee a victory by statistical fiat. That's as silly as thinking that kneeling on the ball is the key to success.

Aaron Schatz founded Football Outsiders three years ago to combat just this sort of lazy analysis. Boston media pundits consistently hammered the Patriots in 2002 for not running the ball often enough; the team ran far more often in their Super Bowl season than they did during that disappointing 9-7 campaign. Why couldn't they see that they were throwing the ball too often? Aaron ran the numbers and proved that the Patriots run-pass ratio early in games changed little from 2001 to 2002, but in 2001 they ran to protect leads in the second half while in 2002 they passed to catch up. The experts had it backwards. While conducting his research, Aaron laid the groundwork for the statistics that would become DVOA and DPAR, paving the way for him to launch the top independent football research site on the Internet.

Still, the Run to Win theory survives and thrives. It has corollaries, too. There's the He Throws Too Much corollary: Brett Favre isn't that great, because the Packers are 7-15 (or whatever) when he throws over 40 passes per game. Of course, Favre only throws 40 passes when the Packers are already trailing, but never mind. And there's the Good Teams Run and Stop the Run postulate, which extends Run to Win to defense. And there are other nuggets of conventional wisdom, like Good Teams Find a Way to Win Close Games, that have little factual basis but live on in the minds of fans and sportscasters.

Of course, if everybody took a careful, thoughtful, analytical approach to football, Football Outsiders would have a hard time standing apart from the crowd.

Learning is Fundamental

Kneel to Win is a snarky parody of the misuse of statistics. But it's also a cautionary tale. Even the most careful, well-meaning statisticians can screw up. At FO, we like to think that we provide the best analysis on the web. But how do we know that we aren't just advancing pseudoscientific claptrap, Kneel-to-Win silliness embellished with a veneer of mathematical precision?

Numbers can lie, they can mislead, they can obscure, and they can send you on mathematical wild goose chases. Sometimes, there's no way to measure the phenomenon you're trying to study. Sometimes, the statistics you want to use to rate a player's performance – like tackles for cornerbacks – are the wrong tool for the job. Because of this, some analysts throw all football stats into the circular file, except the ones that "prove" whatever point they want to make at any given moment. Other observers do the best they can with the numbers that are available, massaging them every which way in a search for underlying truth. That's more or less what we do at Football Outsiders, and we've come up with some mathematical models that rate player and team performance well and, more importantly, do a great job of predicting future performance.

There are other models and systems available, many of them very enlightening and useful, most of them very complex. But just because a statistical model is complicated and scientific-looking doesn't mean it's correct. Just because a writer has a degree in economics or quantum physics doesn't make him right. And just because a conclusion jibes with our gut instincts doesn't mean it's the right conclusion.

DVOA is a great statistical model, but not because it's built on thousands of calculations, or because Aaron is really smart, or because we all just "know" that the best teams must move the ball consistently and convert third downs. It's not a great model because we have a popular website or a deal with FOXSports.com or because Pro Football Prospectus is our ball and we'll take it home if you don't let us play quarterback. DVOA works because it makes excellent predictions, week in and year out, exposing pretender teams with gaudy records and often recognizing rising stars just before they start their ascent. And last year, when the Colts played their Sun Belt Conference schedule and the flaws in the DVOA strength-of-schedule adjustments were exposed, we got out the wrenches and fixed the system, making it more accurate not only for the Colts or the 2005 season, but for every team in the past decade.

Any complex statistical model like DVOA is only as sound as the ideas underlying it. And those ideas are only worthwhile if they are grounded in reality. That sounds simple, but it's a fact often forgotten by analysts who make the Run to Win or Kneel to Win mistake: they start believing their tables and graphs and ignoring what really happens when young men walk onto a gridiron.

Accurate statistical analysis of football requires three things. One is a knowledge of statistics. The second is a knowledge of football. You don't need a BA in mathematics and a decade of coaching experience to do stat analysis; if that were the case, Bill Belichick would be the only person in the field. But you need to know what regression analysis is, and you should be able to spot zone coverage or a trap block when watching a replay.

But the third necessity is crucial: the drive to constantly expand your knowledge of the first two factors. A good analyst strives to make himself a better mathematician and a better football guy. He watches more games, reads more books, talks with more coaches and other experts, all the while experimenting with new calculations and testing new hypotheses. Without the drive to improve, an analyst might develop a "good enough" system, then spend the rest of his energy defending it instead of proving it. In effect, he makes a modest gain, then kneels on the ball.

That's something we will never do. The watchword at Football Outsiders is always "improvement:" adding relevant variables to projection systems, revising and retesting DVOA and KUBIAK and our other stats, accumulating more data, interpreting that data, achieving better results and more accurate conclusions, eliminating spelling errors (that last one's the toughest). We strive to improve so our content is better, but we also do it to learn. As long as we keep learning, our analysis, while never perfect, will keep moving forward, steering wide of the Run to Win/Kneel to Win ditch of faulty logic.

The Geeks Who Triple Check

The Football Outsiders are sometimes called the "stat geeks." In Pro Football Prospectus 2006, Aaron counters that we are really "philosophy geeks," guys who look for true team qualities that transcend wins and losses. I would shed the geek label altogether (if only) and suggest that we're simply the guys who triple check. We don't take anything at face value, even our own conclusions. We put conventional wisdom to the test. We record hundreds of games, scrutinize every play, verify every statistic. We learn as much as possible, so when we post a new statistic or power ranking or conclusion, you can be sure it's the result of keen observation, the critical starting point for scientific thought.

So as you watch football this season, be on the lookout for bad stats, backwards stats, stats that are too good to be true, stats that are too complex to be deciphered, stats that miss the forest for the trees, stats built on a foundation of ignorance or wishful thinking, good stats used badly, and bad stats used indiscriminately. Get ready to wince when the announcer says that some lousy team has the third best pass defense in the league (could it be because no one has to throw the ball against them?), that Team X is great because they have great roster stability (maybe there's little turnover because the team wants to keep quality players) or that the home team is 16-0 when Joe Workhorse gets 30 carries. Greet everything with a skeptic's eye, even what you read in Too Deep Zone or at Football Outsiders. We're all learning about football together, and the more we learn, the faster we can replace "conventional wisdom" with actual knowledge.

*Note: kneel stats are totally and completely made up.

Posted by: Mike Tanier on 01 Sep 2006

50 comments, Last at 12 Sep 2006, 4:56pm by Stats Inc.

Comments

1
by Pat (not verified) :: Fri, 09/01/2006 - 4:11pm

or that the home team is 16-0 when Joe Workhorse gets 30 carries.

My personal favorite is Brett Favre is undefeated at home if the temperature was under 34 degrees when not playing an African-American quarterback wearing the number 7.

Frighteningly, that stayed true last year: because the one game that they lost at home late in the year was Chicago, and the temperature was 34 degrees exactly. Lucky for the Bears. If the temperature had dropped one degree, they would've been doomed, as they didn't have an African American quarterback on their roster!

2
by AJ (not verified) :: Fri, 09/01/2006 - 4:23pm

"we got out the wenches and fixed the system"

Forget the wrenches and bring on the wenches. As everybody knows a good wench is there to bring beer to you while you watch and analyze football.

3
by The Old Schooler (not verified) :: Fri, 09/01/2006 - 4:27pm

Hehe, kneel to win. That's almost as silly as suggesting NFL team's should always go for it on 4th down. Wait a minute...

4
by Tom (not verified) :: Fri, 09/01/2006 - 4:31pm

Random-ish query: is there any statistical significance to the % of opponent drives that go 3-and-out? This seems like a statistic that would correlate strongly with defensive strength, yet the NFL leader last year was the Tennessee Titans, who ranked a well-deserved 30th in def. DVOA. Anybody have any thoughts on this?

5
by B (not verified) :: Fri, 09/01/2006 - 4:32pm

At first I thought this was a late April fools day article, but then you had to go and spoil it by being all smart and stuff.
#3: Maybe we should combine the strategies, and always kneel on 4th down. That's a fool-proof strategy, right?

6
by Andrew (not verified) :: Fri, 09/01/2006 - 4:46pm

By simple linear regression (a method that relates one variable to another and allows statisticians to extrapolate predictions), it follows that if a team calls 40 quarterback kneels in a game, they will score 92.4 points per game.

You laugh, but the greatest Eagles trick play of all time was set up by kneeling on the ball two times to run out the clock. Then Cunningham faked a third kneel, the Cowboys bit, and he tossed the ball for a final rub it in your face touchdown just to run up the score another 7 points.

7
by zip (not verified) :: Fri, 09/01/2006 - 5:06pm

#6 Didn't Marino do that to the Jets as well once?

No wait, that was the fake spike. Which is also awesome.

8
by Will Allen (not verified) :: Fri, 09/01/2006 - 5:09pm

Guys, I received my Prospectus yesterday, and it is truly a great effort. Thank you and congratulations. What is fun for me is to look for things that fall outside the current statistical realm, and see how that may affect a deviation from projected mean wins. As I've mentioned several times already, in my opinion the Vikings last year were among the most badly coached teams I've ever seen, and badly coached in a very unique way, in that the primary blame did not fall on the head coach, which isn't to say that he was anywhere close to blameless. This squad will be interesting to watch this year from the perspective of how quicky a change from terrible ownership can make a difference.

Having said that, I don't greatly disagree with the 5.9 mean win projection for this team, because so much depends on injury, and perhaps that is where some obsessive can make another advance in the statistical analysis of football; a model which predicts the likelihood of future injury to starters within individual units of a roster, based upon previous injury history and age. For all I know the correlation between future injuries and past injuries and age is not a strong one, but that would at least mildly surprise me.

My concern for this team all along has been the number of key players who have significant injury histories, starting with Brad Johnson, of course, and you folks certaintly seem to agree. Is this factored into your projected mean wins? If it is, and it isn't proprietary, what value do you assign it?

In any case, I will stick with my estimation that this team is most likely to win nine games again, largely due to personnel on defense, especially the defensive line, being used much more intelligently, and the offensive line being much more sound than was the case last year. I was sorely tempted to drop that estimate to eight wins after the Vikings lost their nickel back last night for six weeks due to a broken arm, since I think an injured nickel back may be the most underrated loss, in terms of effect on defensive performance, that a team can suffer. Gibbs and Saunders, I suspect, will come out a week from Monday with a ton of three receiver sets. My contention for years has been, however, that there is nothing better than a good pass rush for improving the play of defensive backs, and since my contention is also that the Vikings defensive line will play far better this year than most have reason to suspect, I'll display a modicum of courage and stick with 9 wins as their most likely outcome, and 11 wins as their best case scenario, and 5 wins as their worst case. At least until Brunell throws for three scores in the first half, as Theismann tells us that ol' Mark is a sure-fire Hall of Famer......

9
by Matt (not verified) :: Fri, 09/01/2006 - 5:18pm

I think what the tremendous but under-appreciated FO staff needed to boost morale was a big ol' self-congratulatory pat on its own back, and I for one can't thank you enough for being the guy who provided it.

10
by LnGrrrR (not verified) :: Fri, 09/01/2006 - 5:27pm

Thanks for the chuckle. :)

11
by primantis (not verified) :: Fri, 09/01/2006 - 5:49pm

Sean Salisbury would strenuously disagree with Mike's assertion that kneeling does not contribute to victory.

12
by Fnor (not verified) :: Fri, 09/01/2006 - 6:13pm

I still think always punting is an awesome strategy.

13
by Trogdor (not verified) :: Fri, 09/01/2006 - 6:26pm

"I think what the tremendous but under-appreciated FO staff needed to boost morale was a big ol’ self-congratulatory pat on its own back"

Or maybe, with the new football season about to begin and readership about to get a big increase, they thought a good introduction to what the point of this site is would be a good idea. And if they could do it while finding a humorous way to introduce the new, wider audience to some of the basic concepts (that some of us ol' timey readers have already learnt), well, all the better.

Nah, it's just ego stoking. Bunch o' megalomaniacal bastids, they is.

14
by Vash (not verified) :: Fri, 09/01/2006 - 6:30pm

4: I doubt it. You'd think shutouts would correspond to a good defense too, but there's almost no relationship. Since there are more three-and-outs than shutouts, there probably is some relationship, but most offenses can complete one ten-yard pass against even the best defense.

15
by Matt (not verified) :: Fri, 09/01/2006 - 6:43pm

I didn't say that they were megalomaniacs. Since this is a statistics-oriented site, I will restate my point as an equation.

Introduction to site = fine.

Humor (or attempts at same) = fine.

"At FO, we like to think that we provide the best analysis on the web" NOT = introduction to site OR humor.

Sometimes people who say "we are the best" are right, but that doesn't mean it's not self-congratulatory to say it. If this site is about science and proving it, can't we just leave the subjective value judgments about who is the best to someone else? Or has Mr. Schatz also devised a revolutionary system for quantifying conventional wisdom testing and overall scrutinizing ability?

16
by ptfe (not verified) :: Fri, 09/01/2006 - 6:50pm

Re 4: is there any statistical significance to the % of opponent drives that go 3-and-out?

If you're talking strictly about percentages, it's important to remember that failure to stop the other team usually leads to fewer drives overall for both teams. For example, Tennessee played 5 top-8 DVOA offenses and forced 7 3-and-outs, but they only had 46 opportunities -- about 15%. Conversely, the Titans' offense isn't very good (-6.2, #18), so we'd expect them to have more 3-and-out opportunities against a poor opposing offense as well: against the 5 bottom-8 teams they played, they had 31 4-down stops in 82 chances, or about 38%.

The problem is the relationship between 3-and-outs (or 4-down stops) and number of drives. If your D is good, you probably get a fair number of them, but if your offense is also good, it will control the game better and your percentage will generally not be boosted as much.

17
by jake (not verified) :: Fri, 09/01/2006 - 6:53pm

personally, i judge the quality of analysis by the spelling mistakes - the more mistakes , the better the analysis.
which is why i am such a big fan of football outsiders.
like they say - it's all in the numbers.

18
by Bobman (not verified) :: Fri, 09/01/2006 - 7:15pm

Mike T, Jonathan Swift would be proud. So when is your football self-help book "Kneel Your Way to a Superbowl Victory" coming out? Hey, this is nothing new--didn't Anna Nicole Smith kneel her way to a $400M fortune? What, it's not quite the same? Never mind.

Wait, I just read your F-ing footnote about making up the kneel stats. Shaaaaame on you, you lying bastard! It's all lies, damn lies! I'll never believe anything I read at FO again (unless written in the Paytom Branning thread by ROBO PUNTER).

19
by CaffeineMan (not verified) :: Fri, 09/01/2006 - 7:24pm

Hey LnGrrrR, where you been, man?

And yes, MikeT thanks for the laugh. And thanks to the FO staff for pursuing good science and not simply good discussion.

20
by Tom (not verified) :: Fri, 09/01/2006 - 7:27pm

Re #16
I'm afraid I'm not understanding what you say when you say "if your offense is also good, it will control the game better and your percentage will generally not be boosted so much." I don't see how an increase in the number of drives leads to an increase in the PERCENTAGE of 3-and-outs. If the Titans had had a better offense, resulting in only 61 opponent drives against the bottom-8 teams they played, then shouldn't you expect about 23 3-and-outs, for the same 38%?

One thing I wondered is that 3-and-outs are more strongly correlated with mediocre but not awful defenses with not particularly good offenses. The idea here is that teams were able to take the lead on the Titans early, and ran the ball in the second half to take time off the clock, often without much success. But it didn't matter if they weren't very successful, because they didn't have to fear a comeback from the Titans offense. The split by DVOA you provided (and thank you for that) would seem to cut against that argument, though.

21
by Jake (not verified) :: Fri, 09/01/2006 - 7:31pm

Where can I see stats on % of opp. drives that go 3 and out?

22
by David Carr (not verified) :: Fri, 09/01/2006 - 7:56pm

I head from somewhere that kneeling on every play will significantly lengthen a quarterbacks career

23
by Larry (not verified) :: Fri, 09/01/2006 - 8:10pm

Re: 4,20
2005PFP had a piece by Jim Armstrong on drive momentum where stopping a team 3-and-out is just as likely as stopping them on the next series of downs. That would indicate that 3-and-out percentage ought to be very representative of overall defensive strength. His study lumped all teams together though, so it isn't clear that this holds true for every team individually, but I'm thinking it probably would. At least to within statistical small number issues, anyway. So, it probably wouldn't reveal anything new except that defenses that can stop first downs from happening are good. Which we already know.

Re: 15
Thought that the 'We like to...' statement was inside a paragraph of pretty strong humility, quite frankly.

24
by Cback (not verified) :: Fri, 09/01/2006 - 8:13pm

How about the oft-repeated comments by announcers that the team with greater time of possession is more likely to win and that winning teams have a better turnover difference (actually, they usually say "differential").

My gut feeling is that it's much more of correlation than causation. A team that's behind will take more chances and usually suffer turnovers trying to catch up, causing a bad turnover difference. And teams that are ahead will run out the clock, racking up their time of possession.

I still like the analysis that shows the team that scores first (or leads after one quarter) is much more likely to win. I don't expect TV commentators to mention this, though, because it just encourages people to stop watching the game.

25
by admin :: Fri, 09/01/2006 - 9:08pm

Re: 15. Um, what's the byline on this article again?

26
by John P (not verified) :: Fri, 09/01/2006 - 9:11pm

Re #4

Losing team defenses are given easy 3 and outs at the end of games.

27
by Will Allen (not verified) :: Fri, 09/01/2006 - 9:28pm

Might not % of three and outs while within seven points of the lead, or at most seven points ahead, be a useful number for evaluating a defense? Or would the sample size be too small?

28
by Basilicus (not verified) :: Fri, 09/01/2006 - 11:31pm

"Or has Mr. Schatz also devised a revolutionary system for quantifying conventional wisdom testing and overall scrutinizing ability?"

Yep, but he's only shown it to a select few. It's really quite impressive, but he can't release it to the general public for fear of Michael Irvin releasing his counter-theory, tentatively titled the "Loudness is in the Ear of the Beholder."

Seriously, #15, calm thyself. Put your feet up, drink some red wine, and pop in some Handel. This is an article of intent - FO does something like this every once in a while to restate its goals and intent. No better time like right before the season. Doesn't Mike also say to read everything with an analytical, crucial eye - even what you read at FO. It's a reminder that they're trying to think about the game on a deeper level and, hey - they like it when other people do that, too. Makes for a better readership and a readership that can point out mistakes and faulty logic instead of just believing everything Aaron and co. decide to tell us. One of the greatest dangers of today's culture is to believe the conclusions that any body of media says without cross-checking those conclusions against your own knowledge and that of sources which disagree.

Gods know I don't agree with every article posted here, but at least every article posted on FO fosters intelligent analytical debate. Rant off.

29
by empty13 (not verified) :: Fri, 09/01/2006 - 11:59pm

The stat that matters is how the team does while the game is still winnable, or better yet, when the game is on the line.

When most teams go 2 TDs down, it's time to watch 'Heidi'...

And most teams, if they go down 3 TDs in the first quarter (or up, for that matter), get so far away from their original game plan that all the stats become meaningless.

30
by Tom (not verified) :: Sat, 09/02/2006 - 12:29am

Re #23/26/27
#23: The puzzle, though, is that the Titans were very good at forcing teams 3-and-out, but not a very good defense. #26 has the obvious answer, that you'd expect this to happen a lot at the end of the game against decent offenses. Except, the stats in #16 seem to indicate the Titans were MUCH better at forcing 3-and-outs against worse offenses, which makes some sense, except that the games were closer, as a general rule. #27 would be an interesting research project; maybe I'll have to check out the old drive charts and see if there's any trend.

31
by Ryan Mc (not verified) :: Sat, 09/02/2006 - 11:42am

I think if the Bears had just knelt three times and punted each offensive possesion last year they might have gone undefeated. Actually trying to move the ball on offense really hurt their chances of winning.

32
by steelberger1 (not verified) :: Sat, 09/02/2006 - 11:57am

This is a pretty good article. I have noticed that certain sites (cough*CHFF*cough) just make the numbers say what they want them to in order to prove their opinions. It is kind of irritating.

33
by Crushinator (not verified) :: Sat, 09/02/2006 - 12:51pm

I just can't wait til "Spiking the Ball to irritation" where a team spikes 1-3 downs and punts on 4th to drag out a game as long as possible and attempt to earn the networks in more money for more commercial slots.

34
by Drew Bledsoe (not verified) :: Sat, 09/02/2006 - 4:45pm

How am I supposed to kneel? I'm an 8' 20" statue! I would rather just get sacked. Besides, it gives our strong-legged punter more room to pin the opponent inside their 50 yard line.

35
by ROBO-PUNTER (not verified) :: Sat, 09/02/2006 - 4:48pm

#12: So do I.

36
by ptfe (not verified) :: Sun, 09/03/2006 - 11:53am

re 20: I don’t see how an increase in the number of drives leads to an increase in the PERCENTAGE of 3-and-outs.

The idea is that, if you have a good offense, you generally don't have as many drives for/against you. Which means that your number of drives per game is consistent whether you're playing a lousy team or a good team. So the percentage stays true game-to-game (higher against crappy offenses, lower against better ones), but your season-long percentage is less affected by any given outing. In fact, the Titans SHOULD expect the same 38% against crappy teams even with a lower drive count, but the relative weight of these games drops.

For an extreme example, consider getting 10 drive opportunities vs 90 drive opportunities. If you go 100% against the 90 drives and 10% against the 10 drives, your total is 91/100. If instead you get 10 and 40 with the same percentages, you drop to 41/50 (82%). Your percentage dropped because there's a smaller "weighting" on that game against the bad opponent.

In the context of the Titans, they had significantly more opportunities (generally about 25% more) against teams like San Francisco than teams like Indianapolis, which means we'd expect a greater weight on those games.

Jake: check the NFL.com website, then look up drive charts; they give the number of plays and result of each drive.

37
by Sophandros (not verified) :: Sun, 09/03/2006 - 12:44pm

Good article, and should be required reading for new posters and Falcons fans.

38
by Willsy (not verified) :: Sun, 09/03/2006 - 9:28pm

#8
Will - I too am a Vikings fan and I am floating in the 6 - 7 win area mainly because of the horrible start we have and the new coaches/schemes that have to be assimilated into the club. Hopefully things are on the right path and they will provide more interest on the field than off it!
Team - I am reading the Prospectus at the moment. Best ever! The individual team analysis seems a bit more punchy and I showed the first Prospectus and the recent one to my wife (a professional writer) and she thought you could see a marked improvement in the tone and style of the writing from everyone. It seems you guys are more in a groove and have a clearer idea of who the audience is. I am finding it really easy to read, funnier and it seems much more focused. A great book and a great read.
I am reading it straight after reading "The Draft" and "Next Man Up" which has provided some interesting context regarding the backroom decisions the various teams would be considering at the moment.
Keep up the great work, you help improve the appreciation of an already great game.

39
by Tom (not verified) :: Sun, 09/03/2006 - 10:26pm

Re #36
Yeah, that makes sense. I feel dense now. The question now is, how could/should you test it to take that into account? I'm thinking % of 3-and-outs in a game v. per-game def. DVOA as a first step, but I'd ideally want something less complex than per-game def. DVOA as a baseline.

40
by DGL (not verified) :: Mon, 09/04/2006 - 12:01am

By my calcuations, in 2005 (including postseason where applicable), the top five teams in percentage of opponent three-and-outs were:

DAL: 31.6% (60 of 190 drives)
TEN: 31.3% (61 of 195 drives)
PHI: 30.6% (64 of 209 drives)
JAC: 29.1% (57 of 196 drives)
NO: 28.1% (48 of 171 drives)

These teams were, respectively, 13th, 30th, 14th, 7th, and 25th in defensive DVOA. The correlation coefficient between percentage of opponent three-and-outs and defensive DVOA is 0.29. So while the there's not a lot of correlation at the top, there's a weak overall correlation.

On the other hand, in 2004 (again including postseason), the top five teams in percentage of opponent three-and-outs were:

WAS: 29.8% (59 of 198)
BUF: 29.65% (59 of 199)
PIT: 28.9% (57 of 197)
PHI: 28.7% (62 of 216)
BUF: 28.65% (55 of 192)

These teams were, respectively, 4th, 2nd, 3rd, 16th, and 1st in defensive DVOA. The correlation coefficient is 0.66.

So I wouldn't say there's a strong correlation (I haven't checked 2003 or 2002 yet), but there's some correlation.

I haven't dug into game-to-game variations.

41
by emcee fleshy (atl/sd) (not verified) :: Mon, 09/04/2006 - 1:12am

bravo.

42
by BillWallace (not verified) :: Mon, 09/04/2006 - 3:35am

Credit where it's due. I was watching just a few minutes of the NFL Network telecast of the Steelers Panthers, and the Steelers homer announcers had this nugget (Paraphrasing).

"Everyone always says that the Steelers are a run-first team, but look at these stats from some of last season's victories"

-- Stats showing run-pass ratios by half, typical was 1st half 13R/16P 2nd half 22R/9P --

"So really they pass to get a big lead and then run to salt the game away"

43
by Mitch Cumstein (not verified) :: Mon, 09/04/2006 - 4:20am

Mike Tanier is the foot(ball) #$@!*# Master!

Although coaches are not ready for your revolutionary analysis because coaches "play not to lose" rather than play to win. Perhaps speak their language by also pointing out what they should avoid doing to win more.

For example, coaches who call more timeouts when their team is on defense tend to lose more games. Teams that have more kickoff returns tend to lose more also. We hear about the turnover ratio and how winning that battle is important, but what about winning the offense vs. defense timeout ratio, and the kick return ratio? Clearly teams need to not only be kneeling more, but also need to avoid calling defensive timeouts, and advising their kick returners to let every kick possible go to the end zone for a touchback.

44
by Trogdor (not verified) :: Mon, 09/04/2006 - 3:25pm

I bet the correlation is especially bad if you call defensive timeouts within the last two minutes.

I just wanted to express my disappointment when I read that the kneeling stats were just made up. Why? Don't we already have that data, by comparing FO-runs with official NFL-runs by QBs? I know that might be a ridiculous amount of number crunching for little marginal value, but I still would've loved to have seen the real numbers. Maybe in the next mailbag...

45
by kleph (not verified) :: Mon, 09/04/2006 - 10:23pm

it might do well to add that there are quite a few of us regular vistors to the site that are not stats-savvy and certainly do not have an archival knowledge of football a lot of the folks around here do - including both FO'ers and commentors. but we love to watch the game and we appreciate a tool that helps us understand it better.

of course, as a philosophy grad i certainly know post hoc ergo proctor hoc when i see it.

46
by Brian L Cartwright (not verified) :: Tue, 09/05/2006 - 5:46am

re 36: to avoid the sample being biased because of the proportions, construct an expected value based on the number of opportunities in each situation, and then compare to the observed value (which i believe is how VOA operates)

47
by Dr. Evil (not verified) :: Tue, 09/05/2006 - 8:32am

thanks for the elementary science tip. Most of us educated peoples understand correlation does not = causation.

A worthy lesson for the rest nonetheless.

48
by Matt (not verified) :: Tue, 09/05/2006 - 6:30pm

Re: 28 -- "Seriously, #15, calm thyself. Put your feet up, drink some red wine, and pop in some Handel."

Why does it appear that I am not calm, or that I might benefit from your NPR-approved stress relievers? I said the article was self-congratulatory, no matter what it's other purposes may have been (introducing FO to new people, teaching us to read things with a critical eye, etc.) Laudable purposes all, perhaps, but it's still a bit self-important. Isn't it?

Aaron -- I know what the byline says. My reference was to Mike's note about your motivation for launching the site in 2002. I said that the article was a self-congratulatory piece by the FO staff, which is a fair attribution (even if you don't agree with my judgment) because Mike clearly says he is writing on behalf of "we" the FO staff. I did not say that Aaron Schatz wrote a self congratulatory piece declaring that "DVOA is a great statistical model . . . ." but unless I am reading wrong the byline is that of a person on the FO staff.

49
by Fnor (not verified) :: Tue, 09/05/2006 - 8:39pm

Matt: so, saying that your model is good in an introductory article is self-congratulary? Should we not have grant proposals? I mean, they involve bragging about your stuff in front of an audience (your peers, grant boards, etc). Heck, you seem to be offended when someone suggests you listen to Handel!

The story had an amusing farce and a good moral about interpreting statistics. It wasn't like reading the transcript from WWE Smackdown. There's precious little here to complain about.

50
by Stats Inc. (not verified) :: Tue, 09/12/2006 - 4:56pm

Re. #48

You nailed it! The FO staff do think they are way too big for their britches! What a bunch of self-congratulatory jerks!