What Makes Teams Win? 2

This is a contiuation of an article discussing the relative importance of factors in winning NFL games. Part 1 is here.


Turnovers can be analyzed in the same manner. If we want to know how good a defense is at intercepting passes, we should look at interceptions per pass and not total interceptions. Offenses may shy away from passing the ball if a secondary is known to be excellent at generating interceptions. Their ‘total interceptions’ stat would suffer as a result of teams leaning towards the run, but their interception efficiency would tell the real story.

Fumbles are slightly different. Once the ball is on the ground, who recovers it is a very random thing. Fumble recoveries are almost purely the result of unplanned conditions, such as where players happen to be and who happens to be facing the direction of the ball. (Note: Skills like running, passing, or generating interceptions are repeatable and enduring skills which tend to carry over from year to year. Fumble recovery percentage does not.) The best measure of the fumble, therefore, is the forced fumble, which does carry over from year to year. Defenses can be good at jarring the ball loose, but who recovers the ball is random. The more often the forced fumble, the more opportunities the defense has of recovering. Forced fumble rates correlate better with wins, which bolsters its case as the stronger factor in winning.

StatWin Correlation
Def Fumbles0.33
Def Forced Fumbles0.41

On the offensive side of the fumble, the opposite is true. Unforced fumbles are the repeatable, enduring skill--or I should say avoiding the unforced fumble is the skill. Therefore, the fumble per play stat is the optimum stat to measure an offense’s tendency to fumble. Keep in mind that fumbles happen on both pass and run plays.


So far I’ve ignored special teams. There’s a reason for that—they don’t really matter on a consistent basis. Of course, we can remember a critical play here, a miracle play there--they can change the outcome of a game on rare occasions. We can remember these spectacular plays because they are so exceptional. But no matter how we measure special teams in a meaningful way, the correlation coefficients are not statistically significant. That’s not to say that teams can ignore special teams. Every advantage a team can find is important, but in the long run, the importance of kickoffs, punts, return yards, and field goal percentages are all dwarfed in relation to things like running, passing, and turnovers.

There are several valid explanations for the relative unimportance of special teams. First, and perhaps most obvious, is that special teams plays are far less frequent than pass or run plays. Second, the difference between the best special teams and the worst special teams is small in comparison to the spread between the best and worst offenses and defenses. Third, special teams stats are very hard to evaluate. For example, average net punt yards negatively correlate with winning at -0.17. This means the better the team is at punting, the fewer wins it can expect. But I believe this is misleading because teams with good offenses would be able to kick more “inside the 20 yard line” punts than teams that often kick from their own 30 yard line.

One last point about special teams: Big special teams plays, such as kickoff returns for touchdowns, are rare. In the past five seasons, there has been an average of 13.6 kickoffs returned for touchdowns--by all 32 teams combined. The best test of skill vs. luck I have come across asks the question, “does it tend to be repeatable?” Kickoff or punt returns for touchdowns, or 60+ yard field goals, unfortunately, do not tend to be repeatable.


Penalty rates, however, do tend to be repeatable. Teams that have committed lots of penalties tend to continue to commit more, and vice versa. Experience, discipline, preparation, and playbook complexity probably all contribute to penalty rates. Other factors probably come into play. For example, it might be expected that weak secondaries would tend to commit more pass interference penalties in desperation to prevent deep passes.

Penalties and penalty yards correlate negatively with winning, as we’d expect. (Penalty rate is defined as penalty yards per play.)

StatWin Correlation
Team Penalties-0.21
Team Pen Yds-0.18
Team Pen Rate-0.37

The penalty rate stats have the stronger correlations with wins than either penalties or penalty yards.

Continue reading part 3 of this article.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

8 Responses to “What Makes Teams Win? 2”

  1. Derek says:

    On special teams: I don't think you can entirely dismiss the importance of special teams. Starting field position does directly affect a team's probability of scoring by a significant amount. The ST stats available in box scores are just really lossy information. To get a good special teams input, however, you'd need to do quite a bit of legwork (no pun intended). You'd need a metric of how much the return/coverage unit increased/decreased the probability of scoring on the subsequent drive. For that, you'd need information on every drive and every game for a few seasons at least to get some good idea of P(score | start at X yard line). And then you'd need the play-by-play data to figure out from where the ball was kicked and to where it was returned. I believe that's more or less how Football Outsiders does it. I don't mean to treat them like demigods, but they've obviously researched this more than I have.

    On penalties: I was thinking about that maybe more defensive pass penalties are indicative of tighter coverage on WRs. Pass interference and illegal contact calls are always inconsistent (IND@NE 2003 AFC Champ.). If the DBs are mugging the receivers throughout the game, pass efficiency is going to go down, and the few penalties they get called for are acceptable losses. Or perhaps better pass defenses are able to cover recievers so tightly without major contact that they are targets for ticky-tack calls.

    If you were able to break down the penalties by type and check correlations with pass def. eff., that might explain the positive correlation of defensive penalties to wins.

  2. Brian Burke says:

    I like your theory on def penalties.

    On ST, I agree drive starting position is important, but it's not really dependent on ST play. Starting field position is overwhelmingly due to offensive or defensive play.

    Consider two extreme examples. The game starts and your team kicks off. Your defense gets a 3 and out and your team gets the ball on your own 39.

    Your offense gets 2 first downs but stalls at your opponent's 40. It's 3rd and 9 so you punt. Your opponent starts his drive on his own 15, or 20, or whatever.

    In the second example, your team kicks off, and your opponent gets 3 1st downs, then punts. Now you're starting from your own 15.

    See the swing. It's huge and has very little to do with special teams. They can affect things at the margin, so perhaps you're starting at your own 25 instead of your own 15. But that difference in field position is dwarfed by the direct effect of the offense and defense.

    To put things in perspective, a single holding penalty causes a 1st and 20 (or could call back a long TD pass). The best kick return team in the history of the NFL gets you and extra couple yards per kick. I exaggerate a bit, but you see the point.

    I agree that you can't dismiss ST. But the effect is very small and very difficult to measure. I guess that's the legwork you mentioned.

    Maybe we could add FOs metrics into a comprehensive model to see if they're on to something.

  3. Derek says:

    Oh, I totally plan on filching FO's stats. It's just a shame they don't have weekly archives of DVOA beyond overall Off/Def/ST.

    An another idea for a ST stat is how much field position is gained by the coverage/return unit on average. This would be a little less dependent on field position because you'd be counting touchbacks against the coverage unit. Punts don't usually go further than 40-50 yards, and team don't normally punt from within the opponents' 35-yard-line. So the pooch punts to pin down opponents near their own end zone shouldn't adversely affect such an average.

  4. Tim Folkerts says:

    "Penalty rate is defined as penalty yards per play."

    Is that offensive plays? Offensive + defensive plays? Including kicks and special teams? After all, penalties can come on any play on either side of the ball.

    If it is all plays, then I can't see why penalties/play should be much different than penalties, since most games have a similar number of total plays. (Of course, one of the great things about stats is discovering things you would not otherwise have suspected. :-) )

    If it is offensive plays, then it would seem that no matter how many penalties you have, you are less likely to win if you have fewer offensive plays. So the combined stat would simply be reconfirming that fact.


    As an extension to what you have done, I would think that separating out "offensive penalties per offensive play" and "defensive penalties per defensive play" would be informative. Then you would know if offensive or defensive penalties are more critical.

  5. Robbie says:

    Kick Return TDs may not be repeatable, but Kick Return TDs Allowed are very repeatable. See: 2009 Steelers.

  6. J. Rolla says:

    @ Brian,

    You have a lot of patience with people who obviously don't understand regression analysis, and who comment without bothering to look at the actual statistics. It is borderline insulting to your research lol. Either way, I can respect your tolerance.

  7. Anonymous says:

    Perhaps a good way to measure special teams efficiency would be to find the expected starting field position for a team based on where they are receiving it from. For example, if you'd expect a team punting from the 50 to give the ball to the other team on the 15 (I'm of course just making up numbers here), and the punter stuck it on the 5, he'd have added to the team's chances of winning. But if he booted it into the endzone, he'd have subtracted from his team's chances of winning. You could measure punters' WPA by subtracting the expected starting field position from the actual starting field position. I'd be interested to see what the correlation between a strong punter WPA and winning is. The same principle could be applied to punt returns and kickoffs/kickoff returns. I am new to the site and am unaware if you are already doing this, so I'm sorry if I sound a bit uninformed right now.

  8. Anonymous says:

    Interceptions by a defense are more "given up" by the opposing qb than "earned" by a defense, no? There are certain defenses that are demonstrably better than average at earning interceptions, but the opposing teams qb stats have a lot more to do.with it I would think. Why not run a regression on defensive picks relative to that defenses past ints/pass attempt and the opposing qbs past ints/pass attempt.

    You could further the results by adjusting the qbs previous picks to how good of defenses he played in those games (a qb strength of schedule adjustment so to speak).

Leave a Reply