There are several alternative titles for this particular article including “How important are penalties in sports”, “Which sport requires the most discipline to win”, and “The Most Powerful Zebras”. Regardless of how we title it, the central question is this: between pro basketball, hockey, and football, where do penalties have the biggest impact on whether a team wins or not?
This question was posed by a friend of The Renaissance Fan. He speculated that penalties have the biggest impact in hockey because it results in a player being removed from play, often resulting in a goal, which is a big deal in a low-scoring game like hockey. But you could make a similar argument for the other two sports. In football, penalties keep drives alive and influence field position, which are both a pretty big deal. In basketball, the effect of penalties is pretty complex because they can lead to players leaving the game or having reduced minutes, changing possession, and leading to direct scoring opportunities (free throws).
To dig into this question we obtained penalty data from every team in the NFL, NBA, and NHL over the last 4 full seasons (2013-2016 for NFL, 2012-2015 for NBA, and 2011 then 2013-2015 for NHL to exclude the year impacted by the strike). We chose to exclude MLB because penalties don’t really exist in baseball. We also chose to ignore MLS because most soccer penalties have little impact on game play unless they result in a penalty kick or a free kick close to the goal. For the NFL our measure of penalization was average penalty yards against a team per game. For the NBA we used average team personal fouls per game. For the NHL we used the average number of penalty minutes served per game. All these metrics were selected because they took into account that football and hockey penalties can have different magnitudes (i.e. a 5-yard false start vs a 47-yard pass interference penalty). Using a regression method we tested whether our penalty metric had a significant effect on a team’s winning (measured as regular season winning percentage). A flat regression line means that the amount of penalization has little effect on winning. A downward sloping line means that more penalties means less winning. Here’s what we found for each sport:
You know that feeling when you watch your football team commit a dumb penalty and you just know it’s going to screw them in the long run? Well, you should trust that feeling. The data show that penalties are, in fact, a big deal in football. The most successful teams are typically giving up less than 55 yards in penalties per game. Those giving up more than 60 yards per game are typically below .500 (though there are exceptions, the 2016 Raiders gave up 77 yards in penalties per game yet won 12 games).
Conclusion: penalties in football are a big deal.
The fouls vs. winning pattern for the NBA looks similar to the NFL, with a downward sloping trend line indicating that fouls lead to less winning. However, the trend is not quite as strong in the NBA, meaning that fouls have a little less of an impact. Teams committing less than 20 fouls per game are slightly more successful than those committing more than 20, but there are plenty of exceptions. One reason fouls may not impact basketball as much may be that there is not much variation from team to team. In other words, all teams we looked at from the last four seasons fouled at about the same rate and all averaged between 17-24 fouls per game. In the NFL, the spread was a lot larger, possibly because there are less games played, giving more impact to one or two really bad games.
Conclusion: penalties in basketball are only sort of a big deal.
Here’s where things really get weird. Remember our hypothesis that penalties would affect hockey most? Well, the data don’t provide much evidence for that. Our trendline is essentially flat, meaning that teams have the same win percentage whether they are penalized a lot or very little. In fact, in the last four (full) seasons many of the teams that were penalized most (more than 12 minutes of penalties per game) did very well. What does this mean?! Are hockey penalties reflective of a more successful strategy of being aggressive? Are the most winningest teams those that are really good at power play defense? Is the entire system by which hockey penalties are determined flawed? Getting to the bottom of all that will require some more in depth analysis. But in the meantime, NHL players, go nuts!
Conclusion: penalties in hockey may not be a big deal