THN Analytics: The statistical argument against fighting
THN Analytics: The statistical argument against fighting
The age old fighting debate rages on, but a statistical analysis of the effects of fighting shows that the true impact of fisticuffs may be less than we could have imagined. In fact, not only is it less effective, it may be making the game more dangerous.
As the debate over fighting continues, it certainly appears that the anti-fighting crowd has gained the upper hand. While the debate has often been framed around a series of defenses for fighting, including the concepts of providing a boost, deterring harm against a team, and the idea that you’ll lose fans if you phase fighting out of the game, all of these issues can be addressed by looking into statistical evidence. I also think there’s a fourth point that hasn’t really been discussed: the difference between employing a fighter that can’t play the game particularly well, and a regular 4th liner. Let’s take a look at these arguments point-by-point:
A “Boost” From Fighting?
Just from a base standpoint, we know that both teams are involved in a fight, so a fight shouldn’t really favor one team or the other. Taking a random sample of 50 fights from last season, the swing in possession before and after a fight ranges plus or minus 3-5% in possession. While this seems significant, taking snippets of play from 50 random non-fight games reflects the same kind of in-game possession shift. The evidence does suggest, on the other hand, that shot attempt rates appear to increase after a fight, from approximately 0.67 shot attempts per minute to 0.83 – but not because of a boost…
Fighting as Deterrence
While hits as a normal course of play seem to go down slightly, the rate of “dangerous” penalties (roughing, elbowing, illegal contact to the head, cross checking, kneeing, checking from behind, boarding) nearly doubles. A potential factor could be referees calling the game closer to keep things under control, though it’s worth noting that these particular types of infractions are pretty cut-and-dry when called. This boost in penalties can certainly result in the increase in shot attempts I mentioned above. It’s also worth looking at last year’s teams to see if fighting – or having a fighting reputation – has a strong relationship to fewer dangerous penalties committed against your players:
Each team is represented by a dot; engaging in more fights pushes the dot further to the right and receiving more dangerous infractions-against pushes the dot upward. Though the relationship is not strong, the data definitely suggests that, if anything, there’s a slightly positive relationship between fighting and your teammates being targeted by dangerous plays. Setting the data aside, from a social standpoint the “fighting as deterrence” idea relies greatly on the idea that the kind of players that would hurt others will not do so for fear of getting into a fight. There’s two problems with that assumption: a) as Sean Avery has taught us, you don’t have to fight if someone challenges you, and b) a majority of young hockey players grow up fighting, whether it’s locker boxing or fighting drills set up by their coach. The irony of the growing popularity of fighters is that it desensitized generations of young players to the very fear a fighter’s career relies upon.
Fighting as Entertainment There is another potential argument that has some legs: that fighting is an entertaining aspect of the game, and by removing it you risk losing fans. Looking at it statistically, there are around 40 instances where teams increased or decreased their fighting rate by 50% or more from one season to the next, and the comparison of its impact on attendance is telling:
The middle bars exclude the teams who were a part of the league-wide drop in fighting from 2003-04 to 2005-06, wherein the league also saw an attendance boost that was more likely due to fans eager for the NHL’s return. This argument is further refuted by
a recent survey from Roy MacGregor at Toronto’s Globe & Mail where two-thirds of Canadian respondents said they would support a ban on fighting at the professional level. On all accounts, there simply is not strong evidence here to support the idea that more people will fill the seats for a team that fights.
Employing the Fighter Finally, it’s important to look at what the trade-off is for teams that include an all-fight, no-play skater in their forward lines. Taking the 27 primary 4th line fighters (20+ GP) and comparing them to the other 82 4th line forwards from last season, the impact is cringe-worthy:
Despite getting more opportunities to start their shift in the offensive zone, all-fight players drastically tip the ice against their team, even worse than your typical 4th line forward. Some of that negative impact is mitigated by the fact that a 4th line fighter will play about 2 fewer minutes per game than a regular 4th liner, but given the dubiousness of the value of fighting, what’s the point?
Conclusion Fighting in hockey is one of the more peculiar actions among the major sports, in that it’s taken on a form and a supposed function that made it a regular and allowable occurrence. Ultimately, I don’t think a fighter is necessarily a bad person, nor are they wrong for choosing this path to the NHL. Really, it was the league that made fighting into what it was, and cultivated generations of players who realized it was the only way they could get to the top. But let’s not get locked up in a discussion of “good people” not getting a shot anymore; instead, let’s remember there were a lot of good, deserving players that didn’t get a shot for decades because they didn’t fight, or were too small, or too “soft.” And for goodness sake, if you have fighters that can’t play the game but you really like them, make them an assistant coach or the head of the Department of Player Safety or something – because fighting just isn’t worth it.