Sustained Outrage

Upsetting upsets: Football and domestic violence


Many of us have heard the oft-repeated canard that Superbowl Sunday is the worst day of the year in terms of domestic violence. (More people have money riding on the game, and/or have been drinking, and if the wrong team wins, tempers flare and violence ensues, goes the conventional wisdom.) This conclusion hasn’t held up to closer scrutiny, and I don’t think very many people put much stock in it these days.

But what does cause a spike in domestic violence, according to a new academic study titled Family Violence and Football, is an upset in the NFL. Using Las Vegas spreads as an indicator of the expected outcome, economists David Card of the University of California Berkeley and Gordon B. Dahl of the University of California San Diego tracked reported incidents of family violence in areas loyal to a particular team. When the hometown favorite lost unexpectedly, reports of domestic violence went up 8 percent. And particularly wrenching losses, like in a playoff game, made it worse.

Sadly, the inverse does not seem to be true: unexpected wins do not decrease the reports of domestic violence significantly, the study found.

(While the Superbowl Sunday brutality may be a myth, the study notes that domestic violence does go up on holidays: Christmas Day +17 percent, Thanksgiving +22 percent, Memorial Day +30 percent, New Year’s Day +19 percent, New Year’s Eve +32 percent, and July 4th +28 percent.)

But back to football. Here what Card and Dahl learned:

Our empirical analysis points to four main conclusions. First, controlling for location fixed effects, the pre-game point spread, seasonality and weather factors, and the size of the local viewing audience, we find that an “upset loss” by the home team (a loss when the team was predicted to win by more than 3 points) leads to an 8 percent increase in the number of police reports of at-home male-on-female intimate partner violence. The spike in violence is concentrated in a narrow time window after the end of the game. Second, consistent with the standard behavioral prediction that losses matter more than gains (i.e., loss aversion) “upset wins” (a win when the team was expected to lose by more than 3 points) have at most a small effect in reducing family violence. Third, upset losses in more salient games (those involving a traditional rival, or occurring when the home team is still in playoff contention) have a larger effect on the rate of intimate partner violence, as do unexpected losses after games involving an unusual number of sacks, turnovers, or penalties. Fourth, we find that NFL game outcomes have little or no effect on police reports of female-on-male intimate partner violence. We also find no significant effect on violence against children, though the low rate of police-reported violence against children limits the power of this analysis.

So, some (but only some) incidents of domestic violence can be attributed to aggravating factors that set the batterer off, Card and Dahl note:

From a broader perspective we draw two main conclusions from our findings. First, at least a fraction of intimate partner violence appears to represent expressive behavior that is triggered by payoff-irrelevant emotional shocks, rather than strategic instrumental violence that is used to control an intimate partner. We emphasize that we are not arguing against an instrumental interpretation of some types of family violence. Rather, our results show that a loss-of-control paradigm may be useful in understanding some part of intimate partner violence. Second, there appears to be empirical support for use of a gain-loss utility framework (with a “rational” reference point) for interpreting the effect of emotional cues on the loss of control in intimate partner interactions. We suspect that the same framework may prove useful in other settings where economists are trying to model the effect of visceral influences on observed behavior.

Something to keep in mind with the playoffs right around the corner.