Is There a Grain of Truth in the Super Bowl and Domestic Violence Urban Legend?

I once characterized the statistic that domestic violence increases by 40% on Superbowl Sunday as one of those “vampire numbers” that gets continually repeated in policy debates even though there isn’t and never was evidence of its veridicality. The logic model underlying the Superbowl violence urban legend is that victorious male fans feel powerful and thus entitled to beat their wives, whereas defeated male fans vent their frustration by engaging in the same despicable behavior.

In raw form the urban legend has no evidence behind it, but my friends in the London MET tell me that domestic violence calls (indeed all violence calls) soar in many boroughs on football match days when the local team win but not when they lose. This makes me wonder if, with some modification, the urban legend might actually have some merit. The missing link in the logical chain could be testosterone levels in male fans.

More testosterone is clearly linked to an increased tendency to violence. Sporting events are often characterized as mixing bowls of the stuff, but that stereotype hides a more complex reality: Only men rooting for the winning team get a boost of testosterone. If the outcome of the game is unknown until the final minutes, this boost is massive and sudden. Meanwhile, those men rooting for the losing side, whether in sports or in an election, experience a drop in testosterone. And when men’s testosterone drops, they become more submissive and less aggressive.

This implies that a sporting event, particularly a closely contested one, could appear either to increase or decrease male violence, depending on which set of fans were examined. When I google news reports about sports and increased domestic violence (there are many…the original urban legend continues unabated), they provide what I consider uninformative data, namely someone noting that violence is believed to have gone up in place X after big game Y. But what the news accounts don’t tend to do is examine whether violence went down in place Z after the same match.

I would love to see a careful study of this issue, focusing on two cities which are sports rivals and play each other many times a year. Time-series data on domestic violence reports, arrests for violence and other critical variables (e.g., beer sales) would be required for both cities through a few years of wins and losses for both cities’ teams. The hypothesis would be that even though there is no net effect of sporting events on male violence (i.e., the original urban legend remains a legend), violence goes up in the winning team’s city and down in the losing team’s city. If that was your dissertation, please post the results here. If that’s the dissertation you are planning, I hope I can be on your committee.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College Lonon. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over ten thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

2 thoughts on “Is There a Grain of Truth in the Super Bowl and Domestic Violence Urban Legend?”

  1. Speaking of grain what about alcohol ? It is roughly as associated with violence as testosterone. I’ve been known to drink a bit while watching the game (I haven’t personally engaged in actual physical violence since I turned 10, but I suspect that the radio-immuno assay was developed to measure my personal testosterone level).

    Grape too for the more upscale violent thug.

  2. Robert: Yes indeed, that’s why I said beer sales should be in the model. You could add wine and hard liquor also if those data were available.

Comments are closed.