Ray Lewis has gained a lot of attention with his recent comment that crime would increase if there was no football in the fall:
In an interview with ESPN, Lewis suggested that the NFL lockout could cause a spike in crime. “Do this research,” Lewis said. “If we don’t have a season. Watch how much evil, which we call crime, how much crime picks up, if you take away our game. There’s nothing else to do.”
Lewis offered no evidence from prior work stoppages that crime rose when the NFL season shut down. “I don’t know how he [Lewis] can make that kind of determination without having more facts,” New York Giants defensive end Mathias Kiwanuka told CBS Sports, in response to Lewis’ comments. If the current dispute between owners and the players crept into the season, forcing the cancellation of games, the economic impact would be harsh for the stadium workers and local retail establishments who serve fans on game-day. There will be significant spillover effects. But would a bored populace turn to street looting because the Steelers aren’t on TV?
Kiwanuka is in agreement with a lot of others that have suggested that Lewis’ comments were strange and unsubstantiated. But why haven’t we seen many people use data to refute Lewis’ claim? We could look at two possible scenarios that Lewis was describing:
1. The more limited scenario: in the times when football games would normally be played, crime would be higher. I remember news stories during the Chicago Bulls’ championship runs in the 1990s about the decrease in ambulance calls and emergency room visits during games and I recall seeing similar stories about NFL playoff games. Couldn’t police departments or other agencies quickly go through their files to see whether crime rates differ during NFL games compared to other times? (Run a comparison between Sunday afternoons of bye weeks or weeks when the team plays on Sunday, Monday, or Thursday night compared to typical Sunday afternoon game times.)
2. The broader scenario: overall, crime would be higher when there are no football games. This would take some more data work to find comparison periods in the fall/early winter when there is no football at all. How about looking at cities by year when they had a NFL playoff team compared to years that they did not? I’m sure someone could figure this out with the appropriate crime data and years of football records.