On April 29th, Josh Hancock, a St. Louis Cardinals middle reliever, died in a car crash. Hancock was drunk--his blood-alcohol content was almost twice the legal limit--speeding and talking on his cell phone at the time of the crash.
time of the crash.
Since the revelation of Hancock's condition prior to the accident, a wildfire of introspection regarding alcohol and players has swept across the big leagues. The New York Yankees recently revised their policy on beer in the visiting clubhouse, banning it. The team already prohibited beer in the home clubhouse. Conversely, the Baltimore Orioles banned beer in the home club house "temporarily" but still decided to allow it in the visiting clubhouse. The Oakland A's had already banned beer. The Tampa Bay Devil Rays may be next. Other clubs have varying policies but most, it has been reported, are reviewing those policies. The trend has a self-reinforcing momentum--as more clubs revise their policies and are celebrated for the move, still more clubs to follow the lead.
In every story about such policies, including ones on the Orioles, the Kansas City Royals, the Chicago Cubs, the Boston Red Sox, and, almost laughably, the Milwaukee Brewers, Hancock is mentioned explicitly, as in, "In the wake of the Josh Hancock tragedy..."
Put aside for now the irony of St. Louis, a stadium formerly owned by and named after a family that runs a beer company, banning beer in the clubhouse. Leave out for the time being the blatantly mixed message this sends to fans who sit through seven-or-so innings of beer service surrounded by signs advertising Anheuser-Busch's beers (Bud and Bud Light are the "official Beer Sponsors of Major League Baseball"). Never mind the fans who go to sports bars to drink beer and watch the game on TV and the beer commercials between innings.
Forget all that. Let's just look at these clubhouse policies from a risk management perspective. Let's just ask if banning alcohol in the clubhouse is the proper response "in the wake of the Josh Hancock tragedy." Does it significantly reduce risk and improve employee safety?
Banning beer in clubhouses follows a common theme in risk management. Sometimes, when emotions are raw, our brains want us to feel safe as much as they want us to be safe. Socially, we also feel compelled to match the grief or loss we feel with an equal measure of response. There's a notion that it would be heartless, perhaps downright negligent, not to do something to prevent another such tragedy.
But the problem with responding emotionally is that it tends to yield knee-jerk policies that overcompensate and, from a risk management point of view, carry high costs with limited real benefit. The classic example here is banning sharp objects and liquids on planes as a deterrent to terrorism. Experts will tell you that those policies are expensive to implement and actually do little to reduce risk, even if they make people feel better about flying. For more on this, read "Sharp Object Lessons" and listen to this podcast with Schneier.
What's more, these emotional responses also tend to be amplified by a public relations effect. That is, companies are compelled by media coverage to demonstrate they're doing something, because it would seem callous to the public (read: potential customers) not to react. So one team says something, then other teams don't want to appear to be less thoughtful or caring or responsive to the tragedy. So they follow suit, and you have the domino effect you currently see in baseball. The emotion, the necessity of public relations, media coverage and fans' expectations, they all conspire to create a response that's disproportionate to the risk.
This is understandable in a way. I mean a team could come out after the death of a player and explain that, according to their analysis, the risk of bad outcomes from drinking in the clubhouse is, in fact, relatively rare. They could point out that there are 2,430 games in baseball's regular season and an average of about 35 people per clubhouse (including coaches, staff, September call-ups and roster turnover) per game. That's 70 clubhouse uses per game, or a total of 170,100 clubhouse uses over the course of a season. The team could point out that even if 170 people were involved bad incidents (drunk driving arrests, accidents, death) resulting from drinking in the clubhouse--and 170 seems rather high--that's still only one-tenth of one percent, or .001, of all clubhouse uses that result in bad outcomes, so banning alcohol despite a 99.9 percent non-incident rate might seem a bit drastic.
But what team is going to do that kind of cold calculation in the wake of a tragedy? After a death in the family? None. You can understand how the emotional response emerges. What you can't understand, what's so utterly perplexing about this case, though, is this: Josh Hancock was not drinking in the clubhouse. He was at a bar. It's beyond puzzling to hear teams constantly invoke Hancock's name when announcing the bans, as if by taking the cooler out of the clubhouse, they've addressed the tragedy and are decreasing the chances that it will happen again. If beer were not available in St. Louis' club house the night of Hancock's death, it wouldn't have changed anything. In fact, the ban on beer in the clubhouse not only doesn't decrease the risk of another tragedy; it may even increase it.
Here's how. As Bruce Schneier and others have pointed out, policy enforcement doesn't necessarily decrease risk, it just moves it somewhere else. Schneier's classic example is adding police patrols to street corners with a crime problem. You can suppress the crime at that spot, but all that does is send the criminals to a spot with less enforcement, and the overall crime rate stays steady. In this case, you ban beer in the clubhouse and the behavior, drinking, just moves to to where alcohol is available, like where Hancock was, a bar.
If anything, banning beer impels players to a bar. And when you move the drinking out of an environment the employer can control, you may increase the risk of bad outcomes. In a clubhouse, the team could create controls: limit the amount of beer served. Monitor how drunk players are and provide rides home or cabs to players who need it. Once they leave that controlled environment though, those risk mitigation techniques are reduced or lost.
Someone thinking deeply about risk might argue that teams shouldn't ban beer in the clubhouse, rather they should encourage players who are going to drink to hang around and drink in the clubhouse where controls are available. But this is a tricky argument to make, as anyone who's debated abstinence-only versus safe sex education will tell you. It becomes moral analysis rather than risk analysis. Some will argue that providing alcohol, like providing condoms, encourages a certain behavior that shouldn't be encouraged. Others argue that people will engage in the behavior regardless of what you say or do, so you might as well control it, make it as safe as possible.
This last train of thought comes from former president of the Society of Risk Analysis Baruch Fischhoff, a professor at Carnegie Mellon University. "What might be happening [by banning alcohol in clubhouses], like with abstinence advocacy, is you're making a statement about changing societal values," Fischhoff said. "You're creating a climate of opinion. Same with banning smoking. Only sometimes this can backfire."
But Fischhoff wasn't really buying any of that as an explanation for this particular risk problem. Banning alcohol in clubhouses probably isn't simply a case of disproportionate emotional response. Maybe it's not just a cosmetic attempt to give the appearance of risk reduction while not actually reducing risk meaningfully. Maybe it's not a conscious choice to establish a different social climate. "It could be an agent problem," Fischhoff suggested. "You're assuming these moves are done on behalf of the players." What if the baseball clubs are, in fact, acting on their own behalf? Then, by banning alcohol, they've offloaded risk. To the player. The bar. The driving public. "If the problem is protecting the players, that's one thing," he continued cooly. "If the problem is protecting the club from liability and reputation damage, then it's probably a great decision to ban alcohol in the clubhouse. They're no longer liable."
Think of the head of security at an airport, Fischhoff said. His job is to mitigate risks for the flying public. It's very likely that when he reduces the risk of terrorist attacks at an airport or on an airplane that those risks don't disappear, they, like street-corner crime, move elsewhere. What happened after airport security tightened up post-9/11? Buses and trains were attacked. But that's not the airport security director's problem. "That's someone else's problem," Fischhoff points out.
"In the wake of the Josh Hancock tragedy," I assumed teams were making policy decisions that would change (and presumably reduce) the risk of another such tragedy. In fact, it's more likely that the ban on beer in clubhouses is a policy decision that will change (and definitely reduce) the risk of the team being held liable for such a tragedy.
But that doesn't mean such a tragedy is less likely to occur. And it might mean it's more likely to occur.