Black Swans, Tragedy, and the Limitations of Decision Making

When my twitter feed Wednesday afternoon broke the news of a another school shooting, I immediately prayed that the impact would be limited. As the news continued to seep out, it was clear that those prayers would shift to the families and friends who would be processing the loss of 17 lives.

It was less than a day before the recriminations began. There was a record of school disciplinary trouble. The police were called frequently about Cruz over the years. The FBI had a record of a YouTube comment in September of 2017. (Details on all this here, here, and here.) Students interviewed remember Cruz as a troubled child who acted in scary ways.

This retrospective recollection of isolated events give us some sense of comfort while also allowing us to blame someone for such senseless violence. If there was a trail of breadcrumbs that made this predictable, then we aren’t at the mercy of random events. And if someone should have been following the breadcrumb trail, then maybe there could have been an intervention through either mental health or criminal justice organizations.

Black SwanBut this search for causation is doomed to fail. Because an event like Parkland is a perfect example of the Black Swan problem in probability. Nassam Nicholas Taleb points out the difficulty of extremely rare events (from the Wikipedia entry defining Black Swan events)


1. The disproportionate role of high-profile, hard-to-predict, and rare events that are beyond the realm of normal expectations in history, science, finance, and technology.  2. The non-computability of the probability of the consequential rare events using scientific methods (owing to the very nature of small probabilities). 3. The psychological biases that blind people, both individually and collectively, to uncertainty and to a rare event’s massive role in historical affairs.

According to the National Center for Educational Statistics, there were just over 24,000 public high schools in the US in the fall of 2013 enrolling just under 15 million students. In 2017, there were 9 school shooting incidents with fatalities occurring in 5 of those (a total of 15 killed). The probability of having a shooting in a given school was just under .04%. The probability of being killed was literally 1 in a million.

That is not to throw up our hands and do nothing. It is rather to point out the folly of thinking we can predict any incident. It is true in retrospect that Cruz abused animals and that police were called to his house and that he posted a random comment five months back on a YouTube page. Let me illustrate with the most mundane example.

School Shooter
Yes No
Torments Squirrels Yes

It is true that the school shooter is a Yes-Yes in my table above. But there are lots of boys who torment animals and never go on to threaten their peers. And it’s logical to argue that some of those 9 school shooters in 2017 loved all sorts of animals.

Taleb’s work influenced Daniel Kanneman’s work on how we make decisions. Kanneman finds (as summarized in Michael Lewis’ The Undoing Project) that the decision rules that professionals use seem to work but are less accurate the what an algorithm would predict. It is simply not possible for the human brain to consider all of the potential variables impacting a decision.

It is unreasonable, therefore, to expect the FBI to take a random comment from 2017 and use it to pinpoint a high-risk situation. It is unreasonable to blame school officials for not doing more to see their disciplinary situation as something more than a policy decision like those made daily.

For that matter, our search for less tangible causes is equally flawed. Politicians have referred to Nicholas Cruz as “evil”, on the reasonable assumption that normal people don’t shoot their classmates.

Those who blame cultural changes (lack of respect for authority, Hollywood, athletes, video games, hip-hop music) for school shootings are no different from those looking at the abuse of animals. Such claims of correlation are countered by the millions and millions of young people who experience the same cultural changes and don’t respond violently. (These same cultural critics were praising the faith commitments of Eagles players two weeks ago.)

In fact, the reactions of those very young people — who grew up with a sense of voice, access to social media, and a remarkable aversion to BS — who may be the most positive way forward out of this horrible situation. They are dismissing easy or trite answers and calling for honest engagement of the issues of school violence.

They may not know the probabilities but they want steps put in place that will keep them safer. That may involved discussions about automatic weapons or building security. It may involve actual changes in high school culture that makes it less likely that the angry social isolate remains cut off and feeling victimized.

So the media, the politicians and the religious leaders can tie themselves in knots looking for someone to blame. I’m putting my hopes in the kinds of kids who will call out Tomi Lahren on a daily basis.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s