Tuesday, December 18, 2012

Simulating the effectiveness of guns

As far back as I can remember, any time there is a mass shooting, there are always these two generic responses, among many others: "gun control/ban would have prevented it" and "more guns would have prevented it."  The general logic of each is that one tries to stop the shooter from ever becoming a shooter and the other tries to stop the shooter once he's started.  Of course no law has the total intended effect, so the first idea would in practice act as a reducer, not eliminator.  And the second, well that's what this post is about.

Why not run some simulations of the effect of particular staff having weapons available?

Make someone the shooter.  Make a bunch of people students or whoever is the victim in the particular incident.  Make a few people the armed but otherwise normal members.  The shooter tries to maximize body count or reach some goal and escape, whatever was the goal at a particular incident.  The victims try to survive.  The armed members try to keep victims alive.

On the most basic level, it would be just a custom first-person shooter level.  If we want to make it more advanced, go all the way up to actual training exercises used by military and police, though obviously the first is a lot cheaper.

Run multiple trials.  That's key: get a sample of more than one, so we can see not just how a particular incident turns out, but how they turn out in general.  Find the trend.  See what the average is, just as we'd do in a scientific experiment.  That's the goal, to see whether adding guns helps, harms, or has no overall effect, in a manner similar to a drug trial.

 I would not make this open to the general public.  Making it a training ground for shooters would defeat the purpose.

No comments: