“…In effect, the fatalities would be drawn from an urn and, for a large sample, would tend to be similar to the universe of all gunshot cases. But in reality we expect some differentiation among cases, certainly with respect to such immediate factors as the number of shots fired, the caliber of the gun, the skill of the shooter, whether the shooter was in control of the situation, and so forth (Braga & Cook, ). In that sense, the average ex ante probability of the fatal cases would be higher than the average for the nonfatal cases.…”