[easp]

You are relying on anecdotes. However, given that shooters of this sort are quite rare, is there even enough data to reach statistically significant conclusions about the risks from people returning fire?

In any case, if we were to follow this argument further, we would expect that many more citizens would be armed, which would likely change things. Given the inherent confusion of such a situation, I would expect we would start seeing armed citizens accidentally shooting other armed citizens because they misread the situation.

Even trained, battle seasoned soldiers make mistakes.