I can't fault the math, as presented, but the validity of this approach is based upon more than just the formulaic assumptions used to create the models. The goals of this research are also based upon assumptions, the key one being that gun availability is the major driver of gun homicides. Another assumption is that minimizing gun homicides should take precedence over success thwarting of major crimes, e.g., rape, felony assault, armed robbery.

That gun homicides are universally a bad thing is not necessarily a valid assumption. To use an example from current events, if a man is brutally attacked and is facing the chances of severe injury or death at the hands of an unarmed attacker, is it a bad thing that he pulls a legally-obtained and carried handgun and shoots his attacker to stop the attack... even if it results in the death of a 17-year-old? In short, should we be instead looking for models that help us to minimize unjustified homicides (murders or negligent homicides) instead of justified gun homicides (self-defense)?

Other assumptions include treating the mindset/propensity of violence as a random or shared factor (not influencing the rate of unjustified gun homicides regardless of the rate of legal gun possession). There is enough data to show that this assumption is invalid; one has only to look at the per-capita homicide rate of Chicago or Detroit (where legal gun ownership is very low) versus Seattle or Portland (where legal gun ownership is very high). The major factor relating to gun-related unjustifiable homicides must be something besides gun availability.

I hope the authors of this study extend and revise their models to examine other, perhaps more influential, variables leading to unjustified gun homicides, and thus provide valuable information to guide effective policy choices. As it stand, however, the current hypothesis and the models based upon it appear tautological at best.