For the record, if you graph this table of access to mental health care (using the 2014 values):
To this graph (using the 2016 values):
Then the best fit line of those x,y coordinates says that the best state for mental care would be predicted to have half the murders of the state with the worst mental care. The average state would only have a reduction in homicide by about 1/4th.
The national homicide rate is 5.3 per 100,000, meaning that about 16,000 people were killed in 2016. With equivalent mental health care as the best state spread to every other state, we would expect that number to reduce to 12,000. 4,000 lives would be saved.
It should be noted, though, that even though the graph does look decent and a visual inspection leads one to trust that this prediction isn’t complete baloney, you would really want to throw in average income by state and/or gini coefficient by state as a third variable before making any real predictions on the data. (But I’m feeling lazy today.)
The total number of gun homicides in 2014 (according to the Wikipedia) was 8,124.
Now if a gun ban would reduce gun homicide, then the 8,124 gun homicides would go away.
However, gun homicide is a fake statistic. It’s of no value to reduce gun homicides if it doesn’t impact homicides. The only positive links that I have found (positive meaning “more guns = more deaths”) between gun ownership rates and homicide are in 3rd world countries. In the US, gun ownership rates are negatively correlated to homicide rates and on average across the globe gun ownership is generally correlated to lower homicide rates.
Basically, banning guns doesn’t save lives. At best it makes people kill each other with other weapons - bombs, knives, trucks, etc. At worst, it encourages criminals to use force against the innocent (or whatever real-world factor it is that is causing the negative correlation).
So the option isn’t 8,124 lives versus 4000. It’s 0 or 4000.
Also, lithium in the water supply would save lives. It’ll prevent suicides as well.