ingliz wrote:18,518 - 780 = 17,738
Divided by 26 = 682
Divided by 19 = 36
Just under 6 times the US gun homicide number and nowhere near the 100-fold numbers you arrive at.
The problem is that you are only dividing the numerator here. That doesn't really make sense, since a rate is already an average.
ingliz wrote:If you don't want to divide by 19, you must take US gun deaths from 1993-2022 and add them together to arrive at a comparable figure.
If you wanted to do that, you would need to add up the population of each year to build the denominator. That is, you would take the total number of gun homicides in the US and divide it by the sum of the estimated population for each year in 1993-2022. I don't think it would make much of a difference to be honest.
You could measure the rates in terms of 100k person-years, but you would need to have info on how long was each person considered in the denominators deployed in Vietnam (for the soldiers) and alive during the period (for the US gun homicide rate).
ingliz wrote:I consider non-combat deaths and friendly fire on a par with suicides in the US gun death numbers. So if we are not removing non-combat deaths and friendly fire from the Vietnam numbers, we must add suicides by gun from 1993-2022 to US gun deaths.
Friendly fire is more like accidental homicide. Like, for example, killing someone else while playing with a gun or killing your partner thinking he or she was trying to forcibly enter your home. Not that friendly fire was a major driver of small arms deaths in the Vietnam war from what I've seen - that 398 figure you know - so it doesn't make much of a difference to include it or not. It seems friendly fire incidents involved mostly heavy weaponry, which isn't all that crazy if you ask me - get the target wrong for a bombing run and you can perfectly kill dozens of your own soldiers.
From what I can see, only around 1% of gun deaths (and therefore around 2% of gun homicides) in the US are accidental so I didn't remove them either.
Suicide is, well, suicide, which is clearly beyond the scope of this thread (the American tourist in Chile didn't kill himself, the claim was also clearly specifically about homicide). I think suicides don't consider accidental self-killings. Not that conclusions would drastically change, instead of a 100x ratio it would fall to 50x, meaning the Vietnam rate was still several times over the total US gun death (homicide + suicide) rate.