[tlrobinson]

I don't understand the downvotes. If we're talking about tourists doesn't it make sense to look at... tourists?

EDIT: to be clear, my hypothesis was the murder rate of tourists was lower than the overall murder rate, which could plausibly make the "You’re 100x more likely to die in a car accident" statement correct. See hsitz's comment and my response. 100x is likely an exaggeration but you still are probably "more likely to die in a car accident on your commute tomorrow than by a cartel member in Mexico as a tourist".

[hsitz]

Mexico has somewhere around 18 million tourists per year, about half of whom are Americans [1]. I can't remember any being murdered, but I suppose a handful have. U.S. media loves to publicize these kinds of things so I'm pretty sure I'd be well aware if it were more than a handful, certainly if it was more than a handful being killed by cartels, who commit most of the murders in Mexico.

So say, 10 tourists get murdered in Mexico each year. That's (very roughly) 10 per 18 million, or 1 per 1.8 million, or 0.06 per 100k.

BTW, adding "per year" to the metric would be misleading, add nothing, because it falls out of the unit analysis in the calculation, which is done based on

                      10 tourists      10 tourists
     1 year            murdered         murdered
 ---------------  x -------------- =   ---------------
 18 mil tourists        1 year         18 mil tourists

This is not that helpful, because tourists stay just a short time, not a full year like local residents. If you want to somehow compare this to a metric of inhabitants murdered per 100k per year then you probably want to do more. Assume the average tourist stays 1 week. Then we need to multiply by 52 to get the number of tourists murdered per year of tourist time in Mexico, which yields result of around 3 (0.06 x 52) per 100k of tourist-years. This is significantly less than the U.S. average of 5 murders/100k per year. So it could be that U.S. tourists are safer in Mexico than they are at home. ;)

[1] https://en.wikipedia.org/wiki/Tourism_in_Mexico#Statistics

[tlrobinson]

Right, I was suggesting the murder rate for tourists might be lower than the overall murder rate, which could make the "You’re 100x more likely to die in a car accident on your commute tomorrow than by a cartel member in Mexico as a tourist" statement correct.

0.06 is indeed 100x less than 11, but to compare to the car crash death rate above you'd need to estimate the amount of time tourists spend in the country per year. If you say 1 week per year that brings it up to 3.12 per 100k per year, so still less than the risk of dying in a car accident (in the US), but not 100x less.

Also, technically the statement was "by a cartel member", so you'd have to know how many of those murders were by cartel members.

[hsitz]

Exactly, I think I was editing to add that as you wrote this. Also, my guesstimate of 10 tourist murders/year may be way off. I'm guessing it's lower, certainly if only by cartel members, but not sure.