Quote Originally Posted by TheSpoonald View Post

Before the yale data, you could have deported everyone and expected to save 100 lives. But now you know that exporting all the illegals would save TWICE that many people.
This is problematic for a number of reasons.

First, the assumption is made that the murder rate is equal among illegal immigrants and everyone else. Maybe it is, maybe it isn't.

But more importantly, assuming an equal murder rate, you could lower the total number of murders by reducing the population in any number of ways. E.g., deporting any number of randomly chosen people, deporting anyone over 50, deporting all legal immigrants, or (eventually) by lowering the birthrate to zero so the population dies off.

So, if you randomly deport half of the US, you will have half the number of murders. Half the murder rate!

This is why murder rates are typically reported as x per 100,000 people.

What you've also forgotten to do is incorporate the fact that the census-reported population of (let's say) 330 million doesn't count illegals. So either you have 330+11 =341m people with x number of murders, or you have 330+22 =352m people with x murders. And with the larger number, the murder rate/100k is actually lower than with the smaller number. That's not an argument in favour of illegal immigration, but it does make a difference in any arguments about how dangerous illegals may or may not be (which actually depends on their intrinsic murder rate, not the national murder rate, but w/e).