Hacker News new | past | comments | ask | show | jobs | submit login

The overwhelming majority isn't great when you're talking risk of death.

Someone survives playing Russian Roulette the overwhelming majority of the time too. 5 out of 6 times, nothing bad happens at all. 1 out of 6 times, you die. Is that also not something you would consider a "life threatening situation?"

It's actually an interesting question. At what point does the risk of dying in a situation make it "life threatening"? I believe something like 1 in 291 police interactions end up with someone in the hospital or dead.

https://www.eurekalert.org/pub_releases/2016-07/b-upk072116....




1/6 is literally 2 orders of magnitude smaller than 1/291. Nonsensical comparison.

More importantly, how many of those hospitalizations/deaths are justified? You realize that police are typically responding to crimes and dealing with criminals who tend to be a little more violent than the average person, right?

Don't get me wrong, I'm not a bootlicker by any means, and I acknowledge there are serious policing problems in the US. But seriously questioning whether you're going to survive a traffic stop if you act reasonably is irrational paranoia.


I do formal risk assessments as a part of my job, in a context where people are often risking their lives. Both 1:6 and 1:291 as the chance of of severe injury or death are both firmly in the highest risk category under most calculations, meaning that for decision making, you should assume that it will result in death. Yes, they are two orders of magnitude different. But the comparison is not nonsensical.


Well then you are no doubt aware that these statistics skew overwhelmingly toward those who are noncompliant, combative, and/or felons, correct? Such that the actual rate for "average" people (including minorities, although their rate may be higher) is probably miniscule.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: