Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It’s atypically sloppy of Graham to throw out the 4% number without mentioning that it’s just one estimate of a hard to ascertain number. One survey puts it much lower: https://dc.law.utah.edu/cgi/viewcontent.cgi?article=1130&con...

> This Article challenges the seemingly developing conventional wisdom that the error rate in America’s criminal justice system is 1% or even higher. In fact, looking at the best available and current data, a conservative estimate of the error rate is somewhere close to the 0.027% posited by Justice Scalia.



If I remember it correctly, it's the 4% that were supposed to be the conservative estimate.

Anyone who says the rate is 0.027% is obviously bonkers, given what sample you'd need to specify the result with a 0.0005% error. I don't see how anyone could practically do that.




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

Search: