Bad science and bad statistics in the courtroom convict innocent people

Anyway, this article is spot on and this is a really rampant specific issue.

“expert witnesses have often overstated the probative value of their evidence, going far beyond what the relevant science can justify.”

This is the absolute center of the issue. Experts are used by the prosecution AND defense, but the experts used by the prosecution are often times so hilariously and openly biased that its insane how little pushback they get. Career cops with poor credentials that spend 30 years going from witness stand to witness stand saying literally whatever. You can definitely find biased 'pay for opinion' style experts for defense (its a profitable industry after all), and ive had the misfortune of trying to course correct them and their clients, but they are less frequently biased via a long career in law enforcement, and more importantly, get absolutely flambéed when they display the same level of bullshit as prosec witnesses. There are actual possible consequences for defense experts who lie, you get roasted in a news article that shows up on every google search and nobody with integrity will hire you ever again. That just doesnt happen for prosecution experts, ever. most media is terrified to highlight those bad actors because of the huge pushback they get from cops/cop unions/and more importantly the rabid facebook populace that will NOT stand for police criticism. and judges will very rarely hand down sanctions. after all, over a third (37 percent) of sitting justices (state supremes) are former prosecutors, while only 7 percent are former public defenders. (https://www.brennancenter.org/our-work/research-reports/stat...)

The article cites a few specific examples of experts creating a huge wake of bullshit in the legal field as they trudge around making stuff up, but a more convincing writeup with broader examples of fraud in the field was done by the intercept here (https://theintercept.com/2015/04/24/badforensics/) in 2015. I can assure you the problem is still bad.

Dont think its better with digital evidence just because there are hash values and log files. I have seen (insert f.agency here) plainly and clearly fabricate evidence (and get away with it). Countless instances of smart people who, definitely, beyond any doubt, know and understand what their evidence ACTUALLY indicates, and instead inflate its value to a ludicrous degree.

TLDR; If its physical evidence, and not exact DNA match(think tire marks, blood spatter, bite marks), its wouldnt remotely be accurate enough to be court admissible if the courts were functioning properly and vetting expert witnesses properly. if its digital evidence, and there are claims like "person was at location x,y,z at time a,b,c" and not in the format "the device was more likely than not in the general vicinity at the general time" then they are talking out of their ass. in absence of security footage or the like.

my bias: if it wasnt obvious, i work with the defense the majority of the time, on digital evidence, and i am going to see and remember the transgressions of the other side more often. i try to account for that. not a lawyer. so much to learn in the field still. just an internet ignoramus

If this is your genuine belief, it begs the question: why are Americans, on average, such terrible people when compared to other first world citizens?

I don't know the solution. Ideally, everyone would be good at math, but I don't see that happening in the near future. My best answer is for us to provide every court have an on-call statistician, reviewing every case. I would hope big law firms are already doing this, but you should get a mathematically-sound judgement even if you have to depend on an underfunded and overworked public defenders.

(More recently, I've been wondering this about scientific publishing. Misusing statistical tests--or choosing the wrong once seems--like a fairly common source of error in peer-reviewed science. It feels like there, too, we need expert statisticians carefully reviewing every paper of note. Given the amounts charged by some journals, it feels like a reasonable service to demand of them.)

Its not like those kinds of processes, the "enhancing" filters, are inherently bad or inadmissible, you just have to be clear with the judge and jury what the hell is actually happening.