Bad science and bad statistics in the courtroom convict innocent people • Hacker News Board

Re: Bad science and bad statistics in the courtroom convict innocent people

#91

Post by qingcharles » Mon, Dec 11, 2023, 6:56 PM UTC

s1artibartfast wrote: post #22

graphe wrote: post #7

Earlier quoted context omitted.

Innocent meaning they didn’t do the specific crime, or innocent as in someone with a clean record that was railroaded? If you’re looking for the criminal that killed a bunch of people but not that specific case it’s not that uncommon. If you’re looking for a squeaky clean guy that was there at the wrong place at the wrong time it’s very rare. Most criminals are repeat offenders.

I believe, and I think the data supports, that the US prison population is due to the number of laws we have, and the harshness of our penalties. A false conviction rate of a few percent does not explain an order of magnitude level differences in incarceration between countries. I think an interesting question is what portion of the US population would be incarcerated if the totality of our laws were applied to every…

Opinion of a criminal: I don't think there are too many criminal laws at the State level. That is definitely not the reason for high incarceration rates. The guys who are in prison didn't randomly trip over a law they didn't know existed. 99% of the laws are the same as they've been since pre-history: don't hurt someone, don't steal etc.

Sentencing to prison is probably the key problem. Criminals need help. Outright punishment just hardens them. From my experience the bulk of crime has its origin in mental illnesses which are not being treated. The medical system in the USA almost certainly is the cause of this.

Re: Bad science and bad statistics in the courtroom convict innocent people

#92

Post by claytonjy » Mon, Dec 11, 2023, 6:57 PM UTC

twisteriffic wrote: post #62

pc86 wrote: post #52

Earlier quoted context omitted.

> My best answer is for us to provide every court have an on-call statistician, reviewing every case. Is this not the exact reason for expert witnesses?

Expert witnesses can be cripplingly expensive. Having a gratis sanity check sounds like a great idea, assuming you can find some way to insulate them from influence.

Can confirm. I worked at a small statistics consultancy where the owner did some expert testimony, and you have to reveal what you charge. The opposing side was often paying 500+/hr for shoddy work!

I think we've seen stories of Ivy League professors charging thousands per hour, and making the bulk of their income this way. Not sure how high it goes in the most high-profile cases.

Re: Bad science and bad statistics in the courtroom convict innocent people

#93

Post by csa » Mon, Dec 11, 2023, 6:57 PM UTC

cool_dude85 wrote: post #77

qingcharles wrote: post #47

Earlier quoted context omitted.

OK, having spent practically 10 years inside and having helped 1000s of people with their cases, here's my single-person perspective: at least 95% of the people locked up are guilty of something related to their charges. The justice system isn't completely off-kilter. My main issues were that a) many people were overcharged (i.e. charged with extra crimes they were not guilty of) simply to obtain a plea deal; b) ofte…

1 out of 20 is still pretty nasty given the size of the US prison population.

He said “at least”.

And he specified “guilty of what they were charged with”.

You also need to factor in the people who are guilty of crime A (that they did commit but were not convicted of) but are convicted of crime B (that they did not commit) for whatever reason.

I’ve seen “street justice” like this. In some cases I think it a net good (right outcome for the wrong reasons, typically involving a serial criminal who manages to avoid the “smoking gun”), but sometimes it’s just an abuse of power that needs to be rectified.

Re: Bad science and bad statistics in the courtroom convict innocent people

#94

Post by brigade » Mon, Dec 11, 2023, 6:57 PM UTC

Lord-Jobo wrote: post #20

adamsb6 wrote: post #17

Even photo and video evidence suffers from courtrooms that don’t understand how nearly all cameras will record and encode images. JPEG encoding (and thus MPEG encoding) takes 8x8 blocks of pixels and represents them as a discrete cosine transform. In a very real way these pixels don’t actually exist. If your video evidence relies on such a small portion of MPEG encoded video, you can easily mislead a jury to convicti…

This is one of the better litmus tests to see if the expert witness is knowledgeable in the field, and more importantly, up-to-date on the tech. some really smart but definitely older expert witnesses will fumble this very hard. 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.

And conversely, it's a good litmus whether to listen to the "experts" on websites like this one. So many would discount literally any image/video evidence just because of JBIG2's pattern matching or Samsung's ML model of the moon, without bothering to think about what's actually happening in the relevant pipeline.

Re: Bad science and bad statistics in the courtroom convict innocent people

#95

Post by qingcharles » Mon, Dec 11, 2023, 6:59 PM UTC

WaitWaitWha wrote: post #46

This a bit misaligned, blame shifting. I will be the first to admit there are unscrupulous, fame seeking, and overzealous expert witnesses out there. But, there are mechanisms in the courts to filter most of them out. I just think it is not done for money, time, career, or political reasons. I have seen this fail both in the adversarial (e.g., US) and elsewhere in inquisitorial (e.g., EU) court systems. In the advers…

Most criminals can't afford expert witnesses. It is as simple as that. The court is supposed to intervene and force the State to pay for them, but it's such a nightmare it rarely happens.

I had an expert witness testify for the prosecution against me and I thought he was fantastic, very fair and very knowledgeable.

Re: Bad science and bad statistics in the courtroom convict innocent people

#96

Post by oooyay » Mon, Dec 11, 2023, 7:00 PM UTC

qingcharles wrote: post #81

chaps wrote: post #53

Very, very much yes. Criminal courts really, really aren't a good place to address technical matters. For example, ankle electronic monitoring (obviously) doesn't work well in faraday-cage like apartment complexes. And yet time and time again, GPS skew is used to argue that someone left their home -- sometimes even months after the fact. I've read emails from a local sheriff's office that said that they consider a su…

Also wrote the previous one which ended with me serving five months in jail for retweeting the public defender's Tweet about it (not your fault): https://chicagoreader.com/news-politics/false-alarms/ This might have been the Tweet I reposted: https://twitter.com/CookCoDefender/status/153970586223037644... https://nitter.net/CookCoDefender/status/1539705862230376448

Correct me if I'm off base here, but this seems to be absolute nonsense:

> Like many on EM, Ross was never granted movement outside his home even once, not even to buy food. His sister, a manager at Securitas, helped when she could, but work got in the way sometimes, he said.

So, they replaced the box you sit in with a box that you pay for and provide no additional services with all the same restrictions. What the fuck is the point of these programs? I can understand wanting to use EM to ensure someone shows up to a trial, but this geofencing stuff is absolute madness. It just reintroduces the problem that made prisons and jails problematic in the first place. If someone is a non-threat enough to let them live in society with a bracelet then let them live.

Re: Bad science and bad statistics in the courtroom convict innocent people

#97

Post by qingcharles » Mon, Dec 11, 2023, 7:01 PM UTC

rightbyte wrote: post #54

It wouldn't surprise me if computer forsensics is about as bad science as bullet forsensics. Ever since I had a porn virus hosting kinda lame porn videos over BitTorrent like 20 years ago, I thought about exactly how computer forsensics prove intent in child sexual abuse material cases.

I don't know. I've spoken to FBI computer forensics guys and their computer knowledge was up there with mine in terms of hardware, software, OS etc. I've been very, very impressed with them.

The problems come when they pass their data to the prosecution and defense who are not computer trained and it all goes sour from there on.

Re: Bad science and bad statistics in the courtroom convict innocent people

#98

Post by qingcharles » Mon, Dec 11, 2023, 7:03 PM UTC

csa wrote: post #93

cool_dude85 wrote: post #77

Earlier quoted context omitted.

1 out of 20 is still pretty nasty given the size of the US prison population.

He said “at least”. And he specified “guilty of what they were charged with”. You also need to factor in the people who are guilty of crime A (that they did commit but were not convicted of) but are convicted of crime B (that they did not commit) for whatever reason. I’ve seen “street justice” like this. In some cases I think it a net good (right outcome for the wrong reasons, typically involving a serial criminal wh…

A lot of people in that 5% take a plea for a crime they didn't commit because the plea is so sweet it saves them from trial on all of the charges, some of which they clearly did commit and might receive a higher sentence for than the plea deal.

The overcharging is a trick by the prosecution to secure the plea.

I see people charged with murder who plead to manslaughter when the case was manslaughter from the beginning, but who risks going to trial and being found guilty of murder? Minimum on murder is usually 20 years straight, whereas manslaughter is usually probationable.

Re: Bad science and bad statistics in the courtroom convict innocent people

#99

Post by qingcharles » Mon, Dec 11, 2023, 7:05 PM UTC

Dyac wrote: post #71

Lord-Jobo wrote: post #12

Wow, a hackernews topic thats actually directly related to my specialty. 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…

I think the geolocation stuff is particularly concerning. Expert witnesses try to use it to prove which side of a road someone was on, or what route they took from point A to point B or something. My phone (Pixel 6) regularly gets my location wrong. I have lived in the same house, with the same router and same phone for years, yet every week it randomly tags incorrect coordinates on my photos, in some cases putting t…

I got arrested every day for months because GPS said I wasn't at home on house arrest.

https://chicagoreader.com/news-politics/false-alarms/

Re: Bad science and bad statistics in the courtroom convict innocent people

#100

Post by Dalewyn » Mon, Dec 11, 2023, 7:06 PM UTC

causal wrote: post #10

Note that proportionately few of our federal judges have spent any time as public defenders, and a disproportionately high number have served as prosecutors. Jackson is the first supreme court justice to have served as a public defender, whereas 3 other justices have been prosecutors. Prosecutors are incentivized to get convictions, and that slanted thinking could easily be carried on to the bench. [1] https://www.pb…

Obligatory IANAL.

>Prosecutors are incentivized to get convictions,

Their entire job is to get convictions. It literally doesn't matter if the prosecutor thinks someone should be convicted or not, that's for the judge to decide. The job of a prosecutor is to make as strong a case for conviction as he possibly can.

Likewise, the job of a lawyer representing a defendant is to make as strong a case for acquittal as he possibly can. Whether the lawyer thinks someone should be acquitted or not literally doesn't matter.