5 ms·

Once you are an expert at any topic, you can figure many outcomes from very small amounts of information. That does not mean the outcome was biased or flawed e

Once you are an expert at any topic, you can figure many outcomes from very small amounts of information. That does not mean the outcome was biased or flawed etc. Is it difficult to believe that a judge would be able to predict at least some outcomes from a single paragraph? All it takes is to predict above random chance to have a statistically significant effect.

mattxxx 2y ago
> Is it difficult to believe that a judge would be able to predict at least some outcomes from a single paragraph? It's difficult to believe that the decision should be made quickly. When making decisions about the trajectory of someone's life, they should be made with care. Having a system that removes snap judgements and bias is important, and I would say that even a quick-scan and re-ordering of documents is a form of bias.
- 1992spacemovie 2y ago
  In general I agree with your attitude - however my sympathy is limited by my “lived experience” of interacting with those in the criminal justice system. Most of them belong right where they are - regardless of what the brain dead politico class has embraced.
  - benj111 2y ago
    So how far do you want to take that. If the judges lived experience of suspects is that they're normally guilty.... Having interacted with the criminal justice system, that's the view they seem to take.
    - 1992spacemovie 2y ago
      I will say two things: 1) Most people charged with a crime these day are usually guilty of it. The public has made it pretty clear they would rather see no one charged if the crime is a legit whodunit -- no one desires to see innocent people arrested and charged just to give the impression of safety. Prosecutors, and downwards (police) feel pressure to only file and prosecute cases that are legitimately believed to be of truth regarding the suspect. That said, you are innocent until proven guilty. The burden of proof is on the state. Even if we can mostly feel that you did it, if it can not be put before a jury to return a guilty verdict, you are not to face penalty in our society (mostly; ignore OJ's civil case... that's a pretty rare exception to be honest). 2) I understand why judges seem automatic -- but remember this: in a criminal trial (minus your option to do a bench trial), you are found guilty or not-guilty by a jury of your (location) peers. The judge has his own opinion, but he is restricted based on the guilty/not-guilty finding of the jury, and sometimes state/federal sentencing guidelines. Of course there are prick judges who SENTENCEMAXXXX people just to be an asshole - no one with an ounce of common sense will endorse that. In general, I trust and expect judges to apply their experience of both law and life in determining what is the appropriate penalty after someone is found guilty.
      - benj111 2y ago
        Well my experience is a solicitor pleading guilty on my behalf which isn't generally allowed in this jurisdiction. So no trial, no presumption of innocence, and absolutely no interest in anything other that processing 'criminals' as quickly as possible. So your comment "Most people charged with a crime these day are usually guilty of it" May well be true. the issue is that it is self reinforcing, most people are guilty, so the system treats you as probably guilty, so the people that aren't guilty don't get the protections theyre supposed to have the right to.
NiloCK 2y ago
This reads almost verbatim as my conception of bias. "Once you've abandoned principled, wholistic reasoning for your pet heuristics, you can figure out many outcomes from the inputs to your pet heuristics". Do you mean something different by "bias"?
- DataDive 2y ago
  Sorry to say, but I think you have put the wagon ahead of the horse. Being able to predict an outcome has nothing to do with the process of deriving the outcome, and it has nothing to do with bias. There may be many signals that correlate with an outcome. If you are out of shape and move ploddingly, you probably can't do a triple axel even though you believe you can. Is that prediction biased? Bias would be if you could demonstrate that predicting the outcome has influenced their decision-making.
  - intended 2y ago
    Yes? To expand on the point using your analogy - Your point is that there would be some things an expert would see, and then judge to be highly improbable. While this point can inform our thinking, it is the lesser point that exists within a bigger issue: First - This is a court of law, not the court of public opinion, or processes. There is an expectation of exactness, and of a fair, unbiased and attentive hearing of the facts. Second- While I don’t know what a triple axel is, I have seen people who seem utterly out of shape dance with grace, and people who appear to be incredibly fit, turn out to be frauds. In this scenario, I would say that the assumption that each case is similar, is not valid. I will grant that it becomes human to behave this way though.
- swid 2y ago
  If I am testing something I believe works and is ready, prior to testing, I will tell you it is almost ready. Then if we test it and it fails acceptance testing, I might learn there is a problem that takes some time to fix. I did not arrive at a biased decision; I had priors that I used to make an estimation, which turned out to misleading. The judge is exactly this case. They guess the time it will take to rule; it doesn’t have to mean their eventual ruling is biased.
tikhonj 2y ago
Developing that sort of expertise requires getting clear and—ideally—timely feedback on the quality of your decisions. Do parole judges get that? I'm sure they get feedback on whether their decisions are consistent with what other judges would have decided, but that's qualitatively different from feedback on whether those decisions were fair or right. If anything, that is the kind of feedback that would propagate biases in the system! You would end up becoming an expert on making consistent, defensible decisions, even if those decisions were consistently and defensibly bad.
- thaumasiotes 2y ago
  > Developing that sort of expertise requires getting clear and—ideally—timely feedback on the quality of your decisions. Do parole judges get that? The expertise in question is predicting, from a short summary of the case, what's going to happen in the trial itself. So yes, every judge gets clear and timely feedback on prediction quality.

mattxxx 2y ago

> Is it difficult to believe that a judge would be able to predict at least some outcomes from a single paragraph? It's difficult to believe that the decision should be made quickly. When making decisions about the trajectory of someone's life, they should be made with care. Having a system that removes snap judgements and bias is important, and I would say that even a quick-scan and re-ordering of documents is a form of bias.

1992spacemovie 2y ago
In general I agree with your attitude - however my sympathy is limited by my “lived experience” of interacting with those in the criminal justice system. Most of them belong right where they are - regardless of what the brain dead politico class has embraced.
- benj111 2y ago
  So how far do you want to take that. If the judges lived experience of suspects is that they're normally guilty.... Having interacted with the criminal justice system, that's the view they seem to take.
  - 1992spacemovie 2y ago
    I will say two things: 1) Most people charged with a crime these day are usually guilty of it. The public has made it pretty clear they would rather see no one charged if the crime is a legit whodunit -- no one desires to see innocent people arrested and charged just to give the impression of safety. Prosecutors, and downwards (police) feel pressure to only file and prosecute cases that are legitimately believed to be of truth regarding the suspect. That said, you are innocent until proven guilty. The burden of proof is on the state. Even if we can mostly feel that you did it, if it can not be put before a jury to return a guilty verdict, you are not to face penalty in our society (mostly; ignore OJ's civil case... that's a pretty rare exception to be honest). 2) I understand why judges seem automatic -- but remember this: in a criminal trial (minus your option to do a bench trial), you are found guilty or not-guilty by a jury of your (location) peers. The judge has his own opinion, but he is restricted based on the guilty/not-guilty finding of the jury, and sometimes state/federal sentencing guidelines. Of course there are prick judges who SENTENCEMAXXXX people just to be an asshole - no one with an ounce of common sense will endorse that. In general, I trust and expect judges to apply their experience of both law and life in determining what is the appropriate penalty after someone is found guilty.
    - benj111 2y ago
      Well my experience is a solicitor pleading guilty on my behalf which isn't generally allowed in this jurisdiction. So no trial, no presumption of innocence, and absolutely no interest in anything other that processing 'criminals' as quickly as possible. So your comment "Most people charged with a crime these day are usually guilty of it" May well be true. the issue is that it is self reinforcing, most people are guilty, so the system treats you as probably guilty, so the people that aren't guilty don't get the protections theyre supposed to have the right to.

NiloCK 2y ago

This reads almost verbatim as my conception of bias. "Once you've abandoned principled, wholistic reasoning for your pet heuristics, you can figure out many outcomes from the inputs to your pet heuristics". Do you mean something different by "bias"?

DataDive 2y ago
Sorry to say, but I think you have put the wagon ahead of the horse. Being able to predict an outcome has nothing to do with the process of deriving the outcome, and it has nothing to do with bias. There may be many signals that correlate with an outcome. If you are out of shape and move ploddingly, you probably can't do a triple axel even though you believe you can. Is that prediction biased? Bias would be if you could demonstrate that predicting the outcome has influenced their decision-making.
- intended 2y ago
  Yes? To expand on the point using your analogy - Your point is that there would be some things an expert would see, and then judge to be highly improbable. While this point can inform our thinking, it is the lesser point that exists within a bigger issue: First - This is a court of law, not the court of public opinion, or processes. There is an expectation of exactness, and of a fair, unbiased and attentive hearing of the facts. Second- While I don’t know what a triple axel is, I have seen people who seem utterly out of shape dance with grace, and people who appear to be incredibly fit, turn out to be frauds. In this scenario, I would say that the assumption that each case is similar, is not valid. I will grant that it becomes human to behave this way though.
swid 2y ago
If I am testing something I believe works and is ready, prior to testing, I will tell you it is almost ready. Then if we test it and it fails acceptance testing, I might learn there is a problem that takes some time to fix. I did not arrive at a biased decision; I had priors that I used to make an estimation, which turned out to misleading. The judge is exactly this case. They guess the time it will take to rule; it doesn’t have to mean their eventual ruling is biased.

tikhonj 2y ago

Developing that sort of expertise requires getting clear and—ideally—timely feedback on the quality of your decisions. Do parole judges get that? I'm sure they get feedback on whether their decisions are consistent with what other judges would have decided, but that's qualitatively different from feedback on whether those decisions were fair or right. If anything, that is the kind of feedback that would propagate biases in the system! You would end up becoming an expert on making consistent, defensible decisions, even if those decisions were consistently and defensibly bad.

thaumasiotes 2y ago
> Developing that sort of expertise requires getting clear and—ideally—timely feedback on the quality of your decisions. Do parole judges get that? The expertise in question is predicting, from a short summary of the case, what's going to happen in the trial itself. So yes, every judge gets clear and timely feedback on prediction quality.