# Huerta88 - your handle hurts you **URL:** \ **Category:** The BBQ Pit\ **Created:** [May 26, 2006, 5:52pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344 "2006-05-26T17:52:23Z")\ **Posts on this page:** 10\ **Page:** 27
**Author:** ![EllisDee](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/ellisdee/32/4531_2.png) [@EllisDee](https://boards.straightdope.com/u/EllisDee)\ **Post date:** [June 8, 2006, 10:29pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/521 "2006-06-08T22:29:07Z")
[QUOTE=Left Hand of Dorkness] Because I’m not. And here’s why not: **this business about the numbers being destroyed afterwards is bullshit**. That’s not how our justice system works: before these three guys can be sent to prison for a long time, their guilt has to be proven **beyond a reasonable doubt** during a jury trial. [/QUOTE] It’s not bullshit, it is a way to replicate the fact that we will never know with absolute certainty what exactly happened. Our legal system is based on the idea that we _won’t know_ what _actually_ happened; that’s why the word “reasonable” is included in the phrase “beyond a reasonable doubt.” If the business about the numbers being destroyed were actually bullshit, it would be because the legal system was predicated on “beyond ALL doubt,” which it is not. ---
**Author:** ![EllisDee](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/ellisdee/32/4531_2.png) [@EllisDee](https://boards.straightdope.com/u/EllisDee)\ **Post date:** [June 8, 2006, 10:51pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/522 "2006-06-08T22:51:42Z")
[QUOTE=Ensign Edison] It’s pretty painfully clear at this point that what’s going on here is what we might call a “Monty Hall Problem Problem”, in reference to people’s inability to accept the solution to the Monty Hall Problem. [/QUOTE] Wait, what? **Malacandra** posed a probability problem that, much like the Monty Hall problem, has an unexpected / counter-intuitive answer. Our resident statistics expert answers it incorrectly, and defends her incorrect answer for multiple pages, steadfastly sticking to “it doesn’t matter how many cars are what color, the witness’s reliability is the witness’s reliability,” which perfectly mirrors a person arguing “it doesn’t matter what door I picked or what door Monty opened, there are two doors left so the odds are 50/50.” And after reading all that, you conclude that _ **Malacandra** _ is the one acting like a person who can’t understand the answer to the Monty Hall problem? As a side note, I’ve seen the red/maroon car probability problem multiple times over the years, so it’s driving me crazy that I can’t find an example of it with google. Can anyone else? ---
**Author:** ![Left\_Hand\_of\_Dorkness](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/left_hand_of_dorkness/32/7156_2.png) [@Left\_Hand\_of\_Dorkness](https://boards.straightdope.com/u/Left_Hand_of_Dorkness)\ **Post date:** [June 8, 2006, 11:04pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/523 "2006-06-08T23:04:14Z")
[QUOTE=Ellis Dee] It’s not bullshit, it is a way to replicate the fact that we will never know with absolute certainty what exactly happened. [/quote] Yes, but your scenario implies that investigation won’t get us closer to the truth, that all we can do is look at percentages and thereby make our judgement call. That’s not how our legal system works. If you can’t come up with something better than statistics, you can’t convict someone on statistics. That’s why you investigate all credible claims, even if they’re claims of a rare event. And that’s why your analogy falls apart: it ignores the ability to investigate to get closer to the truth. Daniel ---
**Author:** ![EllisDee](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/ellisdee/32/4531_2.png) [@EllisDee](https://boards.straightdope.com/u/EllisDee)\ **Post date:** [June 8, 2006, 11:40pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/524 "2006-06-08T23:40:32Z")
[QUOTE=Left Hand of Dorkness] Yes, but your scenario implies that investigation won’t get us closer to the truth, that all we can do is look at percentages and thereby make our judgement call. That’s not how our legal system works. If you can’t come up with something better than statistics, you can’t convict someone on statistics. That’s why you investigate all credible claims, even if they’re claims of a rare event. And that’s why your analogy falls apart: it ignores the ability to investigate to get closer to the truth. [/QUOTE] True, that is where my analogy falls apart. But you seem to be under the impression that anybody ever advocated that the DoJ statistics should be enough to dismiss a claim or provide proof for conviction. Nobody ever advocated that. In fact, nobody ever advocated that such statistics should even be used at trial. So I’m not sure your analogy holds up any better than mine. ---
**Author:** ![you\_with\_the\_face](https://avatars.discourse-cdn.com/v4/letter/y/ecccb3/32.png) [@you\_with\_the\_face](https://boards.straightdope.com/u/you_with_the_face)\ **Post date:** [June 9, 2006, 1:33am UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/525 "2006-06-09T01:33:01Z")
[QUOTE=Ellis Dee] Our resident statistics expert answers it incorrectly, and defends her incorrect answer for multiple pages, steadfastly sticking to “it doesn’t matter how many cars are what color, the witness’s reliability is the witness’s reliability,” which perfectly mirrors a person arguing “it doesn’t matter what door I picked or what door Monty opened, there are two doors left so the odds are 50/50.” [/quote] If I’m wrong and you are so confident that I’m wrong, then surely you can explain to the audience _how_ I’m wrong. Where does my logic fall to pieces, oh wise smart one? Can’t explain it? Thought not. ---
**Author:** ![you\_with\_the\_face](https://avatars.discourse-cdn.com/v4/letter/y/ecccb3/32.png) [@you\_with\_the\_face](https://boards.straightdope.com/u/you_with_the_face)\ **Post date:** [June 9, 2006, 1:38am UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/526 "2006-06-09T01:38:25Z")
[QUOTE=Ellis Dee] True, that is where my analogy falls apart. But you seem to be under the impression that anybody ever advocated that the DoJ statistics should be enough to dismiss a claim or provide proof for conviction. [/quote] Friend **Huerta** suggested that “probabilistic models” should be used to determine which cases get investigated which do not. **LHoD** is not arguing a strawman. The whole issue that got this thread started was the claim that we should use “fact patterns” as a way of triaging allegations and deciding where to shuffle prosecutorial resources. Oh yeah, and as a way of assigning credibility. I also noticed you didn’t answer his question. Wouldn’t you want to verify a claim that a 4 had been drawn just as much as you’d want to verify a 17? Why would you blindly accept one and not the other? ---
**Author:** ![you\_with\_the\_face](https://avatars.discourse-cdn.com/v4/letter/y/ecccb3/32.png) [@you\_with\_the\_face](https://boards.straightdope.com/u/you_with_the_face)\ **Post date:** [June 9, 2006, 4:13am UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/527 "2006-06-09T04:13:11Z")
[QUOTE=Ellis Dee] As a side note, I’ve seen the red/maroon car probability problem multiple times over the years, so it’s driving me crazy that I can’t find an example of it with google. Can anyone else? [/QUOTE] I did a little [digging](http://www.maa.org/devlin/devlin_july.html) and found the word problem authored by Amos Tversky which reads thusly > [@](#): > > Imagine you are a member of a jury judging a hit-and-run driving case. A taxi hit a pedestrian one night and fled the scene. **The entire case against the taxi company rests on the evidence of one witness** , an elderly man who saw the accident from his window some distance away. He says that he saw the pedestrian struck by a blue taxi. In trying to establish her case, the lawyer for the injured pedestrian establishes the following facts: > \*\*There are only two taxi companies in town, ‘Blue Cabs’ and ‘Black Cabs’. On the night in question, 85% of all taxis on the road were black and 15% were blue. \*\* > > The witness has undergone an extensive vision test under conditions similar to those on the night in question, and has demonstrated that he can successfully distinguish a blue taxi from a black taxi 80% of the time. > If you were on the jury, how would you decide? (bolding mine) Very similar to **Malancandra’s** right? Interesting things to note about Tversky’s word problem. 1. Two suspects have been identified in the hypothetical. Blue Cab Company and Black Cab Company. 2. Identification of these two suspects establishes the pool of cars that we are concerned about. The “at-risk” population is “blue cabs and black cabs on the road that night”. Not all cars registered in the city, state, country, and world. 3. Using the “at-risk” population composition, we can establish the probability that the witness was right. **Malancadra’s** hypothetical failed to identify who the suspects were and therefore the “at-risk” population. Without that information, you can’t really answer the question with a sound answer. The main reason why it is illogical to treat the total population as the pool of suspects is because then there would be no real reason why our population should be limited to just the city since borders can be crossed. Limiting the population to just the city would be purely arbitrary, but extending it to the whole world just looks insane. That alone tells me that there has to be a specific sample of cars _that has been defined by the incident itself_ that we should be looking at, not the whole population of cars everywhere. The car in Peking, China has a far less likely chance of being the culprit as the car that was spotted driving by the victim 5 seconds after the accident. Tversky’s word problem does a better job of addressing the problematic assumptions that **Malacandra’s** neglects to do. Now **Ellis Dee** if you see any weaknesses in my arguments, feel free to tackle them. I’m always open to ignorance-eradication. I’ve laid out all my arguments because I’m confident that they based on reason and not a pile of stupid assumptions. If you disagree with my reasoning, challenge me on rational grounds please. BTW, I enjoyed this probability problem, **Malacandra**. It was fun. ---
**Author:** ![Malacandra](https://avatars.discourse-cdn.com/v4/letter/m/45deac/32.png) [@Malacandra](https://boards.straightdope.com/u/Malacandra)\ **Post date:** [June 9, 2006, 11:05am UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/528 "2006-06-09T11:05:57Z")
[QUOTE=you with the face] Yeah. I don’t like illogical answers. Sue me. [/quote] It wasn’t an illogical answer. You were obliged to make up an example in which maroons were 33% prevalent to get away from the conclusion that we could not in fact put much reliance on the witness’s testimony. > [@you with the face](#): > > Because that’s a stupid assumption. If the prevalance of maroons is 2%, that means the size of the “suspect population” would have to be, at minimum, 50 cars in order for us to have a sample that comes close to matching the total population. _Nevermind the fact that we have no logical reason to assume that this sample would look like the total population in the first place._ Fifty cars is unrealistic. What utter, utter tosh. You claim to know statistics? You have a problem with the fact that the way the prevalence of maroon cars is reflected in a small sample is that it becomes _highly unlikely_ there are any maroon cars in the sample at all, and you think this in some way violates mathematics? > [@you with the face](#): > > _pointing and laughing at the willful ignorance_ _pointing and laughing back_. There, that advanced the argument no end. > [@you with the face](#): > > _sigh_ If I ask you “what’s the lotto numbers from yesterday?” And you say “Okay, I’ll look it up on the internet.” and then you report back to me and I say “I don’t believe you!” Wouldn’t you want to know why I’m skeptical? This has nothing to do with the “odds” you would have memorized something!!! Again, changing the hypothetical because the one I gave you didn’t produce the answers you wanted. > [@you with the face](#): > > No. I can’t believe you are being this dense! I, on the other hand, am by now prepared to accept pretty much any display of denseness on your part. > [@you with the face](#): > > If you were tell me that the RNG spit out 16 numbers and those numbers were 1, 2, 3, 4 ,5 … 16, I would not believe you. Even though the math shows that these numbers are just probable as any other string of 16, the fact that they _do not appear to be randomly selected_ triggers skepticism. Not mathematical probability! But you cannot have it both ways. Discussions of randomness are inseparable from discussions of mathematical probability, Doctor. (Note how I use your title sarcastically only when I find the display of ignorance unusually startling from someone of your professional standing. Mind you, from you yourself I’m prepared for pretty much anything.) By **Zoe** ’s model, all strings of numbers must be considered equally unique, and all discussion of their unlikeliness, remarkableness, un-randomness or whatever must be considered equally fatuous. Now, note! I am not opposed to throwing away **Zoe** ’s model; it is not I who have been advocating it. It’s your call. > [@you with the face](#): > > I did a little digging and found the word problem authored by Amos Tversky which reads thusly@ > > > [@Tversky](#): > > > > Imagine you are a member of a jury judging a hit-and-run driving case. A taxi hit a pedestrian one night and fled the scene. The entire case against the taxi company rests on the evidence of one witness, an elderly man who saw the accident from his window some distance away. He says that he saw the pedestrian struck by a blue taxi. In trying to establish her case, the lawyer for the injured pedestrian establishes the following facts: > > There are only two taxi companies in town, ‘Blue Cabs’ and ‘Black Cabs’. On the night in question, 85% of all taxis on the road were black and 15% were blue. > > > > The witness has undergone an extensive vision test under conditions similar to those on the night in question, and has demonstrated that he can successfully distinguish a blue taxi from a black taxi 80% of the time. > > If you were on the jury, how would you decide? > > Very similar to Malacandra’s right? Hoo-rah. Finally **ywtf** digs up something that’s mathematically similar to my problem. And, because it’s not posted by me, she can actually agree with the conclusion. > [@you with the face](#): > > Interesting things to note about Tversky’s word problem. > > 1. Two suspects have been identified in the hypothetical. Blue Cab Company and Black Cab Company. > 2. Identification of these two suspects establishes the pool of cars that we are concerned about. The “at-risk” population is “blue cabs and black cabs on the road that night”. Not all cars registered in the city, state, country, and world. > 3. Using the “at-risk” population composition, we can establish the probability that the witness was right. Same math (slightly different parameters). Different flavour text. All of a sudden you find the conclusion acceptable. > [@you with the face](#): > > Malacandra’s hypothetical failed to identify who the suspects were and therefore the “at-risk” population. Without that information, you can’t really answer the question with a sound answer. The main reason why it is illogical to treat the total population as the pool of suspects is because then there would be no real reason why our population should be limited to just the city since borders can be crossed. Limiting the population to just the city would be purely arbitrary, but extending it to the whole world just looks insane. That alone tells me that there has to be a specific sample of cars that has been defined by the incident itself that we should be looking at, not the whole population of cars everywhere. The car in Peking, China has a far less likely chance of being the culprit as the car that was spotted driving by the victim 5 seconds after the accident. Some truly amazing babbling going on here: first of all to explain why cars from out of town should be considered, and then to explain why all the cars in town should _not_ be considered. I’m back to the man-and-woman-on-desert-island story again here. **you with the face** is now cool with the notion that “85% of the taxis on the road” is sufficiently restrictive, even though there is _still_ not a whit of evidence to show how many taxis were in the vicinity of the accident location that night, or what the breakdown of colours was among those. No, it’s sufficient to know what the colour balance was of all the taxis operating that night. > [@you with the face](#): > > Tversky’s word problem does a better job of addressing the problematic assumptions that Malacandra’s neglects to do. “Tversky enters the field with the considerable advantage of not being the person who’s dared to contradict me.” > [@you with the face](#): > > Now Ellis Dee if you see any weaknesses in my arguments, feel free to tackle them. I’m always open to ignorance-eradication. _ **Mal** is carried out on a stretcher in immediate need of oxygen and some serious drugs_ > [@you with the face](#): > > I’ve laid out all my arguments because I’m confident that they based on reason and not a pile of stupid assumptions. If you disagree with my reasoning, challenge me on rational grounds please. > > BTW, I enjoyed this probability problem, Malacandra. It was fun. Good. I have some more discussion to come on the reporting of unlikely events. Stay tuned. ---
**Author:** ![Left\_Hand\_of\_Dorkness](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/left_hand_of_dorkness/32/7156_2.png) [@Left\_Hand\_of\_Dorkness](https://boards.straightdope.com/u/Left_Hand_of_Dorkness)\ **Post date:** [June 9, 2006, 11:12am UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/529 "2006-06-09T11:12:17Z")
[QUOTE=Ellis Dee] True, that is where my analogy falls apart. But you seem to be under the impression that anybody ever advocated that the DoJ statistics should be enough to dismiss a claim or provide proof for conviction. Nobody ever advocated that. In fact, nobody ever advocated that such statistics should even be used at trial. So I’m not sure your analogy holds up any better than mine. [/QUOTE] If nobody’s advocating either of those things, then what relevance do DoJ statistics have? Daniel ---
**Author:** ![Malacandra](https://avatars.discourse-cdn.com/v4/letter/m/45deac/32.png) [@Malacandra](https://boards.straightdope.com/u/Malacandra)\ **Post date:** [June 9, 2006, 12:29pm UTC](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344/530 "2006-06-09T12:29:32Z")
[QUOTE=Ensign Edison] I may be a dolt, [/quote] Don’t be so diffident. There’s no “may be” about it. > [@Ensign Edison](#): > > but you’re apparently illiterate. Dolt, there is much evidence on these very boards that I am nothing of the sort. > [@Ensign Edison](#): > > What does whether you know the Monty Hall solution have anything to do with what I said? You dragged in the Monty Hall in order to line me up with those who Just Don’t Get It, and I’m obliged to point out that you can’t use _that_ brush to tar me with. > [@Ensign Edison](#): > > I have pointed out your error. No, you haven’t. Saying “You’re wrong!” isn’t pointing out anything. If you said something like “On line 16, **Malacandra** cancelled the unknown variable from both sides of the equation, which is fallacious as it ignores the solution x=0”, then you’d be pointing out something worth paying attention to. Y’all know the difference between an argument and a contradiction, obviously. > [@Ensign Edison](#): > > face has pointed it out in great detail, with some help from others, and I have now attempted to point it out in more general terms. You’re not even familiar with the terminology of the science you’re trying to insist is wrong. It’s not face’s opinion, what she’s telling you, it’s facts about how stats and probability actually works. “You’re wrong. **you with the face** is right. Hah! Pwn3d!!!111!!one!”. Do _you_ actually know enough about the subject to be entitled to advance an opinion? > [@Ensign Edison](#): > > But you’re paying so little attention to everything but your own bullshit that up until a couple pages ago, you were still mysteriously claiming there was some ideological reason for her position, as if she’s not completely correct about how statistics can actually be used. It’s like you’re a flat-earther arguing with an astronaut, or a Creationist arguing with evolution. You sound that exact level of stupid and ignorant at this point. So no, I can’t possibly do anything else to “point out your error” because your error is the assumption you have that you know what the fuck you’re talking about. blah blah ad hominem blah blah you’re ignorant blah blah. Fuck off back to kindergarten and leave the field to **ywtf** who is at least _advancing_ some arguments - that I think them wrong does not detract from this fact - and not just saying without elaboration that I am talking like a flat-earther or a Creationist, 'cos, like, dude, everyone knows they’re full of shit and if I say you sound like one then, like, you must be full of shit too, hyuck hyuck. :rolleyes: [Previous page](https://boards.straightdope.com/t/huerta88-your-handle-hurts-you/358344.md?page=26)