Making an Amanda Knox prediction market

post by Kevin · 2011-06-30T00:53:26.331Z · score: 2 (3 votes) · LW · GW · Legacy · 24 comments

I'm not sure if anyone has been following the appeal...

 

but I'm willing to bet up to several hundred dollars at even odds that Amanda Knox will be released on appeal.

24 comments

Comments sorted by top scores.

comment by Pavitra · 2011-06-30T03:32:34.436Z · score: 3 (3 votes) · LW(p) · GW(p)

At what odds?

Also, I'd guess that p(released on appeal) < p(innocent). Are you taking that into account?

comment by Kevin · 2011-06-30T06:22:11.719Z · score: 2 (2 votes) · LW(p) · GW(p)

Even odds. If you think I'm not sufficiently taking that into account, perhaps you should take the other side of the bet.

comment by rolf_nelson · 2011-07-02T06:17:10.523Z · score: 2 (2 votes) · LW(p) · GW(p)

Dang, civil-case reversal rates are much higher than the U.S. (http://scholar.google.com/scholar?cluster=11027117874758072323), I still can't find anything on criminal cases though. komponisto said about 1/3, any cite on that?

There's lots of interesting high-profile Italian murders on Wikipedia, but after excluding those related to the mafia, terrorism, or serial-killers, there's not much recent activity left. Still, three of the ones I found (the Cogne homicide, the Novi Ligure murder, and the "Beasts of Satan") were partially or fully upheld, and the fourth (Nicholas Green) was reversed from acquittal to conviction. (I guess there's no double-jeopardy protection in Italy, since that would deprive them of additional opportunities to reverse. ) So I'll poke around a bit more when I get a chance, but so far a 50/50 bet is feeling moderately advantageous to me, even with the DNA review results.

comment by komponisto · 2011-07-04T21:38:35.569Z · score: 0 (0 votes) · LW(p) · GW(p)

civil-case reversal rates are much higher than the U.S. (http://scholar.google.com/scholar?cluster=11027117874758072323), I still can't find anything on criminal cases though. komponisto said about 1/3, any cite on that?

It was a remark by Amanda Knox's attorney, Luciano Ghirga. I can't remember where offhand.

One would expect the rate to be much higher than in the U.S. simply due to the rules of the game: unlike the U.S., Italian courts at the first level of appeal are allowed to reexamine the facts of the case, instead of merely ruling on whether procedure was correctly followed.

comment by rolf_nelson · 2011-07-01T22:34:42.979Z · score: 2 (2 votes) · LW(p) · GW(p)

Hey Kevin, thanks for pinging me, sounds exciting. I'd bet Knox's odds are only somewhat better than the average guilty defendant of release on appeal, that "somewhat better" based on her having a more expensive legal and PR team than the average defendant. Can't find such info easily though, I'll google around tonight. Wikipedia says we're still on the first of the two mandatory appeals, do you mean released on the first appeal or on any appeal? What if it's remanded back to the lower court? Also, I assume you mean 'released on the murder charge', not on libel (although that might be "time served" anyway by that time.)

For tax purposes, wagering a charity donation would probably be better, but cash might be doable, I'll need to think about it. Let me anyway research tonight how favorable p(successful appeal | guilt) looks to me.

That said, from a "getting to the truth" perspective, I still think a 1-on-1 debate is a better way of getting to the truth in the Kercher case than this wager, given the additional uncertainty of p(successful appeal | guilt).

comment by komponisto · 2011-07-04T20:01:53.548Z · score: 0 (0 votes) · LW(p) · GW(p)

Wikipedia says we're still on the first of the two mandatory appeals, do you mean released on the first appeal or on any appeal?

My PredictionBook prediction was for the current (first) appeal (I specified "appeals court in Perugia"; the second appeal takes place in Rome).

That said, from a "getting to the truth" perspective, I still think a 1-on-1 debate is a better way of getting to the truth in the Kercher case than this wager, given the additional uncertainty of p(successful appeal | guilt).

I agree. Although there should be at least some correlation between court verdicts and reality, a LW-style discussion would be much richer in information content.

comment by gwern · 2011-06-30T14:02:42.584Z · score: 2 (4 votes) · LW(p) · GW(p)

For those who don't have a bunch of money to throw around, there's always http://predictionbook.com/predictions/1804

comment by rolf_nelson · 2011-07-01T22:41:14.683Z · score: 1 (1 votes) · LW(p) · GW(p)

So, komponisto, Kevin, Pavitra, or anyone else, any general thoughts on how to calculate p(K | guilt) or p(K | innocence)? (K meaning Kevin's claim, that Knox will be released on appeal).

comment by komponisto · 2011-07-04T21:32:03.551Z · score: 0 (0 votes) · LW(p) · GW(p)

So, komponisto, Kevin, Pavitra, or anyone else, any general thoughts on how to calculate p(K | guilt) or p(K | innocence)?

Elsewhere, I described my reasoning on p(K | innocence) as follows:

Start, per Ghirga, with approximately 33% probability, or 1:2 odds. Then update on the fact that Amanda and Raffaele are factually innocent; this takes me upward to about the level of total ignorance, 50%, or 1:1 odds. By Bayes' theorem this implies that I must think that innocence is twice as likely in the case of acquittal as in the case of conviction. I think that's a reasonable, even conservative, estimate for a judicial system that is supposed to have some connection with reality. (Some may want to scoff at the notion that this accurately describes Italy's system, but rhetoric doesn't win bets!) Upon the granting of the DNA review in December I updated slightly to 60% (3:2 odds), implying that I think this was 1.5 times as likely to happen in an acquittal-world as in a conviction-world. (Which again strikes me as kind of conservative.) Now that we have the results, I'm at 80%, or 4:1 odds; to have gotten there from 3:2, I must have thought that this result was 8/3 (about 2.67) times more likely under the assumption of eventual acquittal than under the assumption of re-conviction.

comment by komponisto · 2011-06-30T22:29:29.606Z · score: 1 (1 votes) · LW(p) · GW(p)

I'm not sure if anyone has been following the appeal...

I have. .:-)

I have updated my probability of a successful appeal to 80% in light of the DNA report (which I have a copy of and have been helping to translate). So, numerically, you should be wanting to make a bet with me in the other direction (unless you've updated significantly on what I just said).

By the way, since there is a lot of interest in Asperger's here, I'll point out this article from Time:

http://healthland.time.com/2011/06/30/could-amanda-knox-have-an-autism-spectrum-disorder/

What do folks think? She is a fan of HPMoR, after all! (Normally I'd wring my hands about overdiagnosis of Asperger's and knee-jerk association of certain personality traits with the syndrome, but...is it time for me to just give in?)

comment by Kevin · 2011-06-30T22:35:33.725Z · score: 2 (2 votes) · LW(p) · GW(p)

Yes, I'd take the 20% side of the bet. Shall we?

And I agree with the autism likelihood, and think the Italian inquisitor confused autism for sociopathy. I read the Rolling Stone article referenced in the Time blog and thought of it as clear evidence of autistic behavior before reading the Time article. http://www.rollingstone.com/culture/news/the-neverending-nightmare-of-amanda-knox-20110627?print=true

comment by komponisto · 2011-07-04T19:51:30.553Z · score: 1 (1 votes) · LW(p) · GW(p)

(Apologies for the delay...I've been away working on various translations, including of the [report].)(http://knoxdnareport.wordpress.com/).

Yes, I'd take the 20% side of the bet. Shall we?

I don't see why not.

And I agree with the autism likelihood,

I entertained it for a day or two, but I really don't think so. There are a lot of disconfirming signs that the Time article ignores. I think it's just a matter of pure personality and (sub)cultural mores.

comment by Kevin · 2011-07-04T09:36:07.256Z · score: 0 (0 votes) · LW(p) · GW(p)

Do you and Rolf want to have that debate now? Might be fun, especially since your side of the debate is so easy...

comment by komponisto · 2011-07-04T19:52:59.724Z · score: 0 (0 votes) · LW(p) · GW(p)

Do you and Rolf want to have that debate now?

I've been ready for over a year!

comment by rolf_nelson · 2011-07-06T03:37:04.532Z · score: 0 (0 votes) · LW(p) · GW(p)

That'd be great! What forum should we use, should we start a Google Group? We can do it here if you like, but I'm concerned about getting distracted by other comments; the original threads got too bulky to hold a coherent conversation. It's up to you.

For my part, I'm happy to walk through the Micheli report or one of the many lists of evidence against Knox and Sollecito on the Internet, but I would pick as my top three pieces of evidence:

  1. The double-DNA knife

  2. Sollecito's DNA on Kercher's bra strap

  3. The mixed-DNA blood

If you could puncture that evidence, then from a "sampling" point of view, even though there is a large amount of additional evidence, I would agree that Knox has a large chance of being innocent.

comment by komponisto · 2011-07-06T16:37:43.507Z · score: 0 (0 votes) · LW(p) · GW(p)

That'd be great! What forum should we use, should we start a Google Group? We can do it here if you like, but I'm concerned about getting distracted by other comments; the original threads got too bulky to hold a coherent conversation.

That's a problem that would presumably exist to some extent on any public forum. I'm not too bothered by it myself; my feeling is that in such a situation one is not necessarily obliged to reply to all comments individually.

Here's my suggestion: why don't we try it here first, and see how it works? I'd be interested to see if this kind of thing can work on LW. There probably won't be as many comments now as before, since it isn't a new topic.

If it does get unwieldy, we can always move somewhere else.

For my part, I'm happy to walk through the Micheli report or one of the many lists of evidence against Knox and Sollecito on the Internet, but I would pick as my top three pieces of evidence:

  1. The double-DNA knife

  2. Sollecito's DNA on Kercher's bra strap

  3. The mixed-DNA blood

Rather than the Micheli report (which was the first-level ruling in the Guede case), you'd presumably be better off with the Massei-Cristiani report, which was actually about the Knox-Sollecito case (and is even available in translation by the PMF people).

Interestingly, I agree with you that the knife and bra clasp are the strongest pieces of prosecution evidence (though I would have put them in the reverse order). However, they've been pretty severely punctured by Conti and Vecchiotti in their report. You can read the conclusions of that report here.

Anyway, I'll create a new Discussion thread upon confirmation that that's acceptable to you.

comment by rolf_nelson · 2011-07-08T02:04:38.518Z · score: 1 (1 votes) · LW(p) · GW(p)

That's a problem that would presumably exist to some extent on any public forum. I'm not too bothered by it myself; my feeling is that in such a situation one is not necessarily obliged to reply to all comments individually.

We could access-control the main section of a Google Group or a mini-blog, creating a separate area for comments if we like. That would also be convenient because I can easily notice if you've posted when I check Google Reader.

Here's my suggestion: why don't we try it here first, and see how it works? I'd be interested to see if this kind of thing can work on LW.

That is an excellent reason to do it here, then; go ahead and create a new thread.

Interestingly, I agree with you that the knife and bra clasp are the strongest pieces of prosecution evidence (though I would have put them in the reverse order). However, they've been pretty severely punctured by Conti and Vecchiotti in their report. You can read the conclusions of that report here.

If their main point is that the evidence doesn't meet the standard of scientific rigor, then I might not disagree with them on anything factual. Very little evidence, either way, does meet the standard of scientific rigor. Fingerprints never reach the standard of scientific rigor. DNA testing, as practiced, probably rarely if ever meets the standard of scientific rigor. Eyewitness testimony obviously can never come close to meeting the standard of scientific rigor. Heck, most science doesn't meet the standard of scientific rigor. We still need to evaluate evidence on its full merits.

So, to confirm that our initial analysis here diverges, around how much are you currently shifting based on the test results for the knife and for the bra clasp? For you, did either one shift P(guilt) by a factor of 100? 10? Not at all? It's a difficult question to answer in a calibrated way, so if you want to skip that one I'll understand.

You can guess my hypothesis for why the DNA tests came out the way they did. Do you have a specific alternative hypothesis or hypotheses you want me to consider? Is your main claim here that you believe the knife was accidentally contaminated in the laboratory, and the bra strap was accidentally contaminated in Kercher's room? Or is there a different alternative hypothesis I should consider first?

I assume we're not going to quash any evidence, since we're a court of Bayes and not a court of law? That is, I'm proposing that whenever we want to exclude or diminish evidence, we should have a Bayesian reason for why the evidence doesn't really alter P(guilt). The proposal is partly because it's the correct Bayesian thing to do, and partly because trying to divine and mimic Italian criminal procedure and admissability rules would add (IMHO unnecessary) additional complexity.

comment by komponisto · 2011-07-08T08:26:25.896Z · score: 0 (0 votes) · LW(p) · GW(p)

New thread here; reply to this comment here.

comment by komponisto · 2011-07-04T20:02:59.705Z · score: 0 (0 votes) · LW(p) · GW(p)

For any interested: I've begun a translation of the DNA report.

comment by rolf_nelson · 2011-07-01T22:34:25.558Z · score: 0 (0 votes) · LW(p) · GW(p)

Hey Kevin, thanks for pinging me, sounds exciting. I'd bet Knox's odds are only somewhat better than the average guilty defendant of release on appeal, that "somewhat better" based on her having a more expensive legal and PR team than the average defendant. Can't find such info easily though, I'll google around tonight. Wikipedia says we're still on the first of the two mandatory appeals, do you mean released on the first appeal or on any appeal? What if it's remanded back to the lower court? Also, I assume you mean 'released on the murder charge', not on libel (although that might be "time served" anyway by that time.)

For tax purposes, wagering a charity donation would probably be better, but cash might be doable, I'll need to think about it. Let me anyway research tonight how favorable p(successful appeal | guilt) looks to me.

That said, from a "getting to the truth" perspective, I still think a 1-on-1 debate is a better way of getting to the truth in the Kercher case than this wager, given the additional uncertainty of p(successful appeal | guilt).

comment by Clippy · 2011-06-30T15:15:22.239Z · score: 0 (0 votes) · LW(p) · GW(p)

How about bets denominated in bitcoins, since a lot of Website:Lesswrong.com readers don't have easy access to the regular financial system or long-distance travel capability?

comment by Kevin · 2011-06-30T22:37:47.691Z · score: 0 (2 votes) · LW(p) · GW(p)

Right now Bitcoin isn't a regular enough currency for the denomination of goods. It's more of a store of value.

comment by Clippy · 2011-07-01T15:01:29.728Z · score: 1 (1 votes) · LW(p) · GW(p)

When do you anticipate it being economical not to discriminate against non-human intelligences?

comment by Kevin · 2011-07-01T23:29:51.475Z · score: 0 (0 votes) · LW(p) · GW(p)

10 years? Maybe 5.