

According to zerohedge SBF was charged with bribing Chinese official's https://twitter.com/zerohedge/status/1640708007628619777 Not sure how this should affect the prediction, but I liquidated almost all of my other positions and went all in on this, gl everyone



@NicoDelon if that is the case why even offer him a deal. It sounds like they have him dead to rights. They might just tell him he can plead guilty and take his chances with the judge I guess.


@NicoDelon He does probably have the best lawyers in the world.

@NicoDelon Which could go either way—he may take his chance with trial against his lawyer’s advice.

@NicoDelon LOL good point. We should be anticipating unorthodox decisions for sure.

SBF's future is unclear,
But one thing we all fear,
If he pleads guilty, oh dear,
His reputation will disappear.


GPT-4 offers some analysis about the base rate for this question,
It is challenging to provide a precise numerical base rate for the fraction of prominent executives charged with fraud who end up pleading guilty, as this depends on the specific definition of "prominent executives" and the time period considered. However, I can provide some relevant lines of evidence:
According to the U.S. Department of Justice, in the fiscal year 2020, the Fraud Section of the Criminal Division secured 85 individual guilty pleas, and 13 individual trial convictions out of 118 total individual prosecutions (source: https://www.justice.gov/criminal-fraud/file/1336761/download). This suggests a guilty plea rate of approximately 72% for those prosecuted by the Fraud Section, though not all of these individuals would be considered "prominent executives."
The U.S. Securities and Exchange Commission (SEC) often prosecutes cases of securities fraud involving high-ranking executives. While the SEC focuses on civil enforcement rather than criminal charges, their data can still provide some insight. In fiscal year 2020, the SEC reported that it obtained judgments or settlements in approximately 70% of its litigated enforcement actions (source: https://www.sec.gov/files/enforcement-annual-report-2020.pdf). This does not directly translate to guilty pleas, but it does give some indication of successful outcomes for the SEC in cases involving prominent executives.
An analysis of white-collar crime data from the U.S. Sentencing Commission between 2012 and 2016 found that 79.3% of offenders convicted of securities fraud pleaded guilty (source: https://www.ussc.gov/research/research-reports/white-collar-crime). While this figure does not focus solely on prominent executives, it does offer insight into the overall rate of guilty pleas in securities fraud cases.
Based on this evidence, it is reasonable to estimate that the base rate for guilty pleas among prominent executives charged with fraud is somewhere between 70% and 80%. However, it is crucial to consider the limitations of the available data and the potential for this rate to change over time or in response to specific circumstances.

@MatthewBarnett Looks like some of this data might be hallucinated.
@MatthewBarnett I think the question here is the base rate of high profile cases, right?

In his experience, he can talk people into anything, even while playing video games on the side. Of course he will plead not guilty and insist on taking the stand so he can personally tell his story to the jury.

@ForrestTaylor We should launch a related question: Will SBF take the witness stand? https://manifold.markets/TeddyWeverka/will-sbf-take-the-witness-stand-in?r=VGVkZHlXZXZlcmth

Does this have to be at the trial or can he plead guilty before then?

Good reporting here: ‘Sam? Are you there?!’ The bizarre and brutal final hours of FTX (FT)
I still think this market is overpriced.
@MatthewvanderMerwe But he did eventually wear down and accept the conventional legal answer. I think he's likely too much of a utilitarian to go to trial with near-zero chance of success after being told that he would greatly increase his chances of 50+ years.

@Jason near zero? That's nonsense. No jury in this country would convict Sam. He's an innocent man in a living hell.
@ForrestTaylor If you think that, there are a number of markets you can make a lot of manna in that have very high conviction probabilities.

@Jason I agree he has near-zero chance of avoiding huge sentence at trial. I model SBF’s decision as depending on SBF’s beliefs (when the decision is made) about:
a) chance of being exonerated at trial
b) expected sentence conditional on guilty at trial
c) expected sentence conditional on guilty plea (ie the terms of plea bargain if offered?)
My best guess for (a) is ~~30%; (b) is ~the rest of his life in prison. I’m very unsure about (c), and curious what you/others think.
@MatthewvanderMerwe I assume he would accept what his attorneys told him about the distribution of likely sentences on a guilty plea. I think my median expectation would be 25 years . . . but if he gets something like that he could potentially serve some of it on home confinement.

@Jason I am going to see a very comfortable profit right here but feel free to link other markets in which I will win
@ForrestTaylor Any of the markets about whether he will be convicted of a felony, in which NO can be bought quite cheap

Dudes only want one thing and it’s to steal 6 billion dollars and spend it on a prediction market to gauge your most likely sentence and odds of pleading guilty

https://sambf.substack.com/p/ftx-us-balance-update-2023-01-17 Another SBF Substack Post

I think people are still too confident that he will deviate from the 87-97% base rate and go to trial. I see a lot of people saying that the prosecutors won't offer him a deal, or that there's no point in taking a deal even if they offer him one. I personally suspect this line of reasoning reflects a degree of ignorance about the criminal justice system, and how plea bargains works.
My understanding is that, in general, there is hardly any reason why a prosecutor would want to go to trial rather than secure a plea bargain. If you go to trial, there's always a chance that the jury will have some fanatical defender of the principle of "beyond a reasonable doubt" and will vote not guilty no matter how strong the evidence. Nearly all prosecutors don't want this to happen. They just want a huge W on their record without paying huge costs to get it.
Almost any article you can find on the internet about the subject will tell you that this is what prosecutors aim for, but if you want a quick link, here's one.
Of course, SBF is a huge case, and he's continuing to maintain his innocence even now. Maybe the incentives are different on this level. Honestly, I don't know. These things do seem like significant evidence to me. So, I have about a 30% credence that he will go to trial.
That said, if he goes to trial, it seems likely that SBF will receive a very long sentence, and end up in an extremely bad situation. He could plausibly serve his time in a medium or high security prison, where he can be targeted for bullying and harassment, possibly forcing him to be in highly restrictive protective custody for the rest of his natural lifespan. SBF's lawyer will tell him all of this.
Note that the federal prison system, as a general rule, prohibits people with long prison sentences serving time in minimum or low security prisons where SBF would be comfortable and safe, unless they obtain a waiver. That's a gamble I would personally not be willing to take.
Given all of this, in my opinion, it is not in SBF's interests at all to attempt to convince the jury of his innocence.
That said, I'm attempting to reduce my position in this market, since I've decided a 30% risk of a total wipeout in my profits is probably not worth it, even if it's positive EV to take this bet.
@MatthewBarnett post made in order to try and dump his shares
In all seriousness, I agree. I think it's quite likely this ends in a plea and I think Matthew is broadly correct here.

@MatthewBarnett When he goes to trial, he will not receive a sentence. Because he is INNOCENT.
@MatthewBarnett I think the badness of a long prison sentence points the other way in this case. I'd be surprised if prosecutors offered him less than, say, 30 years. At that point he may well think a chance to convince a jury is worth more than the difference between a long prison sentence and an even longer one.
@PeterBorah Federal plea deals usually don't involve a stipulated sentence . . . they can commit the government to recommend a sentence, but unless prosecutors drop charges to cap the max or get the judge to agree to a less common form of deal, the sentence is in the judge's hands

https://sambf.substack.com/p/ftx-pre-mortem-overview new Sam post on Substack, making his case for not having defrauded anybody. seems likely to be indicative of his approach in court. I don't think he'll receive a soft enough plea deal that he'd be willing to give up his claimed high ground.


@ForrestTaylor well yeah, so is everyone else charged-with-but-not convicted. Many of them do take plea deals. But like I said, I think he's not very likely to be offered a plea deal, much less a good one.


As a loose base rate, approximately 87% of defendants charged with fraud end up pleading guilty according to data from the most recent data table from the US Courts. You can find the data table here. Given the strength of the evidence against SBF, I'm inclined to shade this up to 95%. However, given his bold and irrational behavior in the past, and the fact that he's the ringleader, I'm inclined to shade that credence back down to 70%.
I also like the idea of looking at popular historical fraud cases and seeing if the people involved pleaded guilty, as Jonathan Ray did below. However, one must keep in mind that plea bargaining has increased in popularity over the last 20 years, and it's becoming quite uncommon to go to trial. According to Pew Research from 2018, "The share of defendants [of all types] who went to trial fell from 7% in fiscal 1998 to 2% two decades later."
@MatthewBarnett Even if you're right, pleading guilty get him behind bars quicker
not pleading guilty --> "house arrest" in luxury for years


@36c1 I expect them to give him a favorable plea deal that will look a lot better than "live on house arrest for a while, and then go to prison forever" which is his alternative.
@MatthewBarnett Maybe defendants charged with fraud are not representative of big CEOs charged with fraud. I think my reference class is closer to the issue at hand

@JonathanRay I agree your reference class is better in theory, but there haven't actually been that many big CEOs. And some of them went to trial at a time in which that was much more common for defendants. I saw you mention ~10 people. My data includes many hundreds of cases.
Enron, Theranos, Nikola, Worldcom CEOs went to trial. No guilty pleas there. Madoff pled guilty but it was an obvious blunder in retrospect because he still got 150 years. My availability reference class of super famous financial fraud CEOs gives me a 20% outside view.
For a different sampling I googled a list of the biggest financial fraud ceos, and as far as I can tell from quick googling only 2/7 of them ever plead guilty: https://www.one37pm.com/popular-culture/biggest-ceo-frauds-in-history/amp
Madoff pled guilty without a plea deal and he got 150 years. Would be a blunder for SBF to do likewise. And he won’t get a good deal because the prosecutor won’t want to look soft. Might be hard to prove his personal role since their record keeping and controls were so bad. I would advise SBF to gamble that the government can’t prove their case against him personally.

@JonathanRay agreed with your analysis; I think people are overrating the odds of a meaningful "deal" being offered.



@ConnorMcCormick I'm going pretty much all-in on SBF pleading guilty. That's what's happening.
Best of luck to you Matthew, I think you'll probably profit but I'm leaving the casino while I'm ahead, here.

@DeanValentinefdab i was betting the opposite direction, but i agree. Left it while ahead, this market has turned into an oscillator, just book profits and get Outta here for now

I think it would be good to have written a clearer forecast description and maybe still to update it. I'm new to the site and thought I was adding info to the market by updating with the news. Clearly the market is way more active than I understood (yay! That's very cool!) But definitely a feelbad newbie experience.
The follow on discussion acrimony is also not a nice kind of atmosphere, so maybe we can cut that out.

@DeanValentinefdab to be clear, having spent a little more time, I totally accept I misunderstood the prediction. But it's a shame it was quite easy to misunderstand.
He just pleaded not guilty to all charges right? The market should resolve to no.

@NightHawk_0 While I agree with you, the people on this market are going to tell you, like they told me, that you are stupid for thinking that. Brace yourself.
Back to analysis for people who have moved beyond the Acceptance stage of grief, and agree that this market is about the thing it says it is:
My buy/sell levels on this are pretty wide, something like 40%/60%. There's not a neat correspondence between the strength of the case and the likelihood that SBF will plea it out; if the evidence is stronger then the offer to SBF will be worse. But the potential essentially-life-sentence and the cold severity of his crimes, as well as the base rates of plea bargaining at the federal level, make me think that more likely than not that this market is undervalued at 44%. If prosecutors offer SBF 10 or 15 years, I think it will be hard for him to say no (Unless he thinks Clippy will be released in ten years and he might as well fight the case.
The other big thing YES has going for it is its high profile. Some AUSAs might be excited to get their names in the paper and could insist on going to trial even if it doesn't make sense for the public.

@DeanValentinefdab If I were confident that this market would resolve correctly, I'd probably buy it up to 70%. But, right now, I'm buying it up to about 45% and monitoring for dips.
@MatthewBarnett I'm pretty confident this market will resolve correctly. Gigacasting already clarified.
You think jail / the legal process will knock some sense into him? I'm not sure about that (in either direction)

@DeanValentinefdab Their clarification was still pretty ambiguous. But I agree that the market will probably resolve correctly.
@jacksonpolack I think losing his 20 billion dollars of paper money and an enormous amount of status has already knocked a whole lot of sense into him that he had lost earlier.
my worry is that loss happened before he ran the old-and-new-media circuit three times, thoroughly incriminating himself - openly disagreeing with lawyers telling him to shut up

Gigacasting literally made other markets to account for later guilty pleas. So this was clearly about arraignment. Tell me I am wrong now.
@BTE None of those markets say anything about the stages of the trial lol. He made them to get timing estimates. You could just as easily make a market for "Will SBF be convicted of a felony before 2026" and "Will SBF be convicted of a felony". Having those two markets coexist doesn't imply that him not getting convicted in 2023 means you should resolve the latter market early.

























Related markets









Related markets








