Accusations of cheating in the highest tournament levels of chess

[Moriarte]

As I mentioned in the OP, I am aware of his "peculiar" lack of analytican post-game skills, but if he is going to be accused of cheating, there surely should be some hard proof of it (stressing once more that the important game was the one Magnus lost, and it was not online).

I don’t think there will be proof of otb misconduct. What we can have is better tools for over the board detection and higher standards. Such as prohibiting online cheaters from competing over the board.

No known way of cheating offline? I assume you play chess judging from the interest in the topic?

Take a look at this, people have been cheating “offline” as long as chess exists…

Cheating in chess - Wikipedia

en.m.wikipedia.org

 

[Kyriakos]

Afaik there is no proven case of anyone in such high setting (important tournaments) who cheated by using a chess engine; which is obviously the kind of cheating we are talking about

And no, I don't play chess. I am here for the scandal

 

[Moriarte]

Afaik there is no proven case of anyone in such high setting (important tournaments) who cheated by using a chess engine; which is obviously the kind of cheating we are talking about

And no, I don't play chess. I am here for the scandal

The chess world salutes you!

Here you go, one of the many examples, since you won’t read the link anyway

In the Subroto Mukerjee memorial international rating chess tournament 2006, an Indian chess player was banned from playing competitive chess for ten years due to cheating.[69] During the tournament at Subroto Park, Umakant Sharma was caught receiving instructions from an accomplice using a chess computer via a Bluetooth-enabled device which had been sewn into his cap.[70][71] His accomplices were outside the building, and were relaying moves from a computer simulation. Officials became suspicious after Sharma had made unusually large gains in rating points during the previous 18 months, even qualifying for the national championship.[71] Umakant began the year with an average rating of 1933, and in 64 games gained over 500 points to attain a rating of 2484. Officials received multiple written complaints alleging that Umakant's moves were in exactly the same sequence suggested by the chess computer.[70] Eventually, in the seventh round of the tournament, Indian Air Force officials searched the players on the top eight boards with a metal detector and found that Umakant was the only player who was cheating. Umakant's ten-year ban was imposed by the All India Chess Federation (AICF) after reviewing evidence presented by Umakant himself and the electronic devices seized by the tournament organizers.[69] The penalty was considered harsh, especially considering that those in other sports who have been found to be doping and match fixing did not receive such lengthy suspensions.[72] When officials were asked about the suspension they stated, "We wanted to be frank and send a stern message to all players. It is like cheating on exams."

 

[Askthepizzaguy]

Perhaps it's the mafia player in me, but I believe Magnus Carlsen has the experience of playing with chess prodigies, young and up and coming grandmasters, to be able to feel quite strongly that there's something off about Hans' play.

You recognize certain patterns. When Magnus says that when he's faced off against like a Praggnanandhaa and can tell when they're really expending effort and stress to outthink and outplay him, that rings true.

If you've played every grandmaster and most up and coming IMs, you know when someone is struggling to beat you or not. Magnus has also played countless online games, and he has been able to tell when someone's using an engine on him, either from the start or out of nowhere. Just effortlessly outplaying you out of nowhere.

You definitely, when you've had that level of experience, can spot when someone is not actually expending effort in beating you. No question, no contest.

That's not admissible evidence. It's not a reason why Magnus should be able to single handedly destroy someone's chess career. He should have played the games with hans or withdrawn from the tournament before playing anyone.

He should not have singled anyone out without evidence. It's all wrong.

For the past 14 years I've played games where I catch people lying, pretending to be engaged in the same tasks I am. I've also had experience as a moderator catching permabanned people who come back and pretend to be someone else, and fail at it.

The difference between someone who catches liars, cheaters, and ban dodgers all day, every day, and someone who can judge someone else for it, is vast. It's why prosecutors are not judges or juries in our legal system.

I don't really have any consequences when I've been wrong. And when I am right in the past, it doesn't make me right now.

Magnus can be an expert in detecting someone is cheating against him by brute force of his experience, his intuition, and he's right that the guy has cheated a lot in the past.

The guy analyzed the game against magnus post-game and got a lot more things wrong than someone of his ELO rating. The commentator was outplaying him in the analysis simply by suggesting moves to see what Hans would do.

I am pretty sure Hans cheated. We also know by his admission, he's cheated in the past. Actual strong analysis has concluded he's cheated much more than he's admitted to.

All of that being said, none of that is actual proof of cheating (over the board, in person, ever, or in the Sinquefield Cup, as alleged). Thinking you know something is not the same as knowing it. Being a genius at catching cheaters is not the same as knowing someone is guilty.

Unless you actually catch Hans Niemann cheating over the board, using some legitimate method of cheat detection, and/or absent a confession from him, strong suspicion is not proof.

A while back on a certain site, I believed someone was cheating in a game, by trying to enter the game under two different accounts. My suspicions alone are not proof.

I was right. Admins checked. He was cheating.

I turned around and saw another new account entering the game, and it was behaving just like the other, repeated ban dodgers. An admin checked, and I was wrong. This was someone else, someone I knew for a fact wasn't the ban dodger.

I'd caught that particular single ban dodger over 20 times in my life. He's instantly recognizeable. I've been right so often, I got cocky. I accused an innocent person.

Believing you are right is not the same as knowing.

Magnus is probably right, given everything we know. Hans has a history of cheating, a history of lying about it even after being caught, his games with money on the line are engine moves, or he's the most brilliant chess player ever to exist by far on certain days of the week and a sub-par middling grandmaster other days. Wildly different.

But, unless they can say how Hans actually cheated over the board, Magnus is out of line by using his position to try to publicly destroy Hans.

Just as I was out of line to publicly accuse someone who was innocent.

It doesn't feel right, it doesn't feel like justice, when someone you're "sure" is cheating can get away with it simply because you don't have decisive evidence. But the reason why people are innocent until proven guilty, is because sometimes they are.

Hans should be treated as his earned reputation suggests: He's not above suspicion, and any tournament he is in, the organizers should assume he's trying to cheat and make sure he cannot.

Because that's the kind of guy he is, and has shown himself to be.

Tournament organizers should also keep that in mind. But Magnus is not the rule of law, and he alone shouldn't be able to stop Hans from playing.

It would be like allowing me alone to decide who gets banned on every website, on my intuition and experience. I'm good, but I just told you I've gotten it wrong before. That's why a system exists, to prevent folks from abusing their power and position.

Magnus is almost certainly right, in my mind. He knows Hans almost certainly cheated in the game against him.

He doesn't know it for a fact, and he is wrong to treat Hans as though he does know. That's the flaw. And people who know nothing of the situation can have their opinion, but we also don't know. We shouldn't act as if we do.

If the day comes that Hans is proven beyond doubt to be cheating in person, it's probably the end of his over the board chess career, in tournaments anyway.

That day hasn't come yet, and until it does, he's gotta be allowed to play and treated as above board.

 

[Qactus]

One of the reasons I prefer Chess boxing, haven't heard of any cheating accusations there so far...

 

[Askthepizzaguy]

People don't have to invite him to tournaments. People don't have to be in tournaments where he plays.

But they shouldn't accuse him without evidence, and should play him when they agree to play in the same tournament.

May not be a satisfying position to take, but until further evidence, it's the only right answer.

 

[Kyriakos]

The chess world salutes you!

Here you go, one of the many examples, since you won’t read the link anyway

In the Subroto Mukerjee memorial international rating chess tournament 2006, an Indian chess player was banned from playing competitive chess for ten years due to cheating.[69] During the tournament at Subroto Park, Umakant Sharma was caught receiving instructions from an accomplice using a chess computer via a Bluetooth-enabled device which had been sewn into his cap.[70][71] His accomplices were outside the building, and were relaying moves from a computer simulation. Officials became suspicious after Sharma had made unusually large gains in rating points during the previous 18 months, even qualifying for the national championship.[71] Umakant began the year with an average rating of 1933, and in 64 games gained over 500 points to attain a rating of 2484. Officials received multiple written complaints alleging that Umakant's moves were in exactly the same sequence suggested by the chess computer.[70] Eventually, in the seventh round of the tournament, Indian Air Force officials searched the players on the top eight boards with a metal detector and found that Umakant was the only player who was cheating. Umakant's ten-year ban was imposed by the All India Chess Federation (AICF) after reviewing evidence presented by Umakant himself and the electronic devices seized by the tournament organizers.[69] The penalty was considered harsh, especially considering that those in other sports who have been found to be doping and match fixing did not receive such lengthy suspensions.[72] When officials were asked about the suspension they stated, "We wanted to be frank and send a stern message to all players. It is like cheating on exams."

So that's why Hans has so wild hair

 

[Moriarte]

Btw, has Nakamura ever won any title? He is a bit too willing to massacre Hans imo

5 time US chess champion. Olympiad winner. Currently #1 rated blitz player on the planet.

I think everyone concerned is a bit paranoid of what modern tech can do. That coupled with deep, life-time knowledge of chess patterns. High level players, masters, international masters and above, can reliably spot engine cheater just on intuition (knowledge). I know how it sounds, but I’ve seen this done many times on stream and such: master accuses random player of cheating to receive an automated apology from chess.com anti-cheat engine the following day stating that points were refunded. These masters don’t even need deep analysis, most of the time cheaters are obvious. And there’s a certain point in chess, where you have to be a master to understand irregularities, or take things on faith if enough masters form a consensus. People acquire ingrained “vision” for certain patterns when they develop as a player.

Now, what happens, when a person grows to ability level of grandmaster and then starts cheating? (yes, that happened before) Subtly, once a game, once every two games… In exactly the right spots other grandmasters won’t consider suspect. That is what Carlsen is claiming. Lack of friction too. And that implies some kind of tech. Have they all gone bananas tripping over each other trying to overblow this thing? Absolutely. But it is a curious accusation from a man equipped to make such accusation, who has the ability to smell “engine” game. And given his flawless life and reputation, many people believe him. (without being equipped with faculties to verify proof). So ok, Hans is innocent (lets say). But how do you catch a cheater of that caliber when he arrives? With a small non metallic pin in his ear, or something in his boot. It is a potential problem.

The answer is probably - you can’t. And that’s not good enough. So people are running in circles and trying to come up with fair play solutions for a new era.

 

[Kyriakos]

5 time US chess champion. Olympiad winner. Currently #1 rated blitz player on the planet.

I think everyone concerned is a bit paranoid of what modern tech can do. That coupled with deep, life-time knowledge of chess patterns. High level players, masters, international masters and above, can reliably spot engine cheater just on intuition (knowledge). I know how it sounds, but I’ve seen this done many times on stream and such: master accuses random player of cheating to receive an automated apology from chess.com anti-cheat engine the following day stating that points were refunded. These masters don’t even need deep analysis, most of the time cheaters are obvious. And there’s a certain point in chess, where you have to be a master to understand irregularities, or take things on faith if enough masters form a consensus. People acquire ingrained “vision” for certain patterns when they develop as a player.

Now, what happens, when a person grows to ability level of grandmaster and then starts cheating? (yes, that happened before) Subtly, once a game, once every two games… In exactly the right spots other grandmasters won’t consider suspect. That is what Carlsen is claiming. Lack of friction too. And that implies some kind of tech. Have they all gone bananas tripping over each other trying to overblow this thing? Absolutely. But it is a curious accusation from a man equipped to make such accusation, who has the ability to smell “engine” game. And given his flawless life and reputation, many people believe him. (without being equipped with faculties to verify proof). So ok, Hans is innocent (lets say). But how do you catch a cheater of that caliber when he arrives? With a small non metallic pin in his ear, or something in his boot. It is a potential problem.

The answer is probably - you can’t. And that’s not good enough. So people are running in circles and trying to come up with fair play solutions for a new era.

As I said, I am not a chess player (well, know how to play, but you get the point, I don't play online etc). But the moves Magnus himself does in some of the recent games he won (very convincingly) aren't miles away from those seen in the game where Hans beat him. In all cases we are talking about some opening (in late-mid game) sacrifice of a significant piece, which isn't readily identified as leading to the opponent losing at least as much within the space of 3-4 moves AND making his pawn positioning worse. So while Hans' looked suspicious due to lack of any analysis of why he did that (I have watched that video, it is rather bizarre, yes), no one even asked Magnus to provide analysis for his (imo, as an outsider) similar moves.
Sure, Magnus is the chess champion, but Hans is only 19. People should be able to pull this kind of game on occasion (now if it was done online, I'd also be far more suspicious, eg Hans sort of had a very peculiar series of good and then bad games in the recent online tournament).

 

[Askthepizzaguy]

I do want to say, after having ripped Magnus for ripping Hans publicly-

A lot of casual fans of chess decided, apparently without actually following chess very much if at all, that Magnus is complaining about Hans because he lost a game to Hans.

No.

Magnus has been consistently, for over a decade, extremely gracious and congratulatory toward everyone who has ever beaten him, and the only person he's ever upset at when he loses is himself.

He's probably the single most gracious and sporting world champion chess has ever had, especially compared to the likes of Kasparov or Fischer or Alekhine. Few world champions actually behaved like mature adults, unfortunately.

Shining among them is Magnus. He's quite simply the greatest ambassador for the sport that has ever been world champion. Others can only equal him.

For him to be this out of line, and this upset, and this sure he's been cheated against, is wildly out of character for him. Magnus is essentially willing to take his entire career of being a good sport and throw it away on this one singular cheating accusation.

He doesn't know for sure. But he's gotta be taken seriously. He has a triple platinum reputation for being a good sport. It absolutely was worthy of investigation.

My other point still stands. He still doesn't know. And he's not the law, all by himself.

 

[Samson]

But how do you catch a cheater of that caliber when he arrives? With a small non metallic pin in his ear, or something in his boot. It is a potential problem.

The answer is probably - you can’t

You could play the game in a faraday cage. It seems like you should be able to cause enough wide band EM interference that such things would not work, but I do not know the details of that.

 

[Kyriakos]

Ben's point is (tldw) that any such GM can win a couple of games against the chess champion. Obviously they'd lose in a series of many games, but in a one-off they still can win and shouldn't be suspicious. He also noted that Hans won one game against Magnus in the immediately previous ott (a few days prior to his far more famous win that kickstarted all this mayhem) tournament, and neither Magnus nor anyone else seemed to care.

Also, given the expression Hans has currently, if he is banned from serious tournaments at 19 he'll likely end up a serial killer

 

[Askthepizzaguy]

If a super GM gives a simul, they usually win most of the games, and a few, to save some time thinking and win all the other games, in very drawish positions, the GM may settle for a safe draw.

It is not unheard of for someone who has achieved such a draw to not accept the draw, and play on. On very rare occasions, the player may win, or get close to winning, and have the upper hand, especially if they're IM strength or higher.

One of the people I've watched online is Agadmator, another I've watched is Mato Jelic. Both have played far superior players before, super GMs, or world champions, and are proud of their best game they've played against a far superior player. Being able to even hold a draw is an achievement.

If, in such a game, the GM makes one of those rare one move blunders and drops a pawn or a whole piece, suddenly the IM has a shot at winning the game outright. If they're strong enough, they won't make enough mistakes for the GM to get back into the game.

So much lower rated players can win legit. Ben plays a lot of blitz and I've seen him lose to a 1400 player who was clearly not cheating. GM doesn't mean never blunders, it means boy, those blunders are rare.

Hans can beat Magnus legitimately, he's got enough strength. Magnus has to play badly, but Magnus plays badly sometimes.

What Magnus is saying though, about Hans not even struggling to achieve that win, and not being able to explain his analysis after the game, that's the part where I'm siding with Magnus.

A) If it was easy to do, it should be easy to explain. Boy, he failed at explaining it. So it clearly wasn't as easy as he made it sound, like when he declared the position to just be winning instead of explaining why, and then he got refuted by the commentator who is not a GM who made one move and said what would you do here? No good answer, and it was a line every single GM absolutely should have considered.

B) If it was hard to do, then Magnus is lying about Hans not feeling stress or expending effort to beat him, or he's reading him wrong, or Hans cheated.

I don't believe Magnus is lying. He's got a lot to lose here and nothing to gain, since he chose to drop the next game, and he chose to retire as world champion. He's also able to get his ELO rating back up, and clearly doesn't care about his ELO alone if he's resigning on move 2.

That leaves he's badly misread Hans through an honest mistake (I can believe that) and made a poor decision to attack Hans on suspicion alone, or his instincts are right.

Without evidence there's no way to decide between the two, because past cheating is not proof of present cheating. Chess.com has caught lots, and lots, and lots of cheaters who admitted to the cheating when presented with the evidence. Most of them still play.

The reason is because Chess.com wants everyone to be able to play, every GM, the world champion, hans, all the IMs, all the casuals like me who are like 1400 strength. Even the ones who have cheated before, because if they kick off everyone who has ever cheated once, they just come back with a new profile, a different IP address, and keep cheating under a different name. Or they actually leave the site, and the site is worse off economically.

Chess.com has every reason to want every player, including professional chess players who have cheated, on their site. But it's the rule that you've gotta stop cheating once you're caught, and admit to it, because the second ban is usually permanent.

It's like that for a lot of forums. If they banned everyone who ever broke a rule, half of us wouldn't be here, including me. Even on sites where I've been a mod.

But there's a cutoff, if you never learn, never get better, never try to follow the rules, there's no reason to have any mercy. It's unfair on everyone else.

Chess.com did their homework before banning Hans. He's been given multiple chances to reform both before and after becoming an adult. It reflects poorly on his character that even 2 years ago he was still doing it.

There's not a ton of difference in my mind between 19 and 17. You're not magically different once you cross the 18 year barrier.

 

[ywhtptgtfo]

One of the reasons I prefer Chess boxing, haven't heard of any cheating accusations there so far...

Beat your opponent until they have a concussion, then checkmate him as he struggles to think

 

[Kyriakos]

Beat your opponent until they have a concussion, then checkmate him as he struggles to think

Known as the Sicilian (mafia).

 

[TheMeInTeam]

Perhaps it's the mafia player in me, but I believe Magnus Carlsen has the experience of playing with chess prodigies, young and up and coming grandmasters, to be able to feel quite strongly that there's something off about Hans' play.

what is the % accuracy for town lynchings in mafia, even by the best players at it? from your experience, what happens when people are using a name or behaving in a way that already predisposed people to not like them, in terms of informing that accuracy?

All of that being said, none of that is actual proof of cheating (over the board, in person, ever, or in the Sinquefield Cup, as alleged). Thinking you know something is not the same as knowing it. Being a genius at catching cheaters is not the same as knowing someone is guilty.

i also worry because even geniuses aren't correct often enough to have good confidence. even less so when people are already predisposed to think the person in question cheated.

I think everyone concerned is a bit paranoid of what modern tech can do. That coupled with deep, life-time knowledge of chess patterns. High level players, masters, international masters and above, can reliably spot engine cheater just on intuition (knowledge). I know how it sounds, but I’ve seen this done many times on stream and such: master accuses random player of cheating to receive an automated apology from chess.com anti-cheat engine the following day stating that points were refunded. These masters don’t even need deep analysis, most of the time cheaters are obvious.

there are common tells for this not available to otb chess. such as a nearly-fixed time to make each move. they have a much harder time detecting someone who pulls out the engine for just a move or two, flips a coin on whether to use engine move, or if the cheater has enough knowledge of chess to prefer "human-looking" moves (avoids random king moves in the middle of long sequences, deliberately selects 2nd or 3rd computer option that seems or natural if the engine considers them barely worse). at that point, high level players don't pick up on it, but chess.com's analysis still can.

The answer is probably - you can’t. And that’s not good enough. So people are running in circles and trying to come up with fair play solutions for a new era.

if you can't, you're in trouble (and by you, i mean competitive chess). from what i've heard, a significant % of gms (less than half, but more than 10%) have cheated in online chess at one point or another. not to the extent of hans in some cases, but still cheated. if we suspect it's trivial for a significant fraction of otb chess play to be cheated as well, it creates selective pressure for more cheating and a lot of distrust.

if you are told that 3 people were confirmed cheaters at the event, you might reasonably guess one of them to be hans. but that's just one of them then, and while it's more likely to be true, it isn't guaranteed. on top of that what are you going to do to id the other two?

I do want to say, after having ripped Magnus for ripping Hans publicly-

A lot of casual fans of chess decided, apparently without actually following chess very much if at all, that Magnus is complaining about Hans because he lost a game to Hans.

No.

it isn't just casual fans. i've seen at least one gm muse on whether magnus would have withdrawn had he defeated hans rather than lost to him. he might have similarly angled against cheating broadly, but it's fair to suspect if he'd have handled it differently.

He's probably the single most gracious and sporting world champion chess has ever had

that might be true if you're talking in the context of only the all-time greats/multi world champions, but carlson doesn't strike me as particularly more gracious than anand. i'm also not sure when considering players like tal.

he's been fine for the most part, although comments like his most recent title challenger being "good at playing bad moves quickly" make me wonder if he's actually more gracious than some of the others.

i agree it's worth of investigation though. for hans specifically, but also for the future of otb chess generally. if people start believing that getting away with cheating is feasible/unlikely to be caught and stopped, it's going to be increasingly hard to take the game seriously.

on a side note, one type of cheating i suspect is at least somewhat common and is virtually undetectable by chess.com is just having book moves available in a way where you don't have to click off the browser, and can see them by just glancing over to 2nd monitor w/o clicking on it or down at a piece of paper. there are significant difficulties with catching this:

• unlike random engine strength moves later in games, book moves (or punishes to bad responses to book moves) are reasonably believed to be memorized (aka many players will have done so), and will look the same at various skill levels so long as you're still in a relatively common line.
• if you don't click on anything else, there's no way to differentiate checking book move vs thinking or attempting to remember the next book move
• there's no collusion required, so no getting caught that way
• it won't have the telltale "x amount of time between moves", because especially as a player continues to use the book moves, the player will remember some of them. player also won't need to mirror moves into an engine, which i think is what makes the common cheating method so obvious.
  • time spent thinking would spike up once deviating from the book lines...but again this should be common/expected
• im and even gm get caught by opening traps sometimes, especially in faster timer formats, so somebody throwing off a cheater with atypical moves in the opening again wouldn't look very suspicious. this type of cheater would lose to hikaru's bong cloud opener, but so do most legit players below the upper echelons of blitz etc.

it obviously has multiple limitations, too:

• sooner or later, you'll get out of "prep" (lol) and have to play own moves, cheater or not
• good players have memorized many/most of the opening book lines, depending on skill level, completely nullifying the advantage
• while this is a trivial way to cheat online, people will probably notice if you try it otb lol

i wonder how high book cheating would take someone above what their skill level would be otherwise. probably over 1000, no way to 2000. not sure between those.

 

[ywhtptgtfo]

on a side note, one type of cheating i suspect is at least somewhat common and is virtually undetectable by chess.com is just having book moves available in a way where you don't have to click off the browser, and can see them by just glancing over to 2nd monitor w/o clicking on it or down at a piece of paper. there are significant difficulties with catching this:

It doesn't take much imagination to circumvent a cheat detection engine that only takes browser inputs. All it takes is a phone, a second computer, or an accomplice. For tech monkeys, they can even write their app to just automatically mirror your move and show the next move from the engine without a single click.

I suppose this can be made a lot harder if screensharing and turning on webcams with audio are required. But still doesn't address the possibility of people using anal beads to cheat

 

[Kyriakos]

Tbf, the only people who spoke that much of "anal beads" seem to be web meme-lords like that bootleg version of PDP.
It is making it even more of a laugh and despite the momentary increase in interest by non-chess enthusiasts, it will create a worse environment for new players.

Established players (super GMs etc) already aren't much into playing GM level and below in rated matches, since they risk losing significant rating points even by drawing (which is very possible).
One of the few things Hans said after he defeated Magnus in that ott game, was that Magnus must be really embarrassed to have lost to an idiot like him Well, apparently he was more embarrassed than Hans originally imagined.

 

[TheMeInTeam]

It doesn't take much imagination to circumvent a cheat detection engine that only takes browser inputs. All it takes is a phone, a second computer, or an accomplice. For tech monkeys, they can even write their app to just automatically mirror your move and show the next move from the engine without a single click.

the algorithm will likely pick it up if you keep making engines moves even after book moves are done, because it will notice you're playing engine moves. i think what i said will only work in the context of book moves, because you can't really distinguish between memorized book moves and engine moves yet. even low-talent people can, in principle, memorize stuff, so it wouldn't seem amiss if someone suddenly could do book moves. in contrast suddenly playing 600 strength higher and getting to 2400 would look very suspicious.

one side deviating from book moves in a disadvantageous way can lead to a material disadvantage the overcomes what would otherwise be a skill difference in the other direction. if you start involving an engine after that, i expect it becomes much harder to avoid detection, because you'll have to disregard engine moves frequently enough and with enough atypical patterns to engines to prevent chess.com's algorithm from noticing. there are probably people out there who could manage that, for a time.

 

[Askthepizzaguy]

what is the % accuracy for town lynchings in mafia, even by the best players at it? from your experience, what happens when people are using a name or behaving in a way that already predisposed people to not like them, in terms of informing that accuracy?

It's not great.

A sample of my games from Jul 2017 to Sept 2021 shows me accusing / voting 50 guilty people in 19 consecutive medium sized / regular speed games.

However, I accused innocent people in most of those games, and I think a good third of my accusations were duds, minimum.

That's also probably one of the higher accuracy ratings for guessing based on just cold reading whether people are lying or not. It's obviously not an exact science by any means, despite my scientific approach to the game.

People are people and sometimes they behave like they're guilty when they're not, and innocent when they're guilty. Sometimes liars are good, sometimes the innocent just look freakin weird.

It's always squarely the fault of the accuser when you get it wrong.

For the game of mafia, that's a good haul. For accusing people in real life, getting it wrong that often would get you fired, hopefully.

The comparison to mafia games is good because it shows, even when you're really, really good at cold reading people, you are going to be wrong.

Thankfully moderators and admins have more tools in their pocket than just guessing, for catching cheaters or people breaking the rules. But for someone like Carlsen who is not a professional chess cheat investigator, let's give him the benefit of the doubt that he's the absolute best amateur cheat detection you can find, simply because of his amazing experience.

I'd guess if you pitted him against other GMs and IMs and randomly had those players cheat, over the internet, he'd falsely accuse some of them of cheating. Correctly nail several computer cheats, but also, not be 100 percent correct, or even very close to 100 percent correct, just catching cheats using his experience and his gut.

I'm considered "good" at this type of guesswork. It's fine, in a game. In real life, those numbers are terrible. It's why our justice system has a lot more checks and balances than "the prosecutor is pretty sure."

Thank goodness.

When I was moderating on one site, and the whole staff was sure X was doing something very bad, but we didn't have proof, the admin admonished us for "playing mafia". It took the admin using their admin powers to prove us right for something to be done, but we learned the lesson well.

Guessing and suspecting is not the same as knowing, even when you're right. Keeping that in mind made me a better mod, a better mafia player, and if I should ever end up on a jury, a better juror. We don't guess, when it comes to real life. We find out, and then we know, and if we don't know, and cannot prove, then they're innocent.

That's the rule. It's a good rule.

i also worry because even geniuses aren't correct often enough to have good confidence. even less so when people are already predisposed to think the person in question cheated.

I agree with you there.

One of the big problems with being skilled is you do get overconfident. And the more overconfident you get, the worse your guesswork gets.

Actually being good at guessing means taking that confidence and setting it aside and being as objective as possible. Because Carlsen isn't someone who analyzes chess cheats or is a professional anti cheat kind of person, he probably doesn't have the experience with guessing wrong and having that blow up in his face.

Mainly because he hasn't publicly made a lot of guesses that people are cheating. Which is good. But it also means, he's got some degree of inaccuracy built in, and he's got biases, and he can be overconfident.

Track record of cheating plus he beat me, and I'm the champ? Sure, that's going to bias the objectivity.

It's why someone else, someone more objective, who actually knows how to detect cheating, should be brought in. Everything I've heard suggests an investigation was carried out.

If they come back with nothing, then that's gotta be the decision.

it isn't just casual fans. i've seen at least one gm muse on whether magnus would have withdrawn had he defeated hans rather than lost to him. he might have similarly angled against cheating broadly, but it's fair to suspect if he'd have handled it differently.

Yeah, I heard Ben Finegold say pretty much the same thing. That I am fully on board with. There's really no reason why he should have withdrawn after the tournament started, and even resigning on move 2 was in poor sport because it gives Hans an advantage that Carlsen did not give any other player.

Both withdrawing from the one tournament and resigning on move 2 and not withdrawing from the other tournament are decisions Magnus made that I can't back, even if I think there's a big chance he could be right.

Ben Finegold also made the excellent point that, even if you assume Hans cheated, write it down in ink, that's what happened, Magnus would still have been wrong to withdraw from a round robin and he would have been wrong to resign on move 2 against Hans and he would have been wrong to single handedly make public allegations against Hans.

Like I said, he's not the law. All of those actions wouldn't be right even if his guess was right. And it still hasn't been established that he's right, and it was checked, and no evidence of cheating over the board was found.

That's why this whole thing is driving everyone nuts. There's no party here that looks very good.

Chess.com waited an awfully long time to check Hans' games from 2020 and earlier for cheating considering some of those were prize tournaments. The decision to check seems prompted by the recent controversy. Both things are bad for chess.com and the people who use it.
Magnus never should have withdrawn mid tourney, shouldn't have resigned, shouldn't have made public unfounded allegations without proof
Hans should never have cheated online, lied about it, and cheated some more.

Chess fans lined up to support Magnus and give him a pass for everything he did, others lined up and gave Hans a pass for everything he did.

Other GMs were engaged in Chess-related mafia-like speculation, cherry picking games and guessing that Hans must have used an engine on this move, without knowledge of statistics, without checking all his games, without being experts in cheat detection, and accepting tips and donations from their viewers while engaged in this type of speculation.

that might be true if you're talking in the context of only the all-time greats/multi world champions, but carlson doesn't strike me as particularly more gracious than anand. i'm also not sure when considering players like tal.

Yeah I'm mainly "scoring" it in my opinion my number of unsportsmanlike lashouts at opponents who won or made allegations of cheating that were not proven true, or in the case of older WCs, not granting matches to their strongest opponents (as was the style at the time)

And the score is like, golf scoring. Each instance is a point and it's always bad. Everyone starts at zero, etc.

I feel many were Carlsen's equal, but the latest drama puts Carlsen at one big boo boo that I'm aware of, over a pretty long reign as champion. Anand and Tal, I've not heard much drama about them personally, and I'm a big fan of Tal.

I am not completely informed in all details on chess drama, so I may be missing other examples, and am open to hearing them about Carlsen.

Mainly, I've followed Carlsen for a while, and seen him lose games and matches and so forth against other chess players over the board and online, and I've noticed a pattern where he's very congratulatory to those who find spectacular moves or simply outplay him. I can't name another player that beat Carlsen over the board where I can recall negative comments made about that player.

Is it genuine or not, I don't know. But he sure appears enthusiastic in praising other chess players to me.

If it was about losing to someone in general, that pattern of behavior would have shown up in his attitude before now, a lot. Instead, I've seen the opposite.

Take it with a grain of salt because I'd have to have seen Carlsen's reaction to every chess match played to be able to fairly rate his behavior, but from what I have personally seen, over the past decade, is not that pattern.

If I'm wrong I am curious to know because I don't want to defend a guy's reputation if his reputation should be worse than I am claiming.

Oh, and on a personal note:

@ TheMeInTeam I still enjoy watching you demolishing the AI on Civ IV in your old videos from time to time. You and AbsoluteZero are the folks I've most enjoyed watching play Civ IV.

I know that's very old material but just so you know it still gets enjoyed by peeps!