I attribute part of the reason why this scandal has reached as far as it has is that Niemann was able to retain his reputation and enter professional OTB tournaments up until the Sinquefield Cup.
Nobody knew about any of this until it was alluded to vaguely, and then implicitly after Magnus was defeated OTB
The only disappointing thing was the focus on how fast Hans' rating soared after he hit 2500, and how much it rose between ages 11 and 19.25. These thresholds were very cherry-picked to make Hans look bad, which you didn't need to do. For example, Hans's rating lingered at 2450 for two years and then popped. If you charted ratings rise starting at 2450, Hans would be on the other side of the chart!
To be clear, as there is no equivalent to a physical impossibility here, I am not claiming that the rate of rating increase is conclusive; I am saying it seems to be a legitimate concern regardless of how reasonable the broader averages are.
And also... who cares? there are no stake, just playing against random people, who cares if they cheat as long as you have fun playing. It's less fun for the cheater, but doesn't impact non-cheater.
I would be upset if someone was inconsistently cheating, and so I'm playing someone nominally at 1400 but for this game they're actually playing 2100.
Of course for true tournament play, the rules are the rules, whatever they may be, and they must be followed for it to have any meaning.
So who did chess.com share those with?
Ie, their secret sauce.
I basically made the argument that, in any sport, when a player does statistically much, much better than their previous performance would predict, that in and of itself should be considered evidence of cheating - perhaps not conclusive evidence, but definitely evidence warranting further investigation: https://news.ycombinator.com/item?id=32990022
> All this talk of "Carlsen accused him of cheating with no evidence" reminds me of the blowback against some athletes in the 70s and 80s who accused rivals of taking PEDs "with no evidence".
> Sometimes the evidence of someone doing monstrously better than can be expected by their history is sufficient IMO. I mean, look at this article about swimmer Shirley Babashoff [1], dubbed "Surly Shirley" at the time by the media, for suggesting the East German women were on PEDs in the 70s. Nowadays we look back on those images of the East German women, looking more manly than any dude I've ever seen, and wonder how we considered with a straight face that they weren't on a boatload of drugs. Similarly, it completely baffles me how any sane person can think that Flo Jo wasn't on PEDs in the runup to the 1988 Olympics - her 100m dash record still stands today.
> I'm not saying Carlsen went about it in the right way, because now Niemann is basically in an indefensible position, but I'm also not willing to quickly dismiss it because Carlsen has "no evidence".
Besides, Magnus genuinely has never seemed like the guy to get petty and up and throw a fit, he’s lost plenty of times without doing such.
The way Carlsen described his suspicions reminded me of "connoisseurship" in the art world. Now that's a "skill" that's not as important as it once was but once the science has given its results and there are still no firm conclusions, connoisseurship is all you have.
More than that, Magnus is a very fierce competitor and he doesn't withdraw from tournaments. He's 31 and this is his first withdrawal AFAIK.
People aren't coins, with each toss unaffected by previous landings, but at the same time, not giving evidence shouldn't be considered evidence itself.
There's a rather large contingent of people who use it as a synonym for "proof", thus demonstrating their complete lack of understanding of the very concept even as they think they are looking smart. There's a lot of people who think that it is somehow the responsibility of someone like Carlson to have unambiguous proof that is often physically, legally, or morally impossible for them to have. Not that the Kantian imperative is the be-all, end-all of morality or philosophy, but it's still a useful idea, and if one runs this memeset through that grid it quickly reveals that you can't use this as the bar people must meet. It creates a society where anything that can be hidden, often not even hidden particularly well but just hidden at all since the standard is "proof", will be gotten away with. This is impractical.
(Although there is an extent to which we do indeed live in that society.)
There's another smaller contingent (from what I can see, the former is the dominant group) who thinks that "evidence" doesn't have to be proof per se, but that for something to be "evidence" it must be single-handedly capable of meeting some bar, be it "preponderance of evidence" or whatever is being used. These people are blind to the fact that it is often possible to produce ten pieces of evidence that are individually weak, but together are too improbable to ignore. While this case must certainly be treated with care since it is sensitive to how well done the Bayesian-style analysis of their probabilities is, it is perfectly reasonable to use such evidence to come to conclusions. This contingent will often try to defeat the evidence in detail, decrying each individual one and downplaying it, and completely failing to address the totality, be it either as a method of being disingenuous, or, I am convinced, often just lacking the intellectual firepower necessary to address the totality of evidence holistically.
It's important not to fall into that trap because I think in the real world, this is how many very important questions can be settled... and I don't just mean big philosophical or societal questions, or even news of the day like this, but even little personal things. The signs that your partner is cheating on you often come in like this... you may not catch them in the act, nor even receive some individually huge clue, because there's one or two humans intelligently trying to deceive you. Such situations often end up giving off a lot of little clues, each of them individually explainable-away, but the totality pointing to the truth. Or you can determine some new direction your company may be trying to take by a pattern of little changes long before they announce it publicly. Any number of things in the world where you can carefully assemble a lot of little things to come to some reasonably strong conclusion.
Back on topic, the combination of the fact that Neimann was known to have cheated before and the accusation of the current best player in chess, who was kind of going out on a limb to do so, was quite strong evidence that Niemann did cheat. Neither individually would be enough for me to come to that conclusion strongly on its own, but the combination was compelling. I considered that enough proof for my personal opinion. Proof enough for authority figures to sanction would require a higher standard and I support and applaud them for being more careful.
We shouldn’t have to tolerate this in sports, which are an escape from the real world.
Allowing people back into the sport swings the EV math heavily in favor of cheating if the penalty isn't massive given that the chance of getting caught is so low (as long as you know what you're doing). The only way to make the EV of cheating negative is to make the sanction very, very bad. Losing all of your future earnings from the sport is a good way to do that.
I used to run Magic: the Gathering tournaments, and there was a tremendous amount of "minor" cheating - forgetting the rules when it benefits them, shuffling in suspicious ways, peeking at opponents decks, etc. Many competitive players even openly admitted to doing this. Even if a tournament official could call them on the cheating and disqualify them (which was frowned upon without hard evidence), they would likely not be suspended from sanctioned play at all unless the evidence was overwhelming. Several famous cheaters did it many times and got caught several times. Minor cheating was very common as a result.
The benefits were in the thousands and thousands of scholarship dollars.
I cheat now in my employment. I work three full time jobs remotely and do the bare minimum in each. The risk is getting fired (and if I only get fired from two of the jobs, I am still ahead of honest work). The payoff is decades taken off my working life.
I'm personally a bit frustrated with the ever changing standards for adolescents. They are as responsible or naive as people want them to be for whatever their bias calls for. (not saying you btw, just in general).
If you do something bad when you are 17.9 should we wipe the slate clean once the odometer rolls over to exactly 18.0 where for some reason that age creates a solid barrier where the person emerges like a chrysalis and all their sins are washed clean?
And as a 50 year old, the difference between a 16 year old and a 19 year old are not very big. An unfortunate fact is that if you fuck up pretty big when you're 16 that people aren't going to trust you very much when you're 19. You need to do the time to build up more collateral. I've seen people who were assholes when they were teenagers change, but they didn't wake up on some magic birthday a new person. They were still assholes in their early 20s but their trajectory was such that by the time their early 30s came around they had changed themselves.
Maybe when he's 24 or 30 we can try again.
My guess is that the more decentralized a sport is, the more likely it is for cheating to occur and go unpunished. Chess is unofficially becoming a decentralized sport since the pandemic due to the shift to online playing. Even if some organizations claim to be in power, there is only so much they can do. Banning cheaters permanently may not even be possible.
*Of course this varies a lot by sport, gymnastics careers are obviously very short, a golfer's career may be much longer.
Unless their chess.com scores feed into their FIDE ELO scores or something?
Chess.com isn’t the yahoo chess web app or whatever, there’s more at stake there than “gamerscore”. Some of these 100 games were in online tournaments for cash prizes.
Just because it wasn’t a FIDE run tournament doesn’t mean it doesn’t matter or that no one cares. The chess.com Pro Chess League they suggest he cheated in 12 games of had a $100,000 prize fund.
Still it sounds a bit harsh to me that a child that cheated online can never in his adult life participate in chess tournaments.
If he had cheated as an adult I would have a different opinion.
The penalty for cheating in most major sports is way more lenient than you think. Most leagues will suspend you for a handful of games in the first instance. In the NBA for example you can be caught three times before being suspended for one season.
Especially since there are many others who don’t cheat that can take your place in a competition.
What is this alternate reality you're in, and what is your list of "everything else"? Citations, please.
The first result when you google "Are chess players athletes" says no [1], but I realize that this is more of an opinion piece. I would be curious to hear what more members of the competitive chess community think of the designation as athletes.
Edit: Upon further googling, I have learned that the IOC recognizes chess as a sport. Reading up further on how the matches last for 7+ hours, and how important physical conditioning is, I think it's totally valid to refer to chess players as athletes. In different sports there is wide spectrum of physical and mental demands - and I think chess just falls on the incredibly-demanding-mentally-but-less-demanding-physically end of the spectrum.
[1] https://herculeschess.com/are-chess-players-athletes/
[2] https://olympics.com/ioc/recognised-international-federation...
It's recognised by the International Olympic Committee - https://olympics.com/ioc/recognised-international-federation...
A common definition is: "Sport pertains to any form of competitive physical activity or game that aims to use, maintain, or improve physical ability and skills while providing enjoyment to participants and, in some cases, entertainment to spectators." https://en.wikipedia.org/wiki/Sport
I don't think it worked for him.
https://github.com/IceWreck/Page-Visibility-User-Script
I made this a while ago.
Or just use another computer.
and yeah you could argue that if you're not into web dev it's not hyper obvious that websites are able to know when you lose focus on them, but I'd argue that it is fairly obvious. there are indications that should make it clear. facebook flashing notifications in their tab title until you click on them, to take a simple example.
I imagine there is one chess cheater reading this that will use your repo soon.
If he accidentally showed a picture of his second computer on the same desk during e.g. a stream it would be akin to guilt admission.
They can really only think in front of them. If I put information into this program, I get information out. They don't think that applications can monitor their own meta-state. Or the state of other applications.
So I'm not terribly surprised that he thought running the engine in another browser window would have been sufficient. He might have even had it open in "incognito mode". And since it's incognito, it can't be detected, right?
When there is money in the game, there is incentive to cheat.
> The report says dozens of grandmasters have been caught cheating on the website, including four of the top-100 players in the world who confessed.
There are probably smart cheaters already playing who are able to evade detection.
Even during the Carlsen-Niemann game it was meta-factors that initially clued Carlsen in. Niemann was playing without any significant effort or tension, in spite of playing in a game where he was outplaying the world champion. And after the game he was unable to explain his own ideas, proposed ideas that were simply losing, referenced games that did not exist, and was generally (relative to the class of player here) clueless. None of that final section is definitive proof of cheating to say the least, but it helps create a probabilistic profile of a player (and a game).
The point of this is that even a computer that played human-like (which I would argue will not happen for the distantly foreseeable future), would be just one factor among many in busting cheaters. I expect this is why Magnus was also initially reluctant to directly accuse him of cheating. He felt he was cheating based on the meta-factors and probably got folks more capable than himself to evaluate the technical factors, and when that also came up as a redflag - yeah, the dude's a cheater.
Ok, but what prevents the helper to communicate the difficulty or the number of minutes to think-pretend as well as the move itself?
Everything that can be measured can and will be gamed. That's why anti-fraud units are so secretive.
You can tell how experienced someone is based off the gun they use (some are stronger than others), whether they use cover or just run out into open spaces and shoot, how they move, whether they use 'gadgets' like grenades, and so on. A lot of novice players don't even use the sprint function to run.
When someone who literally just walks around the map but can laser everyone with headshots (which have a significant damage multiplier)? They're cheating.
Even a strong player can benefit from consulting a computer. Chess games can win fail based on a few moves.
A strong player would only need to consult the computer on a few moves to get a considerable advantage.
Aside from Niemann's case, how is it strategically beneficial to a chess player to provide the "inside scoop" on his plays?
You're presupposing incompetence, but another explanation would be a deliberate strategy to throw off future opponents.
Carlsen was making mistakes. That wasn't his best game at all. Are we sure we aren't talking of this because someone's ego was hurt?
> And after the game he was unable to explain his own ideas, proposed ideas that were simply losing,
That doesn't mean anything at all
> referenced games that did not exist,
That did exist close enough to the period he mentioned. Remembering the position and analysis is necessary, remembering when exactly this position happened and even between whom exactly is utterly useless.
> and was generally (relative to the class of player here) clueless.
He didn't make a clueless impression to me. But I'm not Carlsen's fanboy whos accusations can cloud my own reasoning. If I was, I'd probably believe that Niemann is a proven cheater and would look for facts to confirm that bias.
From what I understand, Niemann got into trouble because people thought that he wasn't able to adequately provide the analysis i.e. the reasoning behind some of his own moves. You'd need a live auxiliary AI to tutor the cheater in how to explain why a particular move was made.
At one point he described one of his own moves as "a weird move" without offering any explanation, sounding more as if he was observing the move rather than being the one who had actually analyzed it and chosen to play it!
That is harder, almost as hard as playing for real, but doable. Much easier to just be a mechanical turk for sharkfish or whatever it's called.
If it's indistinguishable from human play, then there is no advantage to cheating with a chess engine. The point of cheating is that the chess engine is stronger than human play and will give you mistake-free moves that put you in winning positions. If all your moves are equivalent to human moves, then you're playing no better than a human, at whatever level that is, let's assume GM 2700-2900. So what advantage does a human GM get from doing that?
And if it now turns out that he lied in his confession, too, then that's a really bad look w.r.t. his trustworthiness.
The article doesn't say that Niemann's admissions to Chess.com were about cheating in prize-money tournaments, nor the other disputed facts. The spreadsheet of incidents they show us isn't what Niemann admitted to cheating in, but was Chess.com's internal anticheat flagged -- we can tell because they label it as "suspect games" and it uses qualifiers like "likely". The inferences that the cheating was for real money prizes, or at an age older than 16, or on for-profit Twitch streams, are drawn from from this list of suspected games.
We don't currently know what facts Niemann confessed too: it's not public whether the facts Niemann is allegedly lying about overlap with the facts Niemann admitted to in writing in 2020. WSJ may have evidence that's dispositive on this point (i.e. those Slack texts), but they haven't printed it yet.
Computers have “nearly infallible tactical calculation,” the report says, and are capable of beating even the best human every single time. The report says dozens of grandmasters have been caught cheating on the website, including four of the top-100 players in the world who confessed.
That being said, if they have literal screenshots of the discussions between Niemann and chess.com admitting to cheating and appealing the ban, those seem like smoking guns in addition to all this other analysis
this isn't to say that I don't think Niemann cheated. I think he probably did, just that chess.com is getting great publicity from all this and it would be un-shocking for them to want to string it out a bit longer. it would be poor business not to, in fact
Otherwise we will probably just dismiss you as hardly impartial ...
We dont generally place full trust in online job interviews so why lower the bar to "honor system" when it comes to the most cheat-friendly competition in the universe?
Also, there has been history of people cheating in IRL games by taking many bathroom breaks and looking up solutions while on break.
Can the "over the board" cheating potential be reduced with a 5 minute "tape delay" of broadcasting the game? Is that enough time to thwart the influence of an external signal?
Seems to me the only way this person can redeem himself is through over the board play under strict conditions ("Come dressed in shorts, a t-shirt and sandals").
But I don't know if how long they're allowed per move, and if 5 minutes is enough time to thwart external influence.
The Sinquefield Cup (the tournament where the drama started) added a 15 minute delay which would be much more noticeable and less forgiving.
Or is it that they only need to cheat at a few points where taking more than 5 min wouldn't be abnormal at all?
Also, if your cheating device allows you to somehow input chess positions, then you wouldn't even require an external signal. Though it would be extremely impressive if somebody could pull that off.
If blitz type game where players have <5 minutes to decide a move, then yes, delayed broadcast might be effective. Other game types allow for >5 minutes per move so tape delay would be ineffective since cheater could just stall.
Cheating type also matters. There's external help (friend in the audience communicating 1 or more bits of information via auditory, optical, radio, or some other signal), and there's also internal help (raspberry pi zero + battery and pressure sensor embedded in your shoe or something). There are so many ways to cheat that it's hard to enumerate all of them let alone prevent all of them.
Given sufficient notoriety and money involved, it would be possible to just hire someone to essentially run a training model of alpha zero against moves specifically selected to be likely to be made by Magnus, and then all you really need is memorization of key scenarios (which for a good chess person should be no problem) to identify the right move to make.
Why not analyze his recent and over-the-board games?
While it says Niemann’s improvement has been “statistically extraordinary.” Chess.com noted that it hasn’t historically been involved with cheat detection for classical over-the-board chess, and it stopped short of any conclusive statements about whether he has cheated in person. Still, it pointed to several of Niemann’s strongest events, which it believes “merit further investigation based on the data.” FIDE, chess’s world governing body, is conducting its own investigation into the Niemann-Carlsen affair.
It's also why there are occasional surges in cheating (or crime, or whatever) after significant instances - subsequent examination then finds other cases because it's now looking for them, but the reality is the cheating (or whatever) was always there and just not noticed.
Thats all going to change now though, and its totally possible to cheat using a second computer with an engine that will be undetectable.
I suppose new algorithms will be designed or trained to account for the user's performance history.
>The report made no conclusions about Niemann's in-person games. But it also flagged his play from six over-the-board events, saying those merit further investigation.
https://twitter.com/andrewlbeaton/status/1577380477807300626
Hikaru coverage on YouTube [2].
2013 Interview with Max Dlugy: https://en.chessbase.com/post/the-shoe-aistant--ivanov-forfe...
Also, before you ask, "if he is already at GM level, why does he feel the need to cheat?" the answer is that the stronger players and athletes often feel more inclined to cheat because they have such high expectations of themselves. Past cheating scandals in sports have proven this.
The difference between being in a top team and a lower ranked team is measured in hundreds of thousands of dollars. Riders who spent 10-15 years of their childhood training to be a pro were faced with a difficult choice - cheat and realize your dream, or stay clean and walk away from the sport.
Magnus Carlsen is a majority shareholder in chess.com.
And it’s disingenious to suggest that he’s actively trying to destroy Hans’ career because his ego couldn’t stand losing to him in that OTB game, because it’s an extremely serious allegation from human perspective.
In fact, the article tries to paint Niemann as a liar while the purported facts pretty much match what he admitted to. One cheating in a titled tournament at age 12 and multiple cheats at the beginning of 2020. He said he was 16, so he was barely 17 according to the article. That isn't a lie, that can easily happen in an interview.
If that is all that chess.com has, their behavior is extremely poor. Also, what about all those other cheating titled players who did not have the misfortune to win against multi-million asset Carlsen?
It is time for Europeans to send GDPR requests for cheating scores etc. and terminate their accounts. The risk is too great.
And yes, while cheating online is shabby, hardly anyone took online chess seriously before the big money tournaments started during the pandemic.
And that the whole chess.com affair is a side show that is exploited for streamer content and clicks. The relevant issue is cheating or not cheating in the Sinquefield cup.
https://news.ycombinator.com/item?id=19563768 ("Chessvision.ai – Analyze chess position from websites, images or video", 49 comments)
https://news.ycombinator.com/item?id=21162466 ("Show HN: ChessBoss – enhancing physical chessboards with computer vision", 36 comments)
As to how he'd cheat over the board, that's the big question. There are a couple of theories floating around, some more realistic than others, but if we knew for sure than this whole debacle would already be over.
OTB cheating: there are many possible ways. The simplest one being having a script reading the moves from the live broadcast and feeding them to an engine and then sending the info to the player. Hans' strength magically decreased in the Sinquefield cup once broadcast delay was introduced for example.
So you discount a possibility that the controversy and allegations against him had at least some effect in him playing worse in the Sinquefield Cup?
Anyway, there were studies out there that there were no statistical differences in his rating gains in tournaments broadcasted with delays and without it. Study by Kenneth Regan also found no irregularities in his play, so the only 'evidence' of him using computer help are allegations by a company that is in business relationship with Carlsen, and his 'bad' analysis in post game interview. I'm not very impressed.
This might be the solution. But then on-site audience would need to be monitored as well.
1. The driving force behind the original accusations is that Magnus felt his opponent wasn't "exerting" himself enough, compared to other young prodigies.
2. Chess.com's case is that his results are "statistically extraordinary."
3. There is a history of cheating
4. Allegations that he admitted cheating privately (though it's not clear to whom)
1, 2, and 3 could easily be cause for suspicion; however, that's not the same as evidence. The one crucial piece absent from this article is any suggestion of how he cheated.
Without providing a means, I find this piece premature and questionable. That said, I don't know anything about chess, lot alone cheating at the master level. So maybe the "how" is common sense and not difficult?
And of course, there's also this:
> The report also addresses the relationship during the saga between Carlsen and Chess.com, which is buying Carlsen’s “Play Magnus” app for nearly $83 million.
There are various ways one might cheat OTB, from taking one's phone the bathroom in the middle of a tournament (some allow this!!), to getting signals from an accomplice who is seeing the game in real time. Signals could be electronic to some device on the player, or visual from an audience member in the room. It's been proposed to introduce a 15-30 min broadcast delay in tournament games as one way to prevent cheating. Some tournaments scan the players for electronic devices - not sure how foolproof this is.
Actually, it isn't! Great chess bots have very different play styles and there are people currently studying them. It's very unlikely someone will come out of nowhere so to speak (as in, not on some amazing rise as a young child) with these types of techniques. I'm nowhere near these levels of chess players but have played competitively for my county as a school-kid and still play a couple hundred games a year so have some idea.