Checkmate: Cheating Detection

https://joeposnanski.substack.com/p/checkmate

by CapitalistCartr • 4 years ago

31 62 4 years ago

joeposnanski.substack.com

tobr• 4 years ago

The HN title is wrong, this is not about cheating detection.

jrmg• 4 years ago

It’s a fascinating article though!

CapitalistCartr• 4 years ago

It's what Bruce Schneier called it because of the part about sprinting. And the one-word title was clickbait.

incomingpain2• 4 years ago

To me this Hans Niemann thing looked more like a fake publicity stunt. I guess he admitted to cheating a few years ago. Whatever, didn't mean in the game he cheated. Yes he outright lied saying he had prepped for magnus but magnus has never played an early move.

Hans is probably hitting stride right now having been going fulltime on chess for about 8 years. Cheating likely not necessary, flipside, streamers having to think and deal with the stream which usually results in streamers being underrated.

Levon's game changed my mind. This is exactly what computers do sometimes. Some odd move that is unexplainable and then they blow you off the board. This is blowing levon off the board in a njadorf... he's the njadorf expert. No I'm now in the boat he's cheating now.

bombcar• 4 years ago

Whatever happened it's certainly brought publicity to chess on HN heh.

roflyear12• 4 years ago

He didn't lie, Magnus played a game like that before?

Levon blundered, mate. Clearly so.

Quote from Levon about this event: "So far I am playing so badly that any points that I get are kind of a miracle"

incomingpain2• 4 years ago

He didn't lie, Magnus played a game like that before?

Hans certainly lied by a significant amount.

There was a theory that someone on Magnus' team leaked their prep to Hans. Something along those lines may make the comment not a lie but I await evidence. To prove an admitted cheater isn't cheating takes extra effort.

Levon blundered, mate. Clearly so.

Definitely a blunder, :snirk: When playing against an engine every human eventually blunders. If you're a supergm you blunder in an unusual way. Like moving a center pawn 1 square. :snirk:

Quote from Levon about this event: "So far I am playing so badly that any points that I get are kind of a miracle"

Levon who is currently in 8th place isn't doing that badly.

Carlsen whose only loss is hans and he either drew or crushed everyone else...

mellosouls• 4 years ago

Last couple of days:

https://news.ycombinator.com/item?id=32901249

https://news.ycombinator.com/item?id=32909106

darepublic2• 4 years ago

There was recently a big 'open chess tourney' (actually https://www.chess.com/article/view/chesscom-global-champions...) to be conducted online. A lot of prize money and you could get to play against the top GMs. I laughed at it because of this very reason -- although touted as 'completely open' as an amateur you would never expect to win, and if by some miracle, like playing the lottery you happened to select all the correct moves, you will still be disqualified as a cheater. There is no point have a prize money tournament where amateur can fight GM (except for the PR) because amateur simply cannot win, but even if God possessed them for one game they would not be able to get away with winning.

pkulak• 4 years ago

You're underestimating cheat detection. It absolutely is not as simple as "this person played above their rating, therefor they cheated."

sooyoo• 4 years ago

Ironically, that's what I expected the article to be about. Discussing cheating detection. Like it's f**ing title suggests.

throwaway815234• 4 years ago

To play the GM's in that event, you have to qualify through some earlier events. And open tournaments are nothing new. They are usually played by the Swiss system, which means you play each round against an opponent whose current score in the tournament is the same as yours. That means the most suspenseful games are in the last round, between the players with the highest scores. That tends to mean GM vs GM, since to be at that scoring level you first have to beat a string of regular masters, IM's, and so on. It's not that simple. (This explanation though is slightly simplified vs the real world details).

darepublic2• 4 years ago

I find this article confusing since the way it is written it seems like Carlsen lost to Niemann twice in the same tournament, which I do not think is the case.

textadventure2• 4 years ago

It's not the case. And the two different tournaments used very different time controls, which essentially makes a different "kind" of chess. One was rapid (and played online with computers, but sitting across from each other), the second one was classical chess (much longer and considered to be far more serious chess), played over the board the old fashioned way.

georgyo3• 4 years ago

I don't play chess, and this story doesn't particularly interest me, but since it keeps coming up and I've now read several articles on it...

It seems to me that the entire chess community is absolutely ruining this guy's life on an accusation without any proof. This chess champ had clout to burry this guy with a tweet.

Carlsen won't play in future tournaments where this guy is also there and won't do a rematch. It seems like he is just sulking.

From an outsider of the game, without any proof and without a rematch, everyone is destroying Niemann unjustly. Comedians ripping into him.

This scandal has removed any interest I may have had in competitive chess not because of the possible cheating, but how the chess world handled it.

googlryas• 4 years ago

He's already admitted to cheating in online tournaments. Someone saying "Yeah I cheat in chess online but no way I would ever cheat in chess in real life" doesn't hold much water in my book.

It's also weird that if someone is good enough to beat one of the best chess players in the world, that they would also resort to cheating against much weaker opponents.

pizza234• 4 years ago

He's already admitted to cheating in online tournaments.

He did when he was 12 and 16. It's impossible to assess whether that holds value (particularly difficult because Niemann is 19). I wouldn't be surprised if other professional chess players cheated as young teenagers.

georgyo3• 4 years ago

In my teens I cheated in counter strike, maybe a total of 20 games. As an adult, I still play FPS games but have no interest in cheating.

Also cheating in online chess is a very low barrier if entry. Anyone can do it. When I did cheat in counter strike, it was as simple as wanting to see how it worked. Downloading some questionable binaries and running them.

A simile might be downloading an illegal copy of windows and stealing a car. Both are technically stealing, but stealing windows is not a predictor that you are going to steal a car.

Cheating in person so discreetly that no one is even sure that it happened is the opposite. It would have required custom hardware, discreet placement, and a lot of practice reading the data. Nothing suggests that he was a hardware engineer if any sort. How many accomplices did he have? The more people, the harder to keep a secret.

Similarly, his offers to rematch and play even naked are really doubling down. Magnus seems scared to entertain these matches.

isitmadeofglass2• 4 years ago

Remember, he only ever cheated the two times that it is public knowledge that he caught. Never before, never after, and never in the 4 years between those two events. If that doesn’t sound like the whole truth and nothing but the truth, then I ain’t no chess grand master!

alpaca128• 4 years ago

Show me one person who didn't do something stupid in the 12-16 age range. Chess.com are the only ones claiming to have evidence that he cheated more than he admitted, but they didn't show it and they have a partnership with Carlsen.

This is a situation where both parties have reasons to lie and there's no solid evidence for anything. There's no reason to do anything or blame anyone until any new information comes up.

rgoulter2• 4 years ago

Carlsen's acting with strong compulsion that Niemann is a cheater.

Withdrawing from a round-robin and losing a game on purpose are quite strong indications of this.

It's obviously a bad situation for the chess world. ... But, I think if you strongly believe cheating is going on without being addressed, protest seems an entirely understandable.

roflyear12• 4 years ago

That still does not mean what he's doing is the right thing to do.

rgoulter2• 4 years ago

"Only here’s the problem: As far as I know, there is not one shred of evidence that Niemann cheated — not only that, nobody even knows how he COULD have cheated. This was, after all, an over-the-board game, in person, he wasn’t allowed any communication devices."

There's no concrete, hard evidence of Niemann cheating.

I don't think you can go as far as "it's unfair to suspect him without hard evidence, or without figuring out how he could have cheated".

There are things which are suspicious about Niemann. (e.g. His history of cheating in online games, including where prize money is involved. During the recent brouhaha, chess.com banned his account, and stated that his cheating has gone beyond what he publicly stated).

roflyear12• 4 years ago

chess.com is garbage and has done worse in the past. Don't take their actions as adding to suspicion. They also have a financial relationship with Magnus... See: https://old.reddit.com/r/chess/comments/xjeuwt/a_few_cases_m...

heartbreak3• 4 years ago

Chess.com also has attorneys and has had to fight lawsuits over their cheating accusations before. They wouldn’t issue that statement if they couldn’t back it up.

permo-w5• 4 years ago

well this was a waste of 5 minutes. what part of the article was about cheating detection?

this is just a short summary of the Carlson-Niemann cheating accusation, with a tiny rant about technology in sport

permo-w5• 4 years ago

how accurate the study was, I don’t know, but I saw a persuasive post claiming that Niemann performs significantly better in games with live broadcasting of the moves. I can try to find it if anyone is interested

refulgentis• 4 years ago

Please! Ive been very unconvinced by the stuff I've seen so far that used statistics. Ex. the deepest post I saw was still pretty shallow and either I'm missing an implication, or they didn't say anything helpful. Iirc it left off at "if he's 'actually' the player before his rise, he's 1 in a 1000, if he's actually the player he is after the rise, he's been playing poorly"

roflyear12• 4 years ago

Good luck getting a link. It doesn't exist beyond some offhand Twitch comments.

rogerb• 4 years ago

Have a look at this:

https://en.chessbase.com/post/is-hans-niemann-cheating-world...

permo-w5• 4 years ago

here we are:

https://talkchess.com/forum3/viewtopic.php?f=2&t=80630&start...

in the post by DrCliche. I have no idea how accurate his stats are, but if they are accurate, it doesn't look good for Hansi

tromp• 4 years ago

World Athletics has determined that it is not possible for someone to push off the block within a tenth of a second of the gun without false starting.
Besides, do you know what Devon Allen’s reaction time was? It was 0.99 seconds. One thousandth of a second too fast, according to World Athletics’ science. They’re THAT sure that .01 seconds

That should read 0.099 seconds and 0.001 seconds respectively.

textadventure2• 4 years ago

We’re talking about a sport — or game, depending on your point of view — where you or I could beat Magnus Carlsen, probably the best chess player in history, if we simply mimicked the moves of the best chess engine.

Except not at all, or not anymore than anyone could "beat" the world's foremost martial artist, with a gun.

For an absolute novice to mimick 100% or even a large amount of chess engine moves, it would be exceedingly obvious that they are cheating, even by the most cursory and rudimentary of cheating detection measures and mildly versed (in chess) observers.

The serious problem of cheating in chess, is not at all that a John Doe with zero chess knowledge would rise up the ranks and become the world champion cheating. The scary prospect is when cheating is done by strong chess players, because they only need two or three tips from an engine a game, to have a massive advantage over everybody else.

Carlsen himself discussed this topic last year: https://youtu.be/VcbHmHHwlUQ

ShivShankaran• 4 years ago

Kasparov did win the overall match 4-2, but he lost a couple of games, which was pretty astonishing at the time.

I think this might be slightly wrong. The original battle was lost by deep blue but the "real" deep blue- kasparov match, the well publicized one in 1997, in which Kasparov accused IBM of having humans behind it was lost by kasparov.

The last game was the shortest ever game in which Kasparov had ever lost. I think it lasted around 19 moves.

What's ironic is that in the current St Louis match the organizers were checking if Kasparov had a computer to assist him and in the deep blue match the organizers were checking if Deep Blue had a human to assist it.

xh-dude• 4 years ago

An undercurrent here is how FIDE, steeped and mired in tradition, shepherds the sport alongside technological disruption, both in how the game is played and how it is enjoyed.

Magnus Carlsen would not be the first world champion to go to war with FIDE, and it’s surprising how recent stories don’t mention the surprising withdrawal-but-not-retirement of Carlsen from championship play here. Put together with the other puzzle pieces, ot feels like Magnus is trying to play a game that’s not on the board.

The last candidates tournaments and championship matches have IMHO been boring, over-prepared, extremely risk-averse, uninspiring stuff, and it’s not fun to watch Carlsen/Caruana/Nepo loaded like Atlas or Sysiphus. More enjoyable things happen in looser formats. So some disruption sounds good to me.

m3kw9• 4 years ago

Has anyone analyzed the cadence of the moves, any patterns there?

roflyear12• 4 years ago

Yes. Consensus is that everything looks like normal play.

fictionfuture• 4 years ago

I'd be checking his shoes... https://incoherency.co.uk/blog/stories/sockfish.html

Checkmate: Cheating Detection

Related Stories