Lichess: The free and open source chess server - https://news.ycombinator.com/item?id=32045763 - July 2022 (65 comments)
Give Nakamura a minute with Stockfish any two times of his choosing in each game, and he would have probably won the Candidates.
Heck, just give a good player a blunder alert that tells them after they have made a blunder that they have done so and it could make a big difference.
There were games in the Candidates where a player would make a blunder that would completely turn the game around if the opponent found the one move that exploited it, but the opponent didn't see it. The first player could have then saved themselves but had not yet realized they blundered so didn't. Then the other player realized what was going on and exploited the blunder.
Also the chess engines people use are accessible, you can compare what a suspected cheater does with what the cheater does and if they’re exactly the same consistently, you have a pretty strong signal.
One of the bigger tells are strange moves that set up a many move series resulting in a victory, things that humans just can’t find quickly.
That's similar to people describing how to catch other frauds, such as fake Amazon comments or bots. It's medieval 'science': They usually have no evidence of their accuracy, either false negatives (frauds who they overlook) and false positives (people falsely accused of fraud). So it's easy to say, 'this is how to identify them' - nobody will ever test your claim.
Regarding false negatives, for example, there is reason to believe that people detect only the obvious frauds, and that our detection becomes tuned for the obvious. Regarding false positives, people will cite the 'obvious' positives - e.g., some humanly impossible property - but even if they are correct, the problem is the cases in the grey area. False accusations are no joke.
Ironically, now we want a bot to solve our problems. What data do we have to say that it's accurate, or any more accurate than we are?
It was some time ago so probably their cheat engine detection, and also lichess's should give less false positives.
This has turned into a bit of an urban legend. The automated system flagged him based on both his rapid rise in rating and reports from several verified titled players.
On inspection by a human, he was cleared.
Danny exaggerates when he tells the story cause it's a funny anecdote.
This was also a very old version of the anti-cheat like you mentioned. Personally, part of the reason I prefer chess.com is their much better cheat detection than Lichess.
As for Irwin, its detection rate, and thresholds, it has fairly high false-positive rates inherent in the model.
From my experience, if you get banned don't expect any kind of due process. They aren't professional, they don't respond. Not even the legal contact on their charity.
Not everyone gets banned for cheating.
They do ban people from the lichess site for many other unprofessional reasons such as contributing to the project (when they don't like what you had to say on an issue and you weren't spamming).
I have to wonder how many similar-named accounts got axed alongside mine when they decided to go after me for what I disclosed. Its not like my github username was connected in any way to my lichess account (or that similarly named).
Needless to say, I don't use their platform anymore because its more bots than anything else, and I don't volunteer my time to people that don't deserve it.
As a side note, their lack of standard professional practices make me wonder what kind of fraud is actually going on behind the scenes.
As a business, the only arguably beneficial reason for not following GAAP+ other standard business practices is to commit a fraud.
As far as I understand, yes. In Chess.com at least (not sure about Lichess), there will be some kind of human verification for very high-level players, asking to prove that they are IMs or GMs. They can stay anonymous, but they will have the rank appear IIRC. I'm not sure on all the details, I'm very (very very) far from that level :)
About all it does is act as a nice PR/gimmick. Magic thinking.
When last I checked about 5 years ago, they were including client-based stats in the analysis (time per move, focus, and timing between clicks to pick a piece up and drop a piece).
The issue with doing that is the client is fungible, you can set any state you want locally; and so by manipulating these stats you could skew the input going into these models.
To give people a layman's short overview of the model:
It uses a several CNN Pooling Layers for feature detection and LSTM (attention embeddings) in a siamese network architecture.
CNN models notoriously fail to detect features that that are larger than its kernel size.
The LSTM embeddings can only be as useful as the features that it is trying to detect.
There were a number of problems with the model at the time, a high false positive rate, overfitting, and too much weight was being given to client-controllable input.
When the client-controllable input was held stable (constant). Certain book openings would be skewed with additional weight towards a false positive, and earlier book moves activated more often than lesser seen positions.
I brought the issues up to ornicar years ago in an issue, but they closed the issue without comment. The posts are mostly gone now.
There were a number of disagreements (mostly about certain people with admin privileges abusing their authority, not sure if it was ornicar or one of their flunkies; either way not important just completely unprofessional).
I guess it was good enough for them despite the high false positive rate (driving account churn).
The relevant section of the video on YouTube is: https://www.youtube.com/watch?v=LZgyVadkgmI&t=1080s
Of course the system could go ahead anyway, displaying a fake rating - within the search criteria - for this GM / bot, but given that there are always false positives (and it would have to be done for merely suspected accounts, as you said yourself), it would essentially mean deliberately pairing some innocent players against, well, effectively cheaters. Which kind of defeats the core purpose.
Once you get to ~1900 FIDE level of play and rework your repertoire for anti-engine strategy you realize just how much rampant cheating is happening.
This continues until at some point you have to ask yourself, do you want to waste your time volunteering to be a victim to satisfy someone else's severe mental illness.
They get their dopamine hit for psychologically abusing you with a tool they didn't create, often with some amount of delusion to justify it.
What do you get out of playing a game against a Super-GM rated computer? Nothing but time theft.
The current state of the game is without a guarantee that the game is between two humans, the game isn't actually chess. The providers for online chess don't make that guarantee.
So based on the incentives, the game you play on their services is more likely to be not-chess but they still call it chess. Deceitful bunch.
Specifically, in this context receiving outside assistance directly or indirectly, that in any way impacts your decisions made during the currently on-going game.
This includes more than just lines and moves, such as the exact evaluation score of a given position.
the latest ("real") commit is from 3 years ago
Though ofc you're certainly right that this isn't a recent development at all and at the same time it's certainly also not perfect yet. And I doubt it ever will be. I don't think cheating in online chess is something that can ever really be solved completely.
I don’t know why people do this. It’s not like it makes them better at chess.
I have no idea why people do it, it's not even like video game cheats where it gives people a advantage but you're still in control - you're literally just copying.
The only solution to this problem I'm aware of is developing/becoming a part of a community of people you trust. Of course, that's hard and comes with plenty of limitations.
Lastly, please don't get your expectations up. This is a hugely unsolved problem on lichess and the history of its and chesscom's efforts do not inspire confidence. I noticed this repo has no updates in the past three years, during which lichess has maintained its reputation of having tons of cheaters. Chesscom has a team and probably more tools (proprietary, larger company, etc.) but the best case scenario still involves cheaters getting several games in, wasting your time along the way.
Having now played nearly 2000 games on Lichess, I'm pretty happy with their ability to weed out cheaters. I've very rarely had cause to suspect my opponent wasn't playing fair. Maybe once or twice someone with a lower ranking than me seemed to be a bit too good --but then we all have our flashes of inspiration, as well as our off days.
However, on a couple of occasions, I've logged onto the site to see a notification that I've had some ranking points restored, as a previous opponent had cheated. But, unfortunately, Lichess doesn't tell you which opponent or which game, so I'm still none the wiser.
PS: Agree with you about the people who run the clock down, when they're losing. Really annoying. When this happens, I usually switch to another browser tab and read intarwebs instead, switching back to the Lichess tab every 5 mins or so, to see if they've made a move yet.
Yeah, unless you want to play a longer game than a 5 minute blitz.
The rating of a player will reflect their strength, regardless of whether they are an unaided human, a full computer, or a human with some heuristic to consult an engine. Whatever the player is doing, their rating will reflect their average strength, and the website matches us to players of similar rating. Whether or not a player is pure human, pure engine, or centaur, we get matched to a player that we have a roughly even chance of beating.
Pure cheaters will quickly skyrocket to the top of the ratings and I will never see them. Hybrid cheaters, who do not have a perfect rating, have a probability of losing because their heuristic to switch to an engine is imperfect, and their rating reflects this imperfection.
Whatever the player is doing, their numeric rating reflects their average strength, and if you have a similar rating you have roughly even chances of beating them.
This is like boxing in the lightweight division, but your opponent can, at will, use Mike Tyson in his prime to hit you in the face.
Unless they cheat in unrated games, or discard their account and create another before the rating ascends too high (not necessarily for that purpose, but to avoid detection).
Such a person often ends up making an absolute fool of themselves from the perspective of someone with a little bit more self integrity. But maybe it is worth extending a little compassion, after all a lack of compassion is probably how they ended up like that anyhow.
Happened to a good friend of mine up in Seattle. Good fella, incredibly smart, graduated top marks, a bit ostentatious sometimes but you can tell he really wants to do right by others in spite of his sometimes high falutin tendencies.
Anyway.
Lost a game of chess to his blue collar dad, guy couldn’t handle it. Fell into a hole of depression that lasted for at least an entire episode of Cheers
* Overperforming kids (whether smart, in athletics, etc). They don't have to confront failure and work hard like everyone else. They get a lot of praise and self-worth gets tied to it, and it all feels kind of fake/too easy. Then, one day, the work ethic and tolerance of failure everyone else has been learning become critical...
* People raised in abusive or quasi-abusive environments, where success at any cost, appearances over all, etc, are emphasized. The ends justify the means to get the required outcome.
To give the opposite answer, why you should not cheat. Chess is fun if you play against players of similar strengt. Mostly a score of around 50% will be what you get. If you have a 1400 rating, you can have fun games against 1400 players. If you cheat and go towards 2300 rating, you will still have a score of 50% against other 2300 players. But all the fun of the game won't be there, since it is just cheating. You will lose something (fun), but not really win anything within the game itself.
Or simpler said, in a long gone past I got bored with playing Doom. I hunted down cheat codes and got bored even faster.
This is a life hack if ever you find yourself too distracted with a game (for me, Rimworld). Mod / cheat to the point of ruining the game, and you'll likely never play it again. That being said, I do not advocate doing this in multiplayer games.
So yes the turnover may be pretty high, but it doesn't make much difference from the perspective of a regular user.
You can, but the fun will be different because the challenges are different.
The cheaters aren't playing chess, but some other unnamed game
the game might a big part of their life, and being perceived as a good player puts you higher in the pecking order within your friend group
I think something like chess is different; cheating in video games feels more similar to hacking around some software just because you can and it's interesting.
I have no idea why you would do it on a ongoing basis. It would be so boring, all you do is transcribe moves back and forth.
What does winning do for someone when they didn't actually achieve mastery of the game?
