|
Look, I've commented this before, but chess.com is absolutely not a reliable source for any data or conclusions here. There is no independent audit and they have a financial incentive to bolster the reputation of Magnus Carlsen, who is Hans's chief accuser and whom they recently entered into a major partnership with. Everything they say can be true, but 1) they're also incentivized to omit evidence to the contrary, such as when Hans was cleared of cheating by their systems and 2) they have no business analyzing OTB games. You also don't know that everything they're saying is true. I don't know chess as well as I know money, which is why I just lurk, but the money speaks for itself. Morrow fucked around with this message at 21:09 on Oct 4, 2022 |
# ? Oct 4, 2022 21:07 |
|
|
# ? Jun 5, 2024 06:28 |
|
Morrow posted:Look, I've commented this before, but chess.com is absolutely not a reliable source for any data or conclusions here. There is no independent audit and they have a financial incentive to bolster the reputation of Magnus Carlsen, who is Hans's chief accuser and whom they recently entered into a major partnership with. Source your quotes
|
# ? Oct 4, 2022 21:08 |
|
Morrow posted:Look, I've commented this before, but chess.com is absolutely not a reliable source for any data or conclusions here. There is no independent audit and they have a financial incentive to bolster the reputation of Magnus Carlsen, who is Hans's chief accuser and whom they recently entered into a major partnership with. oooh we got a money guy over here
|
# ? Oct 4, 2022 21:12 |
|
Super Jay Mann posted:Source your quotes I'm not sure this is a joke post, but uh Morrow posted:Chess.com also has financial links to Magnus Carlsen now, so they're not a reliable source of information on who is cheating. https://www.chess.com/news/view/chesscom-playmagnus Consider that Magnus's value as a brand is about to dip in the coming year once he retires as world champion, which reduces the value of chess.com's latest investment. It will dip even further if Magnus Carlsen is a former world champion who false accused another GM of cheating in a huge scandal, so chess.com has every reason to back him to the hilt. As far as I know they haven't gone public with any other individual accounts banned for cheating as a rule so you need to consider what is different here.
|
# ? Oct 4, 2022 21:14 |
|
Fortunately, they seem not to have made any otb accusations. They flagged a few tournaments that seemed suspicious to them but admitted their lack of expertise and deferred the verdict on those tournaments to the fide investigation. That seems like pretty responsible behavior to me The public airing of their evidence of online e cheating is pretty inevitable when Hans directly contradicts what they know on the subject While the full 72 page report doesn't seem to be public yet, the WSJ summarizes it as basically "Hans lied about the degree, extent, and time frame of his cheating on our website. You should not take his statements on the matter at face value, and here's why. Here's a few things we hope the otb cheating experts investigate." It seems perfectly in line with what an investigation by chess.com can or should be able to say, to me.
|
# ? Oct 4, 2022 21:15 |
|
Morrow posted:I'm not sure this is a joke post, but uh the thing that's different here is that it's a very big, public, messy scandal in which hans directly contradicted information that they had access to. There is a strong public interest in chess.com putting up the evidence for "Hans is lying" in a way that isn't true of other online cheating cases because they didn't become matters of public interest
|
# ? Oct 4, 2022 21:16 |
|
cheetah7071 posted:While the full 72 page report doesn't seem to be public yet, the WSJ summarizes it as basically "Hans lied about the degree, extent, and time frame of his cheating on our website. You should not take his statements on the matter at face value, and here's why. Here's a few things we hope the otb cheating experts investigate." It seems perfectly in line with what an investigation by chess.com can or should be able to say, to me. Do we know if it ever going to be public?
|
# ? Oct 4, 2022 21:17 |
|
Chromatics posted:Do we know if it ever going to be public? if it doesn't become public then chess.com is startlingly bad at this and I'll jump on the conspiracy train
|
# ? Oct 4, 2022 21:18 |
|
When Hans says he cheated two times does anyone even interpret that as cheating in two games? It seems pretty obvious it means he cheated in two separate time periods.
|
# ? Oct 4, 2022 21:25 |
|
tractor fanatic posted:When Hans says he cheated two times does anyone even interpret that as cheating in two games? It seems pretty obvious it means he cheated in two separate time periods. Yeah, but as was pointed out he cheated a bit longer then he initially claimed and in more money games then claimed. The ban on chess.com seems more than fair at this point, but still no evidence of OTB cheating.
|
# ? Oct 4, 2022 21:26 |
|
the report is coming https://twitter.com/ChessMike/status/1577392255245197313
|
# ? Oct 4, 2022 21:27 |
|
Rumor is WSJ got two hours of exclusivity, so about another hour.
|
# ? Oct 4, 2022 21:29 |
|
neaden posted:Yeah, but as was pointed out he cheated a bit longer then he initially claimed and in more money games then claimed. The ban on chess.com seems more than fair at this point, but still no evidence of OTB cheating. No one can seriously argue with the chess.com ban but their claim about his OTB Elo growth is going to require real scrutiny.
|
# ? Oct 4, 2022 21:29 |
|
tractor fanatic posted:When Hans says he cheated two times does anyone even interpret that as cheating in two games? It seems pretty obvious it means he cheated in two separate time periods. But the article shows three different years not to. Also he claimed to only cheat in casual games after the TT while 12 while this shows multiple competitive money games.
|
# ? Oct 4, 2022 21:37 |
|
Sub Rosa posted:But the article shows three different years not to. Also he claimed to only cheat in casual games after the TT while 12 while this shows multiple competitive money games. Ah, you're right, I didn't know he also claimed to have not cheated in money games. That's definitely a bigger stain on his character.
|
# ? Oct 4, 2022 21:42 |
|
Morrow posted:Look, I've commented this before, but chess.com is absolutely not a reliable source for any data or conclusions here. There is no independent audit and they have a financial incentive to bolster the reputation of Magnus Carlsen, who is Hans's chief accuser and whom they recently entered into a major partnership with. WSJ posted: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. The report says that while Carlsen’s actions at the Sinquefield Cup prompted them to reassess Niemann’s behavior, Carlsen “didn’t talk with, ask for, or directly influence Chess.com’s decisions at all.” Rensch had previously said that Chess.com had never shared a list of cheaters or the platform’s cheat detection algorithm with Carlsen. If they're lying about this then it probably tanks their entire business. Put the tinfoil down.
|
# ? Oct 4, 2022 22:29 |
|
neaden posted:Yeah, but as was pointed out he cheated a bit longer then he initially claimed and in more money games then claimed. The ban on chess.com seems more than fair at this point, but still no evidence of OTB cheating. I only cheat in mountain bike races, I would never dream of cheating in a road bike race.
|
# ? Oct 4, 2022 22:39 |
|
Salt Fish posted:I only cheat in mountain bike races, I would never dream of cheating in a road bike race. The whole point is it's so much harder to cheat OTB, which is why online chess is full of cheaters. They caught Hans tabbing away to consult an engine, and I suspect the reason they punted on the OTB analysis is because they're not sensitive enough to detect the type of cheating he's accused of either.
|
# ? Oct 4, 2022 22:44 |
|
As a member of Team Chaos, I am delighted to be able to read this after work. Will the US championship rescind their invite or will they hope public pressure forces Hans Moke to withdraw?
|
# ? Oct 4, 2022 22:46 |
|
I have my 700 blitz rating on chess.com fair and square goddammit and I will not stand for this
|
# ? Oct 4, 2022 22:53 |
|
tractor fanatic posted:The whole point is it's so much harder to cheat OTB, which is why online chess is full of cheaters. They caught Hans tabbing away to consult an engine, and I suspect the reason they punted on the OTB analysis is because they're not sensitive enough to detect the type of cheating he's accused of either. My interpretation is more 'online chess isn't real chess, online money isn't real money, and online cheating isn't real cheating' which I'm not saying is your point of view, but at least its implied in some of these discussions.
|
# ? Oct 4, 2022 22:54 |
|
tractor fanatic posted:The whole point is it's so much harder to cheat OTB, which is why online chess is full of cheaters. They caught Hans tabbing away to consult an engine, and I suspect the reason they punted on the OTB analysis is because they're not sensitive enough to detect the type of cheating he's accused of either. While not covered in the article apparently the report does look more at OTB. From the WSJ author: https://twitter.com/andrewlbeaton/status/1577380477807300626
|
# ? Oct 4, 2022 23:00 |
|
Salt Fish posted:My interpretation is more 'online chess isn't real chess, online money isn't real money, and online cheating isn't real cheating' which I'm not saying is your point of view, but at least its implied in some of these discussions. If you're worried about cheating it's going to be increasingly hard to take online chess seriously. It will only get easier and easier to cheat and there's basically no way to stop it all. Online money is very much real money, and it might even be criminal to cheat in a money game, but all this just makes it harder to take online chess seriously. With OTB at least you can imagine there will be physical security measures, but there's nothing stopping increasingly sophisticated engine use online.
|
# ? Oct 4, 2022 23:02 |
|
Whan you're saying doesn't make sense. Yes, it's hard to catch online cheaters, but he got caught. The difficulty of catching hans is zero because he already got caught.
|
# ? Oct 4, 2022 23:17 |
|
Are there shorter time control versions of how correspondence chess does it? Like, is there a thirty minute format where people use all the engines and databases they can to try to defeat each other? I suspect not, because it's not long enough to do the study you'd need to do and it would just turn into two people using their own custom automated data analysis tools as front ends to a suite of engines and historical databases. No time to think, so further automate the decision making process. I feel like it might actually be a fun format if you could lobotomize the engines down to a certain ELO or limit depth to a point where there's more judgement. Need to figure out a way to reduce draws though.
|
# ? Oct 4, 2022 23:27 |
|
They caught Hans because he was dumb as poo poo about his cheating. Like, he tabbed away to check an engine. Imagine someone whipping out a phone right at the board and checking stockfish. Hans was dumb as poo poo about his cheating because clearly, even chess.com didn't treat it as a big deal. They caught him cheating in 2020 and he won a Titled Tuesday Feburary of this year. If people start taking cheating seriously online, then you'll just get some smarter cheaters online, but it's orders of magnitude easier to cheat on chess.com and it will always be so.
|
# ? Oct 4, 2022 23:28 |
|
Xtanstic posted:As a member of Team Chaos, I am delighted to be able to read this after work. Why would that possibly happen?
|
# ? Oct 4, 2022 23:33 |
|
I wouldn't be surprised if part of the reason chess.com doesn't reveal their methods (and lichess leaves the details buried in the source code of a github, rather than having a big page summarizing it) is so that the people who might otherwise go to a lot of effort to cheat in an undetectable way instead cheat in ways that they think are undetectable, but are not, actually
|
# ? Oct 4, 2022 23:33 |
|
T.C. posted:Are there shorter time control versions of how correspondence chess does it? Like, is there a thirty minute format where people use all the engines and databases they can to try to defeat each other? I think http://www.infinitychess.com/ has held some competitions like that.
|
# ? Oct 4, 2022 23:35 |
T.C. posted:Are there shorter time control versions of how correspondence chess does it? Like, is there a thirty minute format where people use all the engines and databases they can to try to defeat each other? It is a fun format, because this is basically what TCEC is (well, minus humans, and only a single engine versus engine, but very comprehensively). The limiting fairness factor to this type of competition isn't the number of engines the human can access, but the total computing power the human has available to use, so stuff like TCEC has been a successful, ongoing, automated, and fair tournament to test engines against each other (notably being the first to fairly test GPU-powered NN engines versus the traditional CPU-powered MCTS engines)
|
|
# ? Oct 4, 2022 23:46 |
|
tractor fanatic posted:They caught Hans because he was dumb as poo poo about his cheating. Like, he tabbed away to check an engine. Imagine someone whipping out a phone right at the board and checking stockfish. Hans was dumb as poo poo about his cheating because clearly, even chess.com didn't treat it as a big deal. They caught him cheating in 2020 and he won a Titled Tuesday Feburary of this year. If people start taking cheating seriously online, then you'll just get some smarter cheaters online, but it's orders of magnitude easier to cheat on chess.com and it will always be so. in fact i suspect they told him how he got caught so he stopped doing it that way anymore
|
# ? Oct 4, 2022 23:49 |
|
tractor fanatic posted:If you're worried about cheating it's going to be increasingly hard to take online chess seriously. It will only get easier and easier to cheat and there's basically no way to stop it all. Online money is very much real money, and it might even be criminal to cheat in a money game, but all this just makes it harder to take online chess seriously. With OTB at least you can imagine there will be physical security measures, but there's nothing stopping increasingly sophisticated engine use online. At the money/prizes competition level that might be true. But if cheating gets only you to an ELO where you'd end up in a random 10 minute rapid game with me, you should probably just give up on chess. And if for some reason you turn out to be a bot that perfectly replicates the experience of playing a terrible novice, I guess I don't actually have that much to complain about. It would be a substantial realism improvement for most of Chess.com's bots. But yes, actual money competitions will probably eventually have to be pretty much exclusively OTB.
|
# ? Oct 4, 2022 23:50 |
|
also explains why Ian was on Magnus' side from the start.
|
# ? Oct 4, 2022 23:50 |
|
mfcrocker posted:If they're lying about this then it probably tanks their entire business. Put the tinfoil down. Yeah, the whole collusion angle is way out to lunch when most simply, ccom is in a position where Magnus is on their wagon now and he grossly mismanaged his accusation. So now they have to reveal far more about their cheat detection parameters than they ever needed to and it's going to cause a lot more cheating wankers to take note of subtle tells, like running the goddamn engine on the same PC. Seriously, though: any details they have to provide that could just remain a trade secret are just giving cheaters a specific thing to design around. I guess we'll see how detailed the report gets into it soon enough.
|
# ? Oct 4, 2022 23:57 |
|
https://www.chess.com/blog/CHESScom/hans-niemann-report
|
# ? Oct 5, 2022 00:02 |
|
Captain von Trapp posted:At the money/prizes competition level that might be true. But if cheating gets only you to an ELO where you'd end up in a random 10 minute rapid game with me, you should probably just give up on chess. And if for some reason you turn out to be a bot that perfectly replicates the experience of playing a terrible novice, I guess I don't actually have that much to complain about. It would be a substantial realism improvement for most of Chess.com's bots. Well those guys are already terrible cheaters, I think they're just kamikazeing their way up the ratings until the website catches up with them and bans them. I don't think they're really worried about detection, so I don't think they'll get any more subtle
|
# ? Oct 5, 2022 00:03 |
|
Duck Chess is incredible.
|
# ? Oct 5, 2022 00:04 |
|
Ravel posted:Duck Chess is incredible. I saw the rules on that a while ago but haven't tried it. The endgame seems kind of difficult to me, conceptually. Is king+rook vs king enough to win? I feel like the duck is such a strong defensive piece that it might not be. I guess king+queen still would be, though?
|
# ? Oct 5, 2022 00:38 |
|
If Chess is like any other sport, the entire organization would love it if Hans is cheating. They can provide a huge report & mountains of evidence on Hans specifically, but very intentionally not look further than that because it would uncover just how many cheaters there are. You always need a fall guy to make it look like they're successfully protecting the sanctity of the game from all cheaters. See? We caught Hans because he was cheating. We haven't caught anybody else, therefore nobody else is cheating. stratdax fucked around with this message at 03:07 on Oct 5, 2022 |
# ? Oct 5, 2022 03:05 |
|
|
# ? Jun 5, 2024 06:28 |
|
Also Eric Rosen just posted an hour long video of Duck Chess. Looks hilarious. "Ah I should have played Duck d3" https://www.youtube.com/watch?v=1owSL9JbP3E
|
# ? Oct 5, 2022 03:16 |