|
to me this is fascinating because i'm a huge 'most everyone in the olympics is on dope, and im glad of it' guy, because it's still real athletes going at it with their talent, they are all just the amp from spinal tap that can go to eleven thanks to the drugs. But if you cheat at chess you just completely gently caress over the whole sport lol.
|
#
?
Oct 5, 2022 17:01
|
|
|
|
| # ? May 29, 2024 05:50 |
|
|
Duck chess is good. Stockfish doesn't understand duck chess. Therefore the world championship should be a surprise duck chess match to ensure no cheating.
|
|
#
?
Oct 5, 2022 17:02
|
|
|
neaden posted:Duck chess is good. duck chess is a better game than regular chess change my view
|
|
#
?
Oct 5, 2022 17:04
|
|
|
Hand Knit posted:My mind suddenly went to the NFTs of these cheated games. Speaking of NFTs, someone investigated that stuff few weeks ago: A grand total of 6 people have bought one of the chess.com NFTs since their inception 5 months ago
|
|
#
?
Oct 5, 2022 17:47
|
|
|
Hand Knit posted:Also, a thought just occurred about chess.com and cheat detection. It strikes me that their methods will catch things like players hanging out on blitz and talking to each other over chat while playing. If they're talking about the game and that's affecting their moves then it seems like the blur detection might pick that up. And, while that is in fact cheating, it doesn't seem like the sort of thing that cheat detection is actually trying to catch. It's sort of like how those blitz sessions where Carlsen is hanging out drinking with Howell and whomever are in fact cheating but also not what the actual concern is when it comes to fair play enforcement.
|
|
#
?
Oct 5, 2022 17:48
|
|
|
Vegetable posted:I seriously doubt they’re looking at just one signal to make an assertion of cheating, particularly if there’s human review involved. Apparently they never caught Hans again after his new account though, so either he's been clean since then or you can defeat their anti-cheat just by not switching to a separate window...
|
|
#
?
Oct 5, 2022 17:52
|
|
|
Vegetable posted:I seriously doubt they’re looking at just one signal to make an assertion of cheating, particularly if there’s human review involved. Oh I'm sure they're not but something like chatting over WhatsApp is gonna get us all of repeated blur, different pattern/quality of moves during blur, plausibly superior-on-aggregate blurred moves, and skewed time usage.
|
|
#
?
Oct 5, 2022 18:04
|
|
|
dex_sda posted:duck chess is a better game than regular chess change my view a lot of chess games are better than regular chess, see also: shogi
|
|
#
?
Oct 5, 2022 18:07
|
|
|
silvergoose posted:a lot of chess games are better than regular chess, see also: shogi i don't like the 'hunker down' phase of a shogi game but it is a good one
|
|
#
?
Oct 5, 2022 18:10
|
|
|
cheetah7071 posted:basically, chess.com's cheating policy is at best suspect and at worst irresponsible, but, within the confines of that, their recent behavior and the integrity of the report seem to be about as good as one might hope They aren't even following their cheating policy. This is specifically why I suggest they are being unethical. Their own publications say "Chess.com will never discuss cases publicly. Will never happen. Stop asking." They claim to have closed the accounts of more than 500 titled players - none of these have had a 72 page report and all this public disclosure of what should be protected information under their privacy policy. (sources: https://www.chess.com/article/view/online-chess-cheating) And to say "well we had to do this because he was talking smack about us" - seriously? is this high school? STILL not getting into this so-called "statistical analysis" that isn't. Hans is a cheater, chess.com is ridiculous. Both these things can be true.
|
|
#
?
Oct 5, 2022 18:10
|
|
|
Also leaking Dlugy's confession just because Magnus snidely mentioned him in connection with Hans was just as lovely.
|
|
#
?
Oct 5, 2022 18:15
|
|
|
Apsyrtes posted:Hans is a cheater, chess.com is ridiculous. Both these things can be true. well that's a much more reasonable stance than calling their statistics nonsense and refusing to elaborate
|
|
#
?
Oct 5, 2022 18:16
|
|
|
Hand Knit posted:My mind suddenly went to the NFTs of these cheated games. All my Qb3s gone. *hides web search of 'how to chess notation'*
|
|
#
?
Oct 5, 2022 18:26
|
|
|
Apsyrtes posted:
Am I wrong to think that the Chess.com statistical analysis is not transparent at all? Like, what is a Strength Score? And how is it different from the other analysis' that people have done?
|
|
#
?
Oct 5, 2022 18:28
|
|
|
I think Chess.com is reliable enough that you can trust their statistical analysis is good, but it largely just confirms stuff we already knew. I suspect Strength Score is probably very similar to the metric Ken Regan uses, i.e. it ignores openings, forcing moves, etc
|
|
#
?
Oct 5, 2022 18:35
|
|
|
tractor fanatic posted:I think Chess.com is reliable enough that you can trust their statistical analysis is good, but it largely just confirms stuff we already knew. I suspect Strength Score is probably very similar to the metric Ken Regan uses, i.e. it ignores openings, forcing moves, etc I think they literally cover this in one of the earlier sections of the report where they say they analyze moves past the opening and discard simple/known endgames. It may not have been referring to strength score but they did say that about something. Also, a lot of their analysis is comparative which I think mitigates some possible bias they may have in their statistical analysis. Not completely, if their analysis is flawed, but it should help. Not to mention that they clearly point out that some of their analysis points towards Hans NOT cheating according to that metric. totalnewbie fucked around with this message at 18:42 on Oct 5, 2022 |
|
#
?
Oct 5, 2022 18:39
|
|
|
It's worth pointing out that probably a main reason they do the "confess and we won't tell" thing is that confessions validate their entire model and it's efficacy. What can they offer other than confidentiality and a new account to elicit a confession? How can they claim confidence in their model without confessions?
|
|
#
?
Oct 5, 2022 18:53
|
|
|
I think the whole thing puts both the suspected cheater and chess.com between a rock and a hard place. If you're suspected of cheating and suspended by chess.com, you have to either confess to something you didn't do to get reinstated or I guess get banned (make a new account?) neither of which are great, are they. Chess.com on the other hand, they think you're cheating. In cases where it's not obvious like a 1200 suddenly playing top stockfish lines, what are you to do? You have strong statistical evidence they're cheating but they insist they are not. Well.. now what? Do you just take them at their word, even though "I'm not cheating" is basically implied at all times anyway and their explicit statement is basically meaningless? If you accept it then you're letting through someone you are pretty positive is cheating and probably given their efforts in cheat detection, they're probably right much more than they're wrong. On the other hand, how do you go about PROVING they cheated if they deny it while not relying on your cheat detection? Do you get someone to go through their games one by one? That's obviously not feasible either. So.. what should they do?
|
|
#
?
Oct 5, 2022 19:01
|
|
|
Sub Rosa posted:It's worth pointing out that probably a main reason they do the "confess and we won't tell" thing is that confessions validate their entire model and it's efficacy. What can they offer other than confidentiality and a new account to elicit a confession? How can they claim confidence in their model without confessions? They can offer people a chance to exonerate and defend themself against their claims before a neutral third party arbitrator. It's how sports leagues investigate malfeasance in most situations when the Commissioner can't fulfill that role themselves. Pro sports teams, players, leagues all have legal counsel on retainer. Chess.com has a vested interest in maintaining confidence in their cheat detection system, so they will always maintain that their system is good. I don't think I need to explain why the players are invested in defending themselves, as well as seeing their competitors removed. If the chess world broadly aspires to the attention and prestige of a professional sports league, they need to hold themselves to the same standards. Like others have pointed out, I'm not sure how they are going to explain their behavior since the Sinquefield Cup before a third party. They have violated their own privacy policies, and possible maybe conducting data collection that questionable under various privacy regimes, especially in Europe.
|
|
#
?
Oct 5, 2022 19:14
|
|
|
stratdax posted:Also Eric Rosen just posted an hour long video of Duck Chess. Looks hilarious. "Ah I should have played Duck d3" lol neat Agadmator highlights the potent power of the bishop pair in his videos. Am I right in thinking duck chess flips the script a bit and turns knights into more potent attackers? If you create a fork or similar dilemma with a bishop or rook or queen etc the opponent can move one piece and likely duck-block the other, whereas the knight gives no fucks about ducks.
|
|
#
?
Oct 5, 2022 19:15
|
|
|
Yeah in duck chess knights might be on a level with rooks.
|
|
#
?
Oct 5, 2022 19:21
|
|
|
They have a new wanding/scanning device at the U.S. Chess Championship in addition to the standard wand. Of the four or five people I saw get scanned, the only player they had turn around during wanding was Hans Niemann and then the dude scanned his butt.
|
|
#
?
Oct 5, 2022 20:03
|
|
|
duck chess owns, i am addicted
|
|
#
?
Oct 5, 2022 20:04
|
|
|
Is White's advantage bigger in duck chess? It looked really easy to force black into a suboptimal opening and then snowball from there. I almost wonder if black should get first duck move
|
|
#
?
Oct 5, 2022 20:06
|
|
|
poonchasta posted:They have a new wanding/scanning device at the U.S. Chess Championship in addition to the standard wand. Of the four or five people I saw get scanned, the only player they had turn around during wanding was Hans Niemann and then the dude scanned his butt. Please let this be real and for there to be a video
|
|
#
?
Oct 5, 2022 20:34
|
|
|
gret posted:Also leaking Dlugy's confession just because Magnus snidely mentioned him in connection with Hans was just as lovely. It's pretty relevant given he's Han's coach. Chess.com is lovely and they are going for maximum drama but Han's career in chess is over. Watching the US chess champions now and they're using lichess for the analysis, is this new? lol if Rex Sinquefield is gunning for them now
|
|
#
?
Oct 5, 2022 21:03
|
|
|
Loving Africa Chaps posted:It's pretty relevant given he's Han's coach. I don't know if it's Lichess buying the broadcast rights but from time to time you'll see the broadcast team use their interface/engine.
|
|
#
?
Oct 5, 2022 21:09
|
|
|
Hand Knit posted:I don't know if it's Lichess buying the broadcast rights but from time to time you'll see the broadcast team use their interface/engine. Maybe i'm reading too much into it but i thought i remembered on previous broadcasts they had a generic unbranded analysis board. Edit: i'm making it up, they've been using lichess previously
|
|
#
?
Oct 5, 2022 21:10
|
|
|
Sub Rosa posted:It's worth pointing out that probably a main reason they do the "confess and we won't tell" thing is that confessions validate their entire model and it's efficacy. What can they offer other than confidentiality and a new account to elicit a confession? How can they claim confidence in their model without confessions? Recruit some master players, give them titled alt accounts that are flagged as protected from automatic bans, and tell them to go out there and cheat using whatever creative method the player can come up with. Instruct them to record which games they cheat in and what the method was. Random matchmaking only of course, and no arenas and whatnot. At the end of this you have a bunch of labeled cheated games from a high level player, and you also have that player's legit games from their main account to compare to. After some short amount of time you close the cheated account and refund points. This also allows you to collect data about your catch rate for various methods, including subtle cheating or non-engine based stuff like having a friend on discord. Eyes Only fucked around with this message at 21:13 on Oct 5, 2022 |
|
#
?
Oct 5, 2022 21:11
|
|
|
Eyes Only posted:Recruit some master players, give them titled alt accounts that are flagged as protected from automatic bans, and tell them to go out there and cheat using whatever creative method the player can come up with. Instruct them to record which games they cheat in and what the method was. Random matchmaking only of course, and no arenas and whatnot. Hikaru said on stream, about a week ago, that chess.com approached him about doing this exact thing but it made him feel too icky to do it.
|
|
#
?
Oct 5, 2022 22:07
|
|
|
Hans just crushed Christopher Yoo with black in a crazy game. If he’s clean, he’s just good. Is it possible that cheating with an engine can make you better after you quit cheating, similar to how training with steroids still gives you major benefits even after you stop taking them?
|
|
#
?
Oct 5, 2022 22:55
|
|
|
wedgie deliverer posted:Hans just crushed Christopher Yoo with black in a crazy game. If he’s clean, he’s just good. everyone uses engines in training all the time and it isn't cheating. It's not like you're forbidden to use them in analysis and practice.
|
|
#
?
Oct 5, 2022 22:57
|
|
|
poonchasta posted:They have a new wanding/scanning device at the U.S. Chess Championship in addition to the standard wand. Of the four or five people I saw get scanned, the only player they had turn around during wanding was Hans Niemann and then the dude scanned his butt. lol wedgie deliverer posted:Is it possible that cheating with an engine can make you better after you quit cheating, similar to how training with steroids still gives you major benefits even after you stop taking them? also wondering about this: if you have immediate feedback on your moves during online blitz, is that training more effective vs getting the same feedback after the game?
|
|
#
?
Oct 5, 2022 23:16
|
|
|
Loooool Hans just won his Round 1 game, the only person to do so, and said "the chess speaks for itself" again and walked away legend
|
|
#
?
Oct 5, 2022 23:25
|
|
|
dumb rear end shoulda stopped cheating when he had the chance, he's a good player
|
|
#
?
Oct 5, 2022 23:37
|
|
|
cheetah7071 posted:everyone uses engines in training all the time and it isn't cheating. It's not like you're forbidden to use them in analysis and practice. What about in a sort of live/scrimmage environment? Again, I'm a very casual player, but getting that sort of live competition experience with assistance specifically. For example, I know when I compete in martial arts I know I fight much more conservatively and defensively than how I am in practice. Are chess players similar when the stakes are high? I could see engine assistance sort of showing people that certain moves that they wouldn't have the confidence to play in a stressful situation are actually fine, and once those training wheels are off they still retain some capacity to recognize those moments better than they would have otherwise.
|
|
#
?
Oct 5, 2022 23:38
|
|
|
VROOM VROOM posted:Loooool Hans just won his Round 1 game, the only person to do so, and said "the chess speaks for itself" again and walked away https://twitter.com/STLChessClub/status/1577788033767772160 LOL he's totally leaning into it.
|
|
#
?
Oct 5, 2022 23:39
|
|
|
Honestly, Chess needed a good heel and it's good he's embracing the role.
|
|
#
?
Oct 5, 2022 23:41
|
|
|
wedgie deliverer posted:Please let this be real and for there to be a video https://twitter.com/SunG3wd/status/1577788893776207873?s=20&t=XlHD4P-27PdEvuUPpG1wHA
|
|
#
?
Oct 5, 2022 23:53
|
|
|
|
| # ? May 29, 2024 05:50 |
|
|
Duck chess appears to be way deeper than it has any right to be.
|
|
#
?
Oct 5, 2022 23:57
|
|
