Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

wsj.com

Post reply on HN
Print view View original

141–150 of 339 posts

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#141

Post by chrisherring »

EddySchauHai wrote: post #60
tobyjsullivan wrote: post #20

Interesting piece. It highlights how it's hard to differentiate between someone genuinely playing at a level higher, and a cheater. A few things I noted from this article: 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 hist…

> it's hard to differentiate between someone genuinely playing at a level higher, and a cheater 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…

A smart cheater isn't just going to replicate bot moves and make it easy to detect. They may just use it to decide between 2 moves they were 50/50 on already. Do this 2 or 3 times and it would make a big difference at the grand master level. This would be quite hard to detect.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#142

Post by Andrew_nenakhov »

Bud wrote: post #130
Andrew_nenakhov wrote: post #43

Earlier quoted context omitted.

I know the basics how to do it, it is rather obvious. I just feel that operating two boards manually is really a chore, and definitely not while streaming.

Are you familiar with professional Starcraft II? Those guys are executing ten moves PER SECOND. While streaming. This is two to four orders of magnitude faster than chess players in a classical game. One to two orders of magnitude faster than blitz.

StarCraft players also aren't really thinking, all micro is mostly almost reflexes. Chess does require way more analysis, so if you are on a very short time limit, checking up with an engine and operating it would eat all your time.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#143

Post by jonny_eh »

whatshisface wrote: post #93
pclmulqdq wrote: post #83

Earlier quoted context omitted.

This is, I think, the single biggest factor in the proliferation of cheating. Most cheaters are smart enough to do the math on how they should win at the sport, or is smart enough to hire someone who can do the math. In other words, cheaters are rational actors. 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 s…

The alternative - an environment where innocent people were sometimes getting expelled - would be worse when it's just a game.

Sometimes people are caught red-handed.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#145

Post by lamontcg »

antiterra wrote: post #45
selfmodruntime wrote: post #22

The fact that an athlete in a competitive sport was allowed to partake in an event even after admitting to cheating not only once, but twice, is outrageous in itself.

The admissions were from when he was 12 and 16; many so societies generally believe in redemption from childhood transgressions.

This assertion keeps on being made and it seems almost deliberately obtuse to me.

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.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#146

Post by JetSetIlly »

raydiatian wrote: post #134
hn_throwaway_99 wrote: post #101

> “Outside his online play, Hans is the fastest rising top player in Classical [over-the-board] chess in modern history,” the report says, while comparing his progress to the game’s brightest rising stars. “Looking purely at rating, Hans should be classified as a member of this group of top young players. While we don’t doubt that Hans is a talented player, we note that his results are statistically extraordinary.” I…

Yeah you pointed out something special here. It’s like those unsung Russian sub captains who didn’t start the nuclear apocalypse because they had an intuitive understanding between faulty and genuine threats. You can’t explain why you know something you just know it. 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.

> You can’t explain why you know something you just know it.

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.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#147

Post by mzs »

chesscom wrote: post #144

There is so much more to our report than what was focused on in the WSJ article. The full report will be shared shortly…

>Rensch had previously said that Chess-com had never shared a list of cheaters or the platform’s cheat detection algorithm with Carlsen.

So who did chess.com share those with?

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#148

Post by ianferrel »

ActorNightly wrote: post #110
bombcar wrote: post #79

Earlier quoted context omitted.

It might work if there are no spectators live in person. Otherwise you can easily have someone in person relaying the information to a computer, or you have to search everyone, not just the players.

If it happened, I don't think it happened realtime. 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.

I don't think that what you described is cheating.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#149

Post by Andrew_nenakhov »

Bud wrote: post #131
Andrew_nenakhov wrote: post #41

Earlier quoted context omitted.

> 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 b…

Yet earlier in this thread, you were incredulous about even the possibility that a chess game state could be entered into software or communicated during play. Which is clearly a ludicrously blind statement. So pardon us if it's hard to be particularly impressed by, well, whatever you are impressed or not impressed by. Because you don't seem to have much grasp on the basics involved, here.

> Which is clearly a ludicrously blind statement.

This statement came from an experiment I conducted. I just tried playing lichess 3+2 game while running a nearby chess.com/analysis on a second display. ...

... and I was barely keeping up with all the clicks, and in the endgame was unable to keep up. So no, efficient cheating in fast games requires at least a special training, and better some automated software to keep up with the moves and to communicate best moves back to player.

Re: Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times

#150

Post by lamontcg »

paxys wrote: post #75
jrm4 wrote: post #42

Earlier quoted context omitted.

Right? I completely didn't understand this part? In pretty much everything else, you cheat, you're basically done.

> In pretty much everything else, you cheat, you're basically done. 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.

MLS you catch a 10 game ban (out of a 34 game season) for your first PED offense.
Post reply on HN