Occam's Razor Supports that Hans Niemann did not cheat against Magnus, or in OTB in general.

Sort:
rookNoob1982
lfPatriotGames wrote:
rookNoob1982 wrote:
PawnTsunami wrote:

My problem is all the people who are shouting "leave him alone, he didn't cheat!" - are you so sure?  Really?

My position is not so much that Hans didn't cheat, but rather there isn't enough evidence to accuse him of cheating (especially by the world champion), an accusation which could very easily destroy his career.  Hans could be cheating, so could Fabi, Nakamura, or even Magnus. I imagine all of whom would be incredibly angry if someone made that accusation about them.

I also think cheating online and OTB are so drastically different that Hans past online cheating can't really be logically factored into his OTB games. For me the simple answer is Hans practiced and improved, and maybe got a bit lucky, and beat the world champion. 

 

In my engine analysis Magnus had 87% accuracy and Hans had 92% accuracy. So that leaves just a 5% difference in quality of play. So really, Magnus could have won, and none of this would have happened. But neither player played impossibly well.

What, specifically, makes cheating online and cheating otb drastically different? Other than one is easier than the other. 

Well, to your point, ones easier than the other. A lot lot easier, and a lot more common. Its like the difference between stealing a pack of gum and robbing a bank and getting away with it. Finegold says he thinks 25% of GMs have cheated online. Which means Hans isn't all that unusual. 

Sure the intent, the desire to cheat is the same in both. And both are unethical. But the methods are drastically different. To the point where elaborate theories have to be constructed to explain how Hans cheated. And in my mind the simple answer, based on evidence, is he didn't. 

MorningGlory84
rookNoob1982 wrote:
lfPatriotGames wrote:
rookNoob1982 wrote:
PawnTsunami wrote:

My problem is all the people who are shouting "leave him alone, he didn't cheat!" - are you so sure?  Really?

My position is not so much that Hans didn't cheat, but rather there isn't enough evidence to accuse him of cheating (especially by the world champion), an accusation which could very easily destroy his career.  Hans could be cheating, so could Fabi, Nakamura, or even Magnus. I imagine all of whom would be incredibly angry if someone made that accusation about them.

I also think cheating online and OTB are so drastically different that Hans past online cheating can't really be logically factored into his OTB games. For me the simple answer is Hans practiced and improved, and maybe got a bit lucky, and beat the world champion. 

 

In my engine analysis Magnus had 87% accuracy and Hans had 92% accuracy. So that leaves just a 5% difference in quality of play. So really, Magnus could have won, and none of this would have happened. But neither player played impossibly well.

What, specifically, makes cheating online and cheating otb drastically different? Other than one is easier than the other. 

Well, to your point, ones easier than the other. A lot lot easier, and a lot more common. Its like the difference between stealing a pack of gum and robbing a bank and getting away with it. Finegold says he thinks 25% of GMs have cheated online. Which means Hans isn't all that unusual. 

Sure the intent, the desire to cheat is the same in both. And both are unethical. But the methods are drastically different. To the point where elaborate theories have to be constructed to explain how Hans cheated. And in my mind the simple answer, based on evidence, is he didn't. 

Finegold is a provocative, immature edgelord and has no evidence to back up his "25%" figure. It's inconsistent to demand evidence for Hans' alleged OTB cheating and then use an unsourced figure to reinforce a point you wish to defend.

rookNoob1982
Optimissed wrote:
lfPatriotGames wrote:
rookNoob1982 wrote:
PawnTsunami wrote:
 

 

What, specifically, makes cheating online and cheating otb drastically different? Other than one is easier than the other. 


In one, you're also more liable to be caught. If you haven't been caught cheating in the format that you're more liable to be caught in, that means you haven't done it.

This is actually one of the smartest posts I've seen on this topic. Great point.

rookNoob1982
MorningGlory84 wrote:
rookNoob1982 wrote:
lfPatriotGames wrote:
rookNoob1982 wrote:
PawnTsunami wrote:

My problem is all the people who are shouting "leave him alone, he didn't cheat!" - are you so sure?  Really?

My position is not so much that Hans didn't cheat, but rather there isn't enough evidence to accuse him of cheating (especially by the world champion), an accusation which could very easily destroy his career.  Hans could be cheating, so could Fabi, Nakamura, or even Magnus. I imagine all of whom would be incredibly angry if someone made that accusation about them.

I also think cheating online and OTB are so drastically different that Hans past online cheating can't really be logically factored into his OTB games. For me the simple answer is Hans practiced and improved, and maybe got a bit lucky, and beat the world champion. 

 

In my engine analysis Magnus had 87% accuracy and Hans had 92% accuracy. So that leaves just a 5% difference in quality of play. So really, Magnus could have won, and none of this would have happened. But neither player played impossibly well.

What, specifically, makes cheating online and cheating otb drastically different? Other than one is easier than the other. 

Well, to your point, ones easier than the other. A lot lot easier, and a lot more common. Its like the difference between stealing a pack of gum and robbing a bank and getting away with it. Finegold says he thinks 25% of GMs have cheated online. Which means Hans isn't all that unusual. 

Sure the intent, the desire to cheat is the same in both. And both are unethical. But the methods are drastically different. To the point where elaborate theories have to be constructed to explain how Hans cheated. And in my mind the simple answer, based on evidence, is he didn't. 

Finegold is a provocative, immature edgelord and has no evidence to back up his "25%" figure. It's inconsistent to demand evidence for Hans' alleged OTB cheating and then use an unsourced figure to reinforce a point you wish to defend.

Fair enough. I have heard elsewhere that chessdotcom keeps a running list of IMs and GMs who've cheated and it's in the triple digits. But granted, I don't have a concrete source on that. But is it likely Hans is the ONLY one? I doubt it.

lfPatriotGames

I'm not sure robbing a bank and stealing a pack of gum is a good analogy. One has more value than the other. 

If the result is the same, then how can they be drastically different? Lets say both are casual games, why would one be different than the other? Or, lets say the prize is 1,000 dollars. Why would one be different than the other?

If the only difference, is one is easier and more common, how would that impact the severity of the offense? If stealing a pack of gum from a kwiki mart is easier (and more common) than stealing the identical gum from an airport mall, wouldn't the actual offense, the actual loss be the same?

Or maybe a better example is our cities rampant car thefts. If a thief is caught stealing a Kia from a shopping mall vs stealing the identical car from a guarded parking garage is the penalty different? Or should it be different just because one is easier (and more common)?

Or I guess one way to look at it is imagine you are the victim. Lets stay with the car scenario (but it could be your lawnmower, your favorite guitar, your bank account, etc). The thief gets caught, and the prosecuting entity explains to you that since it was easier to steal from you vs. a guarded locked facility the penalty is going to be much less. Do you think that makes sense?

rookNoob1982
lfPatriotGames wrote:

I'm not sure robbing a bank and stealing a pack of gum is a good analogy. One has more value than the other. 

If the result is the same, then how can they be drastically different? Lets say both are casual games, why would one be different than the other? Or, lets say the prize is 1,000 dollars. Why would one be different than the other?

If the only difference, is one is easier and more common, how would that impact the severity of the offense? If stealing a pack of gum from a kwiki mart is easier (and more common) than stealing the identical gum from an airport mall, wouldn't the actual offense, the actual loss be the same?

Or maybe a better example is our cities rampant car thefts. If a thief is caught stealing a Kia from a shopping mall vs stealing the identical car from a guarded parking garage is the penalty different? Or should it be different just because one is easier (and more common)?

Or I guess one way to look at it is imagine you are the victim. Lets stay with the car scenario (but it could be your lawnmower, your favorite guitar, your bank account, etc). The thief gets caught, and the prosecuting entity explains to you that since it was easier to steal from you vs. a guarded locked facility the penalty is going to be much less. Do you think that makes sense?

In my view the penalty is irrelevant. I'm not concerned with the consequence, only the method, and the likelihood based on the method. Traveling to the moon vs traveling to another country are both "traveling" and you end up at a new destination. So in one sense the intent and consequence are the same (to travel). But the method is drastically different, making one far more unlikely than the other.

lfPatriotGames
rookNoob1982 wrote:
lfPatriotGames wrote:

I'm not sure robbing a bank and stealing a pack of gum is a good analogy. One has more value than the other. 

If the result is the same, then how can they be drastically different? Lets say both are casual games, why would one be different than the other? Or, lets say the prize is 1,000 dollars. Why would one be different than the other?

If the only difference, is one is easier and more common, how would that impact the severity of the offense? If stealing a pack of gum from a kwiki mart is easier (and more common) than stealing the identical gum from an airport mall, wouldn't the actual offense, the actual loss be the same?

Or maybe a better example is our cities rampant car thefts. If a thief is caught stealing a Kia from a shopping mall vs stealing the identical car from a guarded parking garage is the penalty different? Or should it be different just because one is easier (and more common)?

Or I guess one way to look at it is imagine you are the victim. Lets stay with the car scenario (but it could be your lawnmower, your favorite guitar, your bank account, etc). The thief gets caught, and the prosecuting entity explains to you that since it was easier to steal from you vs. a guarded locked facility the penalty is going to be much less. Do you think that makes sense?

In my view the penalty is irrelevant. I'm not concerned with the consequence, only the method, and the likelihood based on the method. Traveling to the moon vs traveling to another country are both "traveling" and you end up at a new destination. So in one sense the intent and consequence are the same (to travel). But the method is drastically different, making one far more unlikely than the other.

OK, I can understand your point of view. I think for most applications though the consequence is very important. Lets say there is a fine, a penalty (like disqualification or disinviting someone). Or maybe there is a lawsuit and the issue is money. If the harm is the same (cheated a thousand dollars online vs otb) it seems to me the consequences should be the same. Sure the method is different, but I would go back to the car thief. He used a different method to steal from you, because it was easier. He didn't choose the more difficult route of overcoming guards and security. 

 

Quanber

As I understand the debate, the claim is that Niemann cheats OTB using a signal from a computer or an assistant outside the playing room.

One can philosophize for a long time about whether it is even technically possible. Of course, there are many signal transmissions in frequency ranges that are not so well known and where normal  control equipment does not work. There are even ways to receive signals that do not require advanced technical equipment at the receiver. For example, the ear can be enough if the transmitter  are advanced enough:

https://www.ll.mit.edu/news/laser-can-deliver-messages-directly-your-ear-across-room

But if we are to stay out of the world of speculative agent equipment, the first relevant question to be asked is of course whether there is a statistically significant difference between performance in tournaments that are not broadcast live, and tournaments with live transmission.

 It is obvious that such a difference cannot exist if the person does not receive help.

Quanber

Unfortunately, a statistical study 2019-2020 shows that Niemann OTB consistently lost rating in tournaments without live transmission, and consistently won in tournaments with live transmission.

https://twitter.com/atl_kings/status/1568656197812891653

It has also been followed by several similar studies in other years and the difference is statistically significant.

There can't be as big a difference as 200 rating points in the two types of tournaments without Niemann scoring much better in live tournaments. Which doesn't make sense.

All statistical models rely heavily on whether you have chosen some good assumptions. Here it appears, in my opinion, that Regan has forgotten an important premise. To examine the difference between live and non-live broadcast matches.

 

Quanber

When it is important to think carefully about the prerequisites, it is because if Niemann cheats with an advanced program, if he has an undiscovered method, then of course he does not sit and win 100 games in a row. It goes without saying that it would be completely idiotic, like shooting yourself in the foot. He needs to slow down and make deliberate mistakes.

Other methods would be to investigate whether Niemann makes only win moves in those situations where it is absolutely necessary if he wants to win, or whether, like all other chess players, he occasionally makes a regular blunder.

It turns out again that Niemann has a remarkable statistic here. He can make inaccuracies, my smart moves, but pretty much never a blunder.

In the game against Carlsen, he does not make a single blunder and 54 out of 57 moves are within the choice of Stockfish:

https://www.youtube.com/watch?v=BbEiW-60hf0

Can we therefore conclude that Niemann is cheating OTB No, of course not. The above are just a few pieces in a big complicated puzzle.

 

What I consider absolutely decisive for the issue of cheating is how Niemann would fare in a tournament outside the USA, where security was upgraded to a level it is not today. If the tournament organizers made an effort and allied themselves with the experts who make a living from providing and tracking advanced equipment and conducted the tournament without live transmission, then for me it would be a definitive test that Niemann is not cheating if he performs at normal level 2700.

 

 

Quanber

 

In my opinion, there are two clear results in the upcoming trial and an endless number of  unclear solutions  in between : 

 

01) Niemann has a dyssocial personality structure, which means that he has received help to set up advanced equipment that plays for him without any pangs of conscience. In that case, he loses the case because this type of personality always forgets that it costs dearly to lie in court, and a lot of witnesses will be called. Sooner or later a witness will confess to save his own skin.

02) Niemann is a genius on the chessboard and has never received help OTB. In that case he wins the case, because no witnesses can bring him down. But he won't get any high compensation because he has notoriously admitted fraud, and he can't raise the burden of proof for a conspiracy where chess.com , Nakamura, Carlsen and whatnot have colluded to harm him.

03 ).... to endless ......

CraigIreland

#402: There exists interesting statistical data. A complication of these discussions is that not everyone who comments is aware of the key evidence.

#405: The first thing to learn about personality disorders is that for very good reason, they are not to be diagnosed by amateurs.

Quanber

I think I was misunderstood. I did not diagnosed Niemann. I only made the assumptions that if he do have a personality disorder then is is usually a problem  in court.

 

CrusaderKing1
Quanber wrote:

I think I was misunderstood. I did not diagnosed Niemann. I only made the assumptions that if he do have a personality disorder then is is usually a problem  in court.

 

I haven't seen any evidence of some far spectrum mental illness diagnosis. He seems fairly 'normal'.

Obviously you have to have a psychiatrist diagnose him with in-patient exams. 

Arm chair  internet psychiatrists have no domain.

But no, I would see no reason to send Hans to a psychiatrist if he were my patient. 

rookNoob1982
lfPatriotGames wrote:
rookNoob1982 wrote:
lfPatriotGames wrote:

I'm not sure robbing a bank and stealing a pack of gum is a good analogy. One has more value than the other. 

If the result is the same, then how can they be drastically different? Lets say both are casual games, why would one be different than the other? Or, lets say the prize is 1,000 dollars. Why would one be different than the other?

If the only difference, is one is easier and more common, how would that impact the severity of the offense? If stealing a pack of gum from a kwiki mart is easier (and more common) than stealing the identical gum from an airport mall, wouldn't the actual offense, the actual loss be the same?

Or maybe a better example is our cities rampant car thefts. If a thief is caught stealing a Kia from a shopping mall vs stealing the identical car from a guarded parking garage is the penalty different? Or should it be different just because one is easier (and more common)?

Or I guess one way to look at it is imagine you are the victim. Lets stay with the car scenario (but it could be your lawnmower, your favorite guitar, your bank account, etc). The thief gets caught, and the prosecuting entity explains to you that since it was easier to steal from you vs. a guarded locked facility the penalty is going to be much less. Do you think that makes sense?

In my view the penalty is irrelevant. I'm not concerned with the consequence, only the method, and the likelihood based on the method. Traveling to the moon vs traveling to another country are both "traveling" and you end up at a new destination. So in one sense the intent and consequence are the same (to travel). But the method is drastically different, making one far more unlikely than the other.

OK, I can understand your point of view. I think for most applications though the consequence is very important. Lets say there is a fine, a penalty (like disqualification or disinviting someone). Or maybe there is a lawsuit and the issue is money. If the harm is the same (cheated a thousand dollars online vs otb) it seems to me the consequences should be the same. Sure the method is different, but I would go back to the car thief. He used a different method to steal from you, because it was easier. He didn't choose the more difficult route of overcoming guards and security. 

 

This is valid. But remember Hans hasn't been "caught" cheating OTB, At least not yet. So we're weighing probabilities based on the evidence we have. And in my mind the difficulty of OTB cheating makes that probability extremely low. Not zero, but very low. 

If your argument is that he shouldn't play at all OTB because he cheated in the past online, then that seems like a different conversation altogether. It's a valid point, and would make whether he cheated OTB or not, completely irrelevant.

rookNoob1982
Quanber wrote:

Unfortunately, a statistical study 2019-2020 shows that Niemann OTB consistently lost rating in tournaments without live transmission, and consistently won in tournaments with live transmission.

https://twitter.com/atl_kings/status/1568656197812891653

It has also been followed by several similar studies in other years and the difference is statistically significant.

There can't be as big a difference as 200 rating points in the two types of tournaments without Niemann scoring much better in live tournaments. Which doesn't make sense.

All statistical models rely heavily on whether you have chosen some good assumptions. Here it appears, in my opinion, that Regan has forgotten an important premise. To examine the difference between live and non-live broadcast matches.

 

I'd take that chart with a grain of salt. I've read elsewhere it's not accurate.

CrusaderKing1
Optimissed wrote:

Also, she thinks that general medical practitioners have absolutely no understanding of mental illness and psychotherapy and that they often cause a lot of avoidable harm, through their lack of understanding in a specialised field, together with their assumption that they are actually competent to form an opinion.

The opinion could be that there is no reason to have a competent professional check them over. The next typical mistake made by medical professionals, apparently, is to contact a cbt practitioner.

Opinions are great and all, but opinions among the medical community are also broad and vast among each other. 

I specialize in a surgical field, but that doesn't mean I don't have understanding in some unexpected departments. I know more about the TM joint than 99% of dentists, for example.

I think some physicians are able to form competent opinions about fields they don't specialize in, but can't diagnose to the full understanding.

As a physician, I don't see any reason to send Hans to a psychiatrist. That being said, I don't see any reason that Hans has any mental illnesses past the normal spectrum of personalities. Psychiatrists are the only individuals that can diagnose Hans, and not only that, they have to do it while he is their patient, as they can't even do it from their homes. 

rookNoob1982
Optimissed wrote:
rookNoob1982 wrote:
Quanber wrote:

Unfortunately, a statistical study 2019-2020 shows that Niemann OTB consistently lost rating in tournaments without live transmission, and consistently won in tournaments with live transmission.

https://twitter.com/atl_kings/status/1568656197812891653

It has also been followed by several similar studies in other years and the difference is statistically significant.

There can't be as big a difference as 200 rating points in the two types of tournaments without Niemann scoring much better in live tournaments. Which doesn't make sense.

All statistical models rely heavily on whether you have chosen some good assumptions. Here it appears, in my opinion, that Regan has forgotten an important premise. To examine the difference between live and non-live broadcast matches.

 

I'd take that chart with a grain of salt. I've read elsewhere it's not accurate.

I've just been looking at that chart. I wonder if you realise that if that chart is an accurate and full account of Niemann's tournament activity, then it's statistically highly significant and is more or less conclusive evidence, by itself, that he has cheated.

You say you have read that it's inaccurate. That could be because it's inaccurate OR because it's accurate and someone wishes to discredit it. Superficially it's very strongly conclusive, against Niemann. It's necessary to find out if it's accurate.


I do think its important to verify if the chart is correct. But even if it is accurate, it's actually not all that useful. 19 games isn't nearly enough data to draw any firm conclusions. Maybe if it had 100 games it would be more telling. But even then, the only expert I know of on statistical analysis for Chess cheaters has already cleared Niemann.

rookNoob1982
Optimissed wrote:


Elroch is stats king and I'm not qualified but to me, it looks like a less than 1 in 2 ^15 chance that it would be obtained randomly. That's one chance in about 65500 that it could be simply a random result.

Given you have some expertise in this area, what would that number be if the next 20 games were the opposite?

rookNoob1982
Optimissed wrote:
rookNoob1982 wrote:
Optimissed wrote:
rookNoob1982 wrote:
Quanber wrote:

Unfortunately, a statistical study 2019-2020 shows that Niemann OTB consistently lost rating in tournaments without live transmission, and consistently won in tournaments with live transmission.

https://twitter.com/atl_kings/status/1568656197812891653

It has also been followed by several similar studies in other years and the difference is statistically significant.

There can't be as big a difference as 200 rating points in the two types of tournaments without Niemann scoring much better in live tournaments. Which doesn't make sense.

All statistical models rely heavily on whether you have chosen some good assumptions. Here it appears, in my opinion, that Regan has forgotten an important premise. To examine the difference between live and non-live broadcast matches.

 

I'd take that chart with a grain of salt. I've read elsewhere it's not accurate.

I've just been looking at that chart. I wonder if you realise that if that chart is an accurate and full account of Niemann's tournament activity, then it's statistically highly significant and is more or less conclusive evidence, by itself, that he has cheated.

You say you have read that it's inaccurate. That could be because it's inaccurate OR because it's accurate and someone wishes to discredit it. Superficially it's very strongly conclusive, against Niemann. It's necessary to find out if it's accurate.


I do think its important to verify if the chart is correct. But even if it is accurate, it's actually not all that useful. 19 games isn't nearly enough data to draw any firm conclusions. Maybe if it had 100 games it would be more telling. But even then, the only expert I know of on statistical analysis for Chess cheaters has already cleared Niemann.

Oh I thought it looked like it was 19 tournaments. I discounted four, which were less significant. I think that means it's less than 2 ^15 but they were small deviations from the mean in the "correct" direction, so I suspect 65,000 is about right.

I don't believe it should be assumed that Regan has cleared Niemannn and I know Elroch believes he's guilty.

This seems to indicate Regan doesn't believe Hans cheated https://en.chessbase.com/post/is-hans-niemann-cheating-world-renowned-expert-ken-regan-analyzes Read his verdict at the end.