Came here looking for this

Time stamp?

@@basimcool927814:12

@@basimcool9278 14:09 maybe? I searched the transcription

He is doing Russian math, they have their own Russian truth 🤭

Indeed! He must’ve cheated! Not against Magnus Carlsen. Not in important tournaments. No… He cheated against some low level FMs. 😂😂😂 /s

@@str3123 🤣🤣🤣

@@str3123while doing 10 things at the same time too lol 😂😂😂😂

Incon-theivable!

😂😂😂😅

went to comment this when I heard it but you got there first

Also accurate, given the failure to understand the statistics involved!

Ehh... no I don't think so. I really doubt stockfish has any specific code for humans in it, but I know for sure that Alpha and Leela do not. But chess cheating detectors do have human play related information in them, that's what Hikaru is talking about (source: I have written my own chess engine and I'm somewhat familiar with the code of stockfish). Chess computers work by searching very deep, not by trying to emulate human games. That's why they can stomp humans.

Hikaru looking at the ceiling turns out is not him asking stockfish, it's him telling stockfish what to do in its own games

@@ConsciousExpression You've not seen the metal ports on the back of Hikaru's neck where he plugs into the matrix? Goddamn open your eyes!

​​@@ConsciousExpressionyikes bud, you can code your own chess engine but can't spot a joke ? 😅 you need to get off the PC a bit and go outside mate. Need to level up those social interaction points .

@@Majdar oh its a joke! i didnt understand it was a joke either because there is no humor in it

Absolutely! I'm not sure what he contracted between hallucinations, drugs or jealousy but I hope he recovers soon enough ^^

not sure head injuries are fixable

@@Aienhel Head injuries, yes. Brain injuries, no.

Hikaru: Call an ambulance Also Hikaru: But not for me.

If he plays me blindfolded with queen odds. a bad mouse while drunk i think i could manage a draw in a 40 game set. (on a good day)

he's already got one

The example Hikaru gave yesterday was like the example of the pennies given in the Reddit post. He pointed out that the streak of 44.5/45 happened after a loss. So Krammik was already looking at 44.5/46 when he cherry picked the data and presented it without context.

@@jadonloomis Whats your rating?

@@wooshbait -300

the putin strategy

@@donsurlylyte what?

@@1degenerate it is something that a lot of arrogant and narcissistic people tend to do, they are unable to cope with being wrong, even if it is a minor mistake which often leads to them looking like a even bigger idiot than if they just owned their mistake, typically these people are in positions of power so they can retaliate towards anyone willing to challenge their beliefs (or anyone not willing to feed their egos)

​@@1degenerateHavent read the news? Or incapable of relating those two situations?

What did news say to you? Ps Im not stupid enough to watch thouse but it will be funny to hear u out

No, you just need to work hard then wait for geniuses like Kasparov to do retirement championships 😂😂😂😂

Although it could be that he’s jelous

@@Blinkers2007GameDev I think this is what is happening. He has too much time on his hand and too little attention as a retired player, so he is basically satisfying his narcissism and jealousy towards younger, still active players in this way. It is quiet pathetic, to be honest.

In fact, chess is an ability not attached to intelligence. Like solving Rubik cubes.

True, just look at Magnus

Who would have thought that chess skill didn't transfer to other areas of life? Though you'd think you'd find a second opinion about these statistics before you publicly make an arse of yourself!

Probability is sometimes extremely counterintuitive though, like the Monty Hall Problem-even Paul Erdos got it wrong.

to be fair. humans are not good at probability. most people get coin tosses wrong.

Kudos to Pjotr, I was very happy to see someone take the time and explain it patiently and respectfully (not that it will help Kramnik at this point...). I do happen to also have a PhD in math (game theory, specifically), but it really doesn't take a PhD to see the problems here -- it's Probability 101. Literally anyone after two months in STEM would be able to point it out. Not rocket science.

​@@OlaRozenfeldP.H.D in Physics here, and the wife has a P.H.D in Stats, the math does check out, the chances aren't that low when considering Hikarus respectably large amount of career games.

I mean, I'm a psycholgist who took basic statistics class and play fantasy football and yet laughed at Kramnik immediatly, way before this post. He basically put less effort than our fantasy football league from highschool in analizing the events.

Engineers would only be 800 Elo mathematicians 🙄

Yes, it is called the law of large numbers. Given sufficient games, many odd things that would seem almost impossible if taken out of context of the history of playing many games becomes highly likely. That is not even considering the psychological part of the game, which makes these runs even more likely.

i personally don't even think the chance is worth talking about because this event was really not all that rare in the first place. that is the whole point. my point being, while yes of course rare events do happen, if something is exceedingly rare it does raise big questions, and there are better ways to disregard these accusations rather than just "rare events do happen" because that reinforces the idea that this was rare when indeed it really wasn't and that's the whole point.

If unlikely events never happened, statisticians would be out of a job haha

“Statistically rare event don’t happen” Taleb: “hold my black swan”

This depends on what you mean by rare. Because there is a point where the chance becomes too low, where because it is so unlikely it is unlikely to ever happen within tens of thousands of years. Technically possible, but statistically impossible. Now whether or not hikaru's record falls under the statistically impossible category, I dunno. Haven't done or seen the maths so I can't say, more pointing out that the premise of what you are saying is dismissing something that is really done.

@@jambott5520correct statistics can be used. Though kramniks statistics are no good

Yes, he's a well known nutter.

Paranoid. I have seen other - much lesser titled players - make absurd accusations every time he lost.

Kramnik and Kamsky, both of them completely paranoid about cheating in online chess. I've lost all respect for Kramnik lately, and in my view he is bringing chess into disrepute with these baseless accusations against Hikaru.​

Magnus has even more impressive performances, he 22-4 Fabi, who recently won St Louis Rapid and Blitz. Kramnik went after the wrong person 😂

I agree with your description of Hikaru, but I think that the issue here is that players of their level literally just need a single signal that there is a tactic in the current position and they'll always find it. I doubt Hikaru would cheat but the whole thing is just a result of figuring out how to eliminate any form of cheating in high level online chess since it can make a huge difference on the top level

Yes, you can verify Hikaru has a 4.8% chance of getting 45 out of 46 points if his opposition is 400 points lower rated on average, by using a binomial distribution calculator and entering the numbers for "cumulative probability" X >= 45, N=46 and P= 0.9 (90% chance of victory if you are roughly 400 points higher rated than your opponents). BTW using the same calculator it is said the world has an even chance of another nuclear bomb 'incident' (like in Hiroshima / Nagasaki) once every 75 years or so, if p= 0.01 for the chance of a nuclear event every year (1% chance every year), N=75 years, hit the Enter key, and you'll get a probability, last I checked of roughly 50%.

​@@raylopez99No wonder the percent is so low. You Are calculating a 45 game winstreak out of 46 games... which half of the video explained why not to do.

@@geirtristananton9305 Hi not sure what you mean. I thought Naka had 45 out of 46 games (that was my assumption) that he won. And 4.8% chance is pretty good, happens 1 in 20 times, so Hikaru not cheating.

@@raylopez99 I dont think he has 90% chances of victory though, draw chances are still quite high. but yeah it's still not that unlikely over a large number of games

@@Will-uj7yu I believe number of games is not relevant in a binomial calculator. As I recall the binomial calculators assume sampling without replacement meaning the sample size is assumed infinite. So the calculator will computer out of N=46 games how often you win 45 times, which is what my initial 4.8% chance of 45/46 found given P=0.9. In any event, we seem to all agree that Hikaru is not cheating and that his win streak is due to chance alone.

I will beat 99.99% of all legit 1300 players aswell, thats 400 less than i have, so, yes, you are right

I witnessed in person two people of same rating playing a match and one winning 20-0 these things are not anomalies

My child always complains that i am cheating in chess because i have a 100% win ratio over her. According to stats, there is only a 0.0000000456% chance that a fully grown adult will win over 100 games in a row over multiple years against an opponent that is still learning the rules of chess and trying her best as a 10 year old 😂😂

@@ianmolnar8889 hikaroo was i think an IM at that age

I'm sorry for saying that, I don't think that Hikaru is cheating. But your argument isn't logical. Why? Let's assume he is consistently cheating. Then he would have big elo, so that situation is logical and your argument is not

Ya this is certainly no dream situation. These "professional mathematicians" must have been the same "professional players" he claims support him.

True@@freepimaths9698

I am not familiar with chess statistics, but wrote my thesis from Stock Market anomalies. And I would like to mention that even scientists for decades published various researches about various clendar anomalies - supported by statistics. And thereafter in ~2001 Sullivan, Timmerman wrote a nice article (Dangers of data mining: The case of calendar effects in stock returns) in which they showed "However, this evidence has largely been considered without accounting for the intensive search preceding it. In this paper we use 100 years of daily data and a new bootstrap procedure that allows us to explicitly measure the distortions in statistical inference induced by data mining. We find that although nominal p-values for individual calendar rules are extremely significant, once evaluated in the context of the full universe from which such rules were drawn, calendar effects no longer remain significant." How is it related? Even if Kramnik were doing good enough math (similar to the original researches for market anomalies) the very fact that he searches for them introduces a bias that the otherwise significant results stop being significant. If Kramnik's first hypothesis would be that Of all only player it is Hikaru is cheating and he is most suspect when makes long streaks against same opponent, then the same statistical outlier could be significand, however what really happens that he went in that 'intensive search mode preceding it' and 'hunt for whatever anomaly since I heard some gossip and anyhow I do not like him' therefore even the very same anomalies suddenly much less significant.

yes

Lol 😆 nice 👍

Ikr 😂😂 that was such a God tier insult.

🤣🤣😂🤣😂🤣😂🤣😂🤣😂🤣😂🤣😂🤣😂

Absolute violation loool

They are literally hand picked Hikaru pawns he has been beating for years. Kramnik is a idiot literally.

Apologies i appear to have made the same point you did but with many more words.

I don't believe Kramnik has ever read a book about probability because I read about half of one and I could see the bias and the basic mistake he felt to. The mathematician that rebutted him was massively overqualified for the task at hand.

Best comment

@@sailor5853 I know generally one should not attribute to malice if incompetence is a sufficient explanation, but in this case I would not rule out that the data point in question was cherry picked on purpose just to stir up drama so people talk about Mr Kramnik again.

I'm JuSt ASkinG quESTioNs! Don't give me any answers I don't want to hear, though! That's just hysteria and tears!

@@sailor5853 I find it very odd that " Kramnik has ever read a book about probability", after all he has collaborated with people that develop algorithms to detect cheating, he must surely know something... I mean this is not rocket science, the math displayed here is actually very simple, it has been ~20 years since I had probability classes at university (I studied computer science not math) and I find it very easy to calculate all of this. In any case, Kramnik is wrong, either because he doesn't understand it or he deliberately chose not to.

It may be insane and potentially unfair but it's the best way to detect a cheater unfortunately. That being said, if a 1 in 100 event is your case than you're really just throwing poop and the wall and hoping it sticks. For reference, throwing heads 50 times in 50 throws is a less than 1 in a quadrillion (thousand million million)

You only need 7 consecutive heads or tails to reach something with less than a 1% chance of happening.

the coin flip analogies are the least useful by far. the games are not random variables and portraying a streak as if it had to be random chance isn't neccesary to defend Hikaru.

Maybe the flipping of a coin analogy would work if Hikaru was going against someone of the same or at the very least a similar range of rating to him. Even then, people have good days, people have bad days. Some people play better or worse when playing against a specific opponent. Tilt is a major factor as well. Heck, even chess engines have their strengths and weaknesses in specific positions that can cause them to underperform. tldr; chess isn't simple enough to be able to be represented as just x coin flips.

This depends on the figures. There have been cheaters caught because the "luck" they generated had an extremely low probability happening in the lifetime of the entire universe. That's when you can start counting individual events.

very true! people dont understand statistics at all :D

That would be true if the probability of 90% he used in the model was accurate. That seems way too high, so it should probably be an expected event every once in a while, but certainly not every 20 or so days.

It’s because he played 3 particularly “weak” in relation to his FIDE rating. When he played a GM immediately after, he lost twice. Kramnik really sounds low IQ with his analysis. This was even difficult math

So Hikaru is sandbagging. Got it.

​@@HiFiAwardTourthat's not what sandbagging is lmao, sandbagging is losing on purpose to tank your rating, he is farming

In fact every ten coin flips have a 1 in 1024 of occurring.

But two sides of coins is equals and Hikaru and this 3-4 dudes not equals and Hikaru have more ELO. So, chance to get win steak is more likely.

Well said

Exactly. I did this in middle school as part of a group project and was one of the kids who got a ten of a kind. I think we saw three total out of the 1024 sets.

​@@archsys307its 50/50 you either get 10 in a row or not. Easy math

you don't have to be clueless to be a cheater. The better you are as a player, the easier it would be to understand what stockfish wants you to do and describe that to other people. So explaining moves doesn't really clear hikaru, I don't think he's cheating because of his skill in fast situations where it would be too hard to use stockfish.

​@@ragegaze3482Hikaru has been playing at a super high level online for almost 20 years, I've never found him suspicious, and he can back the level up in bullet too....if he was using an engine in bullet it would be obvious...

@@dexio8601 I 100% don't believe that Hikaru is cheating, but I also think it's dangerous for people to say that he could/would never cheat. His playing should be respectfully scrutinised for cheating, along with all of the other top players, but it should be done with the presumption of innocence.

@@dexio8601 he didnt say he was cheating or even suggest it

@@dexio8601 I agree but I think the point is that hikarus playstyle is completely opposed to what one would think a cheater would play like. Someone using an engine would not be one of the best time scramblers with a few seconds left. that in fact should be their weakest side.

This is every game of low elo league of legends bronze or lower. (It tends to happen less often in silver for some reason, not sure how much it applies on upwards since I'm bad and permastuck silver.)

ye more importantly the money from the views lmao

nah he sees the opportunity from being accused in such a weird way to milk youtube views. I would not blame him.

@@kheireddineattala1281 He's a streamer/content creator, not a chess professional, so obviously.

lol he's jealous they're not using his games... he even asked them to look at his games in his last rant... the true motivation! XD

Indeed. He was 2770 Fide when he defeated Kasparov 2848 Fide in a 15 game WC match!

yes every young, top player will annihilate him...

Probably pissed at the fact that hikaru is god tier under time pressure whereas this guy gets flagged by teenagers.

Same with nepo and dubov. Salty russians

​@@rijulgupta1597At the end of the day Kramnik is 3 times World championship a d Hikaru never won a Candidates

@@redeyexxx1841and despite that, he is making a fool of himself

As a chess player, you have to respect Kramnik for his great chess, but as a human, Kramnik has always been a horrible person. Lying and attacking others.

@frogesh Krimnik trying to save online chess. If nothing is done, professional online chess competition will die in a few years. There will be no reasons to pay prize money for seeing how cheater plays versus cheater.

he got mentally ill (paranoid), it al started when he lost to a lower rated player (the fact the Kramnik himself didnt play well, he cannot see that

​@@lukaswolek7294I dont have to respect anyone. Where is this law I must respect Kramnik?

It's not his fault. It's part of the russian culture to covet other people's belongings and achievements. Have some tolerance for those poor guys, geez...

Animals, like chess players, fight over mating privileges very often. They are driven by high testisterone which depletes on defeat so the animal or player may walk away and rest for the next challenge. I think this is an important factor of why women are worse in chess.

This is a very simple take on the issue tho because many more hormones are involved such as dopamine, adrenaline etc. The base line of my speech is that chess players are just animals that push wood dolls around.

@@paulgoogol2652 wow, you are right about it bro

The difference between conditional and unconditional probability or probabilities of correlated events are very very basic statistics and Kramnik just doesn't understand them, it's mind-boggling

@@za5528 he is a former world champion, the one who dethroned KASPAROV But you????? you are a L

Absolutely agree with you.

I would even argue and say the Generation before Kramnik with Kasparov, Anand, Karpov, Tal was stronger than him

@@piratemercenaryhajrudin6893 Ofcourse

No. He puts a shadow of doubt on himself. Nakamuras reputation as a chess nerd is indisputable.

I think that this is his intention. To be charitable to Kramkik, rather than him having some kind of moment, he might actually be trying to trash top level online chess.

Kramnik isnt stupid, beleave me... If he told someone cheated, it might be 99% correct...

​@@turtle207so Naka and Magnus is that 1% that didnt cheat?

@@turtle207 if you have some spare time , I would like to sell you a bridge

At some point you have to conclude that Kramnik is not handling retirement well. Plus I think he is a little butthurt that chess isn't about him anymore like it was 2000-2007 or so. At some point, you pass the baton to the next generation, that's called chess.

"Me, myself, and I all agree that Hikaru is cheating"

​@@knownas2017😂

Absolutely ! I don’t know why there’s a first such comment, but the speed of his moves is faster than any chess engine, even stockfish thinks on each move for around a second, for Hikaru it takes around 0.3-0.5 of a second very often

@@RomgyrTheBiancotrue but he could still use it to find if the position is actually winning or not and find the right line and just calculate on his own. Then he could communicate that however and move quickly. I don’t think he’s cheating tho just saying he definitely could tho just like anyone else.

@@leonardoruiz2793 at the level of super gm, you need to give the AI like at LEAST 3+ seconds or more for it to calculate at the level that super gm's calculate. It would be really obvious if he was constantly doing something like that

Note: probability is one of the fields I like the least and I’m by no means an expert, but it does have some very cool results to read about!

He gave a big interview to Levitov ~ week ago. His method makes sense. Of course you cannot accuse somebody based on these numbers (and as far as I know Kramnik didn't accuse anybody), but you can make some conclusions, like that some people are starting to win in row last rounds of events if they contend for prizes.

exactly! @@andrewzolotukhin6043

@user-ls3ng5qj2p I agree, but what if very unlikely events are happening multiple times in row?

@@andrewzolotukhin6043 didn't the video point out that we should look at the probability of winning from all his games, not just a portion because that will yield higher results if I'm not mistaken? And yes, he indirectly accused *someone* of cheating, but I think it's either Eric, Levy, or someone who pointed out that the games Kramnik referred to were in correlation with Hikaru's.

Yeah it's not fair that Hikaru gets help from Hikaru!

He made sure there was only one bad guy.

This is quality material. Hikaru would be a fool not to profit off these accusations

​@@SINQUEFIELD83 I mean, when your enemy tries to set you on fire, but only ends up burning himself, is there anything wrong with offering all your supporters popcorn and s'mores they can enjoy toasting in the flames? 😂

Hikaru is a very classy guy. Which is def one of the reasons he's my favorite player... not to mention that his chess skills are just mind blowing.

That sounds awfully familiar. Interesting. @@lightyagami1752

I think a lot of people would find that interesting 😉

Well he has doing it for years, if you watch his postmatch interviews...

IQ (pride x butthurt) Based on how Kramnik percieves math, he probably thinks him being wrong is such a statistical outlier that it should never happen

Exactly. And when you cherry pick a dataset, it’s like waiting until AFTER someone gets hit by lightning and saying “the probability of this happening is so low, you must have cheated”.

Yeah, I used to work with someone who had been hit by lightning twice. Odd bloke. And another who had lost half a leg to a shark bite. What's the probability of either of these, let alone both?

acquisition

Monty hall problem is very counter intuitive to the untrained laymen. You can't blame kramnik. He isn't trained in anything besides chess, which he isn't even among the best now

That mathematician is obviously an Hikaru fan, he even admit it, and his argues come with phrases like "if I am not mistaken", "if I am not wrong" and so on, it feels like a fast post done with the intention of support someone who he admires adding a few mathematical arguments that are subject to review, and shouldn't be treated as an absolute truth. That said, I am not for or against Hikaru or Kramnik in this matter cause I don't really care, but the maths showed should be checked in deep before taken them as valid, in science (and maths are) the opinion of just one person, no matter who is doesn't count much if it's not supported by more experts in the same field.

Argument from authority fallacy

the fallacy is named "argument from false authority" which is exactly the case since not because a genius in chess says there is cheating in chess it means its true since they have no idea how to cheat well at chess in the first place and no clue in mathematics. @@theKashConnoisseur

Also These statistics compare him to stockfish, but stockfish would destroy liamputnam 40-0 without ever getting in trouble, perfectly outplaying out of the opening, while hikaru often got bad/drawish endgames and won because he is very good under time pressure, which is also not factored in

I m not sure how the probability was calculated. Maybe he thinks that the probability is less than one percent given the sample size of the entirety of his online game history. I am only saying that because the alternative would be that Kramnik is a massive moron if he actually thought an unlikely streak even at less than 1 % in a small sample size which is probably going to happen even more than once in a huge sample size, constitutes evidence of cheating. But then again , the probability that Kramnik is a massive paranoid moron seeing cheating everywhere is more likely than Hikaru cheating so what do i know...

Agreed that you cannot use single instance of 45 successes in 46 trials with p=0.78 because the odds are ≈ 0.0001 and a single cherry pick is not worth concentrating on. But what happens if we allow such comparisons for a play that lasts a year or more? Assumptions: Blitz Elos are representative of Hikaru and opponents, blitz Elo stats adhere to the known math, and Hikaru plays 10x every day on weekdays for a year (2,600 same or similar games). Statistically less than 1% chance of a 45.5/46 happening. How massive a sample do we need? It is likely (over 50% chance) that Hikaru gets a streak of 25 wins within 1600 games, but then he must follow up with a streak of 20/20 after the tie or loss, a 0.7% chance. You can increase the first part with more trials, but you cannot increase the second because the x/x success is required after the tie to complete the sequence. Hikaru very likely did not cheat, but note that I am using known numbers and stats with no assumptions, emotions, history or friendships etc. It is far more likely that a 2700 beats a 2860, and we saw this happen a few times in the past year.

Einstein wasn't bad at math, he didn't get accepted to Zurich university because he failed the entry test in zoology and languages.

If there are a thousand GMs, 1 in 1000 performances should happen every time they all play a tournament together.

Well actually hans have actually had 100% accuracy consistency for like 9 games in a row as some people used engines to check his accuracy and made video about, i don't recall which video but i only remember the persons face slightly

So carlsens accuse of hans might actually be correct

devils advocate for that though is why should they include all his games anyway? That would only be important if you were assuming Hikaru is literally always cheating and wanted to show that he plays consistently above expected performance. In the case of the win streak the idea is that he was using cheats in those games and thats why he won and then turned them back off, because no cheater is always cheating unless they want to get banned and it's far too suspicious.

It's got to be at least several memes by now.

I commented something along these lines on the previous video and you did it better here because you actually explain the why - as to why this is an important detail. I too noticed that nobody seems to mention the fact that he literally only had 3 or 4 different opponents during all 46 games.

This is very true, and Hikaru is especially good at this. He also knows when to play it safe and when to be risky depending on what he knows of his opponent.