Since Nakamura havent explained presicely due to low understanding of the subject I suspect, what was he calculating, I ASSUME he (his guys) were calculating MATH PROBABILITY of this overperformance. Because that is the only thing matters in this test. Now start to doubt if he meant this or just EXPECTED score, which DOESNT HAVE ANY RELEVANCE actually in this particular case calculating the probability of some result. Not sure he or his " data scientists" know themselves what did they examine in fact, but either way it has nothing to do with professional statistical and mathematical research
@@anatolyl what makes you think we didn't do that? And btw it is more complicated than you write here,there are nuances, just wanted to show how amateurish and ridiculous HIS calculations are. Ours will publish anytime AFTER chess com or its hero presents their
Just publish full data (game number - game result - player's rating) of his series in last year (7 or 8 long series over 800 games, as far as I remember), and do a small program that computes it's probablilty based on Monte-Carlo simulation. I could do this in few hours if I had the data and exact probability model for win/loss/draw based on rating difference. Then everyone will understand this is impossible, because you will get 10^-10 or 10^-20 probability (in the later case, you won't even get the event happening in the simulation process). And you don't have to be world top statistician to do it, or understand the program (which will be probably 20 lines of code). This will be absolute, 100% proof for everyone who knows math.
You are my biggest inspiration. Please never quit your content, is nice for the people that know to appreciate it. Don't focus only on the haters, you have a lot of fans as well!
Good evening Vladimir. I appreciate the intelligent and logical points that you make. They are not based on emotion or some baseless number crunching and (possibly more importantly) you do not ever try to undermine your opposition immediately by discrediting them in some way as the first course of action. I will look forward to the next parts in your series. Thank you for taking the time to film your clips.
4:01 "I don't make these movements you know I don't try to look well to speak like this "you know guys I'm very pleasant man you know you like me you see I'm handsome you know" no sorry uh that is not my way of doing it".. haaaaaaaahaha.
Hello Mr. Kramnik. It was really interesting video, thank you for your job and investigation. I am on the same page with you, because I am from Ukraine and understanding your views about the difference between "blaming" and "assumption/suspection", but according to American culture I think that is why Hikaru didn't catch it well. Or, as you said, its only deliberately manipulation. I wish you good luck in this topic and hope that truth will be coming! P.S. Hello from Lviv :)
С одной стороны интересно наблюдать, как Владимир Борисович объясняет элементарные вещи по этике, шахматам и статистике. С другой - не покидает ощущение, что всё это лишено смысла. В современном мире в моде супергерои, подавляющему большинству плевать, какая там вероятность. Если Хикару выдаст 100 побед подряд в классику - этот мир будет аплодировать, а не сомневаться. Чесском уверенно на этом зарабатывает. Хочешь быть как Хикару - оформи подписку. Сейчас его за уши тянут к победе в турнире претендентов, которая на 99% сделает его чемпионом мира. Доходы чесскома от этого вырастут в разы. Что бы там Эрик не говорил про неважность денег, именно они являются главным мотивом авторов вселенной Марвела и стримов Накамуры.
Согласен полностью,кроме того,что бессмысленное занятие :). С небольшими шансами на успех,согласен,но иначе обречены на поражение точно, а так еще посмотрим
Последние годы замечаю что многие гроссмейстеры относятся к Хикару неоднозначно, у многих он заблокирован, многие не хотят с ним иметь никаких дел. Наверное не просто так.
Thank you for the mathematical explanations. I'am just wondering if there is a more theoretical approach than Monte-Carlo simulations to evaluate the streak percentage. When trying to implement the algorithm I was also stucked to compute the probability of a drawn game knowing the winning probability distributionn as a function of elo distribution. Still a long way to go be be up to date at math :-).
Владимир Борисович, спасибо за Вашу деятельность и борьбу с "особо честными" игроками виртуального пространства! К сожалению, таких становится все больше и больше в шахматном мире на самых разных уровнях, от любителей до гроссов.
I commend Vladimir's patience in dealing with this thoroughly repugnant person. Hikaru Nakamura is an individual utterly devoid of class. And that's being kind to him.
Wow it takes a lot of courage for anyone to tell the truth about the most toxic person in chess (Nakamura). Respect Big Vlad for fighting for the beautiful game of chess!
Love your game commentary champ. I have a hard time with the idea that Hikaru is cheating. He makes money from sponsors and streaming not prize winnings, what's the motive?
Этому сериалу по борьбе с читерством с Владимиром Борисовичем в главной роли как борца с читерами нужен экшн и новые герои как положительные так и отрицательные
I think there is something wrong with the cited article (or its interpretation) at around the 1hr 13 minute mark. If one looks closely at the table, one sees that for each rating difference N, the probabilities pf the stronger player winning and the weaker player winning add up to 1 - but this would imply the probability of making a draw is 0! For a simple sanity check, the table predicts that you have a 50% chance of winning if your opponent is rated the same as you. This is obviously not true. I think the table must be reinterpreted as the chances of NOT LOSING as opposed to winning.
Could be, it is not clear there, but in any case, if using Gliko ratings and system as we should of course here, the win probability in a single game is around 84 percent in this streak. Which doesn't make it same but lower and lower probability of the performance WITH EACH consecutive win. In general, expected performance is irrelevant here but winning probability ONLY, so Hikarus great examination is just ABSURD
@@VBKramnik Sure, I agree the Gliko is the more relevant measure here. I was only making the point that with FIDE ratings, the expected score of 89% sounds about right; I'm not commenting on its implications to the streak.
@@vikramnadig794 I think the way those percentages are meant is the probability of a decisive result being in favour of one player or the other. Draws aren't counted, as in the case of a draw the players' ratings are adjusted accordingly a little closer to one another. So if we have a rating difference of 300, it means that a decisive result will be in favour of the low rated player 15.1% of the time, and the decisive result will be in favour of the high rated player 84.9% of the time. It's not saying the probability of making a draw is 0, it's just about win/loss outcomes. Consider that this is just about rating _difference_ but not rating _level._ So the numbers work out for a 800 vs 500 rated player, or a 1800 vs 1500, or 2800 vs 2500. All have a 300 point difference, but obviously different chances of drawing, as high level play tends towards a draw much more often than low level play. The Elo system isn't about how often a draw occurs, it's about who wins with rating differentials. Just my understanding, I could be wrong.
Mr. Kramnik. Hikaru Nakamura has arguably the 2 most important games of his career tomorrow and the day after tomorrow. Choosing to publish this video at this time is not a classy move by you.
LIke the Video HIkaru made against Kramnik? Why should he care about the career of someone who has ruined the respectability of chess over the years and turned into a game only cheating kids respect? Its the weekend, he didn't schedule the tournament. Nothing to do with him.
@@kohnr3381 I just looked it up. He did that before his game at the WC against Ding. TO be honest I didn't watch it because I was pissed Magnus wasn't in it. But this is definitely no different and seems like Karma.
Владим Борисыч, раньше я тебя хейтил, но уважал, а теперь просто уважаю. Даже если ты и в чем-то не прав, но ведь ты же веришь в то, что ты делаешь! Успехов и удачи в этой неравной битве. Надеюсь, что как можно людей перестанут быть твоими хейтерами, как я.
If Nakamura has found a way to flag them because of his speed, then this is not chess rating, but some kind of mouse competition rating and there he may have such results? Also maybe he drinks some red bull and then he has these streaks. So not easy to prove.
confirmation bias is very annoying lol never realised there were so many people on the other side supporting kramnik but i will watch the video with an open mind
That is because you live in the websites streamer social media bubble and echo chamber. Ask yourself why 96% of active titled players don't even play in online tournaments. Or why almost none of the million people playing even watch online tournaments as fans. E-sports have been dead and unmarketable in general for over 30 years because society can't respect the anonymity and no game company has ever fully applied the same principles we apply in real life to keep games honest and fair. Thats all Kramnik is asking for, and its pretty outrageous how this community of mostly children and corrupt streamers are trying to ostracize him. FIDE has already cut ties with the website, so they can stop begging for partnerships. And the CEO Erik Allebest has already said he can't get any sponsors and will probably be leaving the company in 5 years. Just the fact HIkaru sees nothing wrong with farming noobs for rating or smurfing for streaming entertainment, speaks to his immoral and unsporting character. He said himself he stopped playing chess24 events in the past because of their anti cheating measures...
Владимир Борисович, недавно Ханс Ниманн сказал, что Вы были с ним в тренировочном лагере. Что думаете про Ханса сейчас? Мнение поменялось? Успехов в борьбе с читингом!
Could be, it is not clear there, but in any case, if using Gliko ratings and system as we should of course here, the win probability in a single game is around 84 percent in this streak. Which doesn't make it same but lower and lower probability of the performance WITH EACH consecutive win. In general, expected performance is irrelevant here but winning probability ONLY, so Hikarus great examination is just ABSURD
@@VBKramnik I think it is pretty clear. At the bottom of the website it says: "Note: This is often the way this table is used (probability of win), but it is actually more accurate to say that the percentage is your *expected score* against an opponent." Expected performance (or same thing, expected score) is RELEVANT for all the calculations. If the expected score of a player against some opposition is 75% and the actual score is 99/100 that player is overperforming, this is obvious. With the utmost respect, I think you are misusing the term "winning probability" when you actually mean "expected score". Please check this with your team.
No because we are talking about a winning streak therefore every "not win" finishing it. So in this case WINNING PROBABILITY is decisive. What is the point of measuring expected score here? He made 45 out 45 and a draw. Since the formula of calculating probability of 45,5 from 46 is MUCH MORE COMPLICATED, simpler to calculate 45 wins and after make certain adjustment. A bit less presice (doesnt affect much the conclusion) but less chance of a mistake and much faster :). Your expected score is nothing else than result against average rating oposition, so who cares if its 39 or 39,5 if you are examining 45? Expected score here has no relevance simply and it tells nothing in itself about the probability. It is a consequence of your and opponents rating,like the probability itself, the BASE is a winning probability
When an opponent in argument makes ad hominem attacks, such as the data scientist comments that Hikaru made, I consider that an automatic disqualification of their position.
Владимир, спасибо что продолжаете освещение этих проблем. Всё больше людей понимают, что это не параноя, личные обиды или что-то подобное, а систематическая проблема. Ну и содействующие лица в ней все как на подбор... Но хотелось представить вам взгляд со стороны на это видео. Обсуждения "чужих" методик в длинном видео с присутствием небольшого языкового барьера, это корм для ваших оппонентов. Накамура как всегда может сказать "Я не аналитик данных + я не особо понял что он имел ввиду". Лучше не комментировать его чушь, а спрашивать его то, от чего не отвертеться. В этом и сила анализа данных, даже если играть в адвоката дъявола, вывод не меняется и очевиден всем. Ну и обвинения Накамуры тяжело делать, даже эти его стрики я проверил, и игра там стандартная по качеству + формат для его сильных сторон создан. Там нужно нормы подстраивать, делать базисы для дисперсии качества игры на каком-то отрезке партий и т.д. А вот спросить его, почему он выигрывает 50 подряд + столько лет в топе, а ни одна из его серий не превосходит по качеству игры Пешкача и Гальченко. Как Пешкач имеет 14.28 сантипешек в равных позициях и зевает -0.5 в 7.08%, а Хикару в своих стриках не показывал лучше чем 20 сантипешек и 11% частоты зевков? На такое уже "Крамник крейзи" не ответишь, ведь надо будет объяснить как ФМ лучше его в разы. Ну и почему в бане Крамник, а не эти "мастера". Ещё один хороший вопрос касается вашей игры. Как так получается, что у вас игра не проседает по качеству относительно других ГМов, одни из лучших показателей в равных позициях, но разбивка позиций по частоте странная. Как будто вам искусственно порог достижения результата поднимают и хорошая игра не гарантирует стандартные % по win/draw/lose позициям. Тут уже не скажешь что кто-то крейзи, т.к. игра с Пешкачами и Гальченко такие аномалии вызывает. Я постораюсь прикрепить скриншоты в следующем посте, чтобы было наглядно. Но думаю, что это гораздо более действенный способ противодействия их чуши. Я для себя сделал исследование по игрокам кого вы подозревали и легко убедился в вашей правоте и то что интуиция вас не подвела. Вот пусть Накамура, вместо обзывательств и обид расскажет всем, как "ФМ 2400 на ничью ОТБ" в онлайне играет на порядок лучше его в 40+ стриках побед. И чем занимается его любимый знаток анализа данных чесском, если это всё можно одному на ПК проверить за неделю. Тогда их докапывания до вас будут бесполезными, т.к. вы никак на эти данные не влияете и от вашего мнения или гипотезы они не изменятся. Изаиняюсь за длинный пост и "советы", но очень хотелось бы увидеть как вы вскроете некомпетентность чесскома и тд, спрятанных за наглой уверенностью в своей правоте или вообще в злом умысле.
GM V. Kraminik in the documentary final gambit communism in check GM V. Anand said the documentary with a spectacular phrase that nowadays this generation plays chess for fun. I'm 45 years old and have been part of the chess community since I was a child. The times of chess and cultural change will never return. 🙄🤔
Hikaru has always used dirty tricks (you can remember the cases with Aronyan or Nepo), this is his element, manipulation, deceiving people, cheap posing, pathetic insults
You said they emphasized the cheating accusation because of PR tactics, but you have to understand that's also just great for content/views on youtube (clickbait).
Было бы корректно сравнить рейтинг оппонентов Магнуса на кубке мира и тех с кем Хикару играл во время Страйка , так как перфоманс легче показать с более слабыми игроками
Нет,во первых, рейтинг перформанс учитывает любую рейтинговую разницу, и достигнуть некий рейтинг перформанс не легче или сложней с разными соперниками Втопое, не станем действовать,как Хикару и смешивать классику и блиц. Хоть перформанс Карлсена на Кубке Мира,конечно, тоже на порядок ниже
Мне кажется,Хикару должен как Ханс начать учить русский язык.Даже необязательно много учить,можно начать с одной известной фразы: "Молчание - золото" Ну а так,возможно,прикольный видеоролик,хоть и долгий,но осознаю,что для понимания всей "картины мира" пояснения необходимы,что и выливается в хронометраж. Спасибо!
Still waiting for anyone of the top 100 players except for Vladimir to openly state that there is the 'anomaly problem'. Modern players are too concentrated on the possible money problems than on the fair play.
MVL, Fabiano, Dubov, Nepo have spoken up about it. Even hikaru sid before he got sued and was focus of suspicions himself. Most titled players just don't play in tournaments online. It's not taken seriously.
Связи никакой, материал подготовил и выпустил,когда был готов Другой разговор,если бы поприличней был персонаж, тогда подождал бы до понедельника. А чувства этого товарища,как он воспримет и тд, мне безразличны. Но не подстраивал специальным образом, много чести слишком 😊
he is the reason I watched my first titled tuesday, after playing on the website for almost 3 years. lmao. I also refuse to play on the website though. Got banned for similar reasons as Vlad "community abuse", basically complaining about fair play measures. refused my 2nd chance, made another account with the same name just to close it myself and now I just play on lichess ever since. Quality of matchups have gone down past year but games I still feel are more consistently competitive there compared to com. They take it a little more seriously but at my level only so much they can do with all the anonymity.
Here are some key data points from the results of statistical analysis: • At an Elo rating of 3100, the probability is approximately 0.000191% • At an Elo rating of 3300, the probability increases to about 2.25% • At an Elo rating of 3400, the probability reaches about 15.98% (significant improvement) So, to achieve a "realistic" probability of around 1% to 5% for scoring 45.5 out of 46 games, a player needs an Elo rating in the vicinity of 3300 to 3350.
@@mgoogyi Here are the probabilities for achieving exact streaks of wins, given a win probability per game of 0.9: Probability of a 30-game winning streak: about 4.24% Probability of a 40-game winning streak: about 1.48% Probability of a 50-game winning streak: about 0.52% These probabilities reflect the chance of achieving exactly those streaks within any specific set of consecutive games. However, to understand the likelihood of observing at least one such streak over the course of 5000 games, we'd need to consider the entire sequence and calculate how the odds increase as more opportunities for streaks arise. The complete calculation would involve a more complex analysis, involving simulation to account for the overlapping probabilities of streaks throughout the 5000 games. But these figures give a basic sense of how likely it is to achieve such streaks purely by chance, assuming a high constant win rate. The probabilities are non-negligible, so seeing multiple streaks in a large number of games could indeed happen naturally, depending on the player's consistent performance and the strength of their opposition.
@@mgoogyi Here are the probabilities for achieving exact streaks of wins, given a win probability per game of 0.9: Probability of a 30-game winning streak: about 4.24% Probability of a 40-game winning streak: about 1.48% Probability of a 50-game winning streak: about 0.52%
1:23:10 Dear Kramnik, I think you can make a video series "Every Russian Schoolboy Know About Maths" :) As an American HN should not mass with an ex-soviet "comrade", as HN says, on these subjects, ha ha. I heard once from great mathematician Vladimir Arnold that American pupils sum rational numbers like (2/3)+(3/5)=5/8. Of corse a joke.
Thanks Vlad for being Honest...ANother interesting topic is how in a world, females with a -2200 fide performs like 90% precision in games bullet TT and Titled Tuesday...I guess is for those 100 bucks prize or in order to get games against some strong titled players
I write comments supporting Kramnik (without abusing) - Teacher deletes comments People write aginst Kramnik - Teacher keeps comments and even replies Not Fair 😥
Владимир Борисович, хорошая с Вашей стороны затравочка перед завтрашней партией - японца совесть замучит, если она у него есть; но корона чесскомовская на голове есть точно. P.S. Оскорблять оппонента в матерной форме в ситуации, когда у тебя нет весомых аргументов супротив, да ещё и экс-чемпиона мира, - это стыд и позор, хотя он этого и не осознает, по всей видимости.
Ладно, это правда умора когда Хикару тупо взял среднее и сказал, что вероятность стрика такая же как и в одной игре. Но на самом деле интересно глянуть на большой выборке количество таких стриков и провести аккуратный стат тест. Хотя и так вроде очевидно, что это оверперформанс, причем маловероятный.
Ну что, Накамурушка, самурайчик наш, не изведал ты ещё силушки нашей богатырской? Сможешь ли увильнуть от правды пылающей? Да восстанет справедливость от богатыря нашего, да сгинет во веки веков ложь хитрого самурайчика.
Владимир Борисович, при всём уважении ваши доводы тоже вилами на воде писаны. То, что вы сделали называется в науке cherry-picking нельзя рассматривать 46 партий и удивляться малой вероятности заработать 45.5 очков в отрыве от остальных игр. Всем очевидно 20 раз увидеть подряд орла жутко малая вероятность, но если смотреть на серию из 1000 испытаний то 20 раз подряд орёл уже не так маловероятен. Мало того, фразы в стиле "я никого не обвиняю, просто обращаю внимание" очень некрасиво звучат. К примеру, можно накопать различной притянутой за уши или просто выдуманной статистики "во всех городах где Крамник выступал в дни выступлений поднимался уровень ограблений" , и в конце добавить "я конечно не намекаю, что Крамник по вечерам обворовывал старушек в перерывах между играми, но хочу обратить внимание на эту связь". Ваш изначальный твит выглядит именно так.
So he tried to spin it that you were questioning why his online rating is so high compared to FIDE rating? when that doesn't have anything to do with anything? Shows what he thinks of his fans too and maybe they actually do indeed believe that lol. I always found it extremely strange that he claims online chess, where he farms noobs for rating and smurfs for streaming entertainment, made him a better chess player. Because that simply goes against Human nature. Vidit on the other hand said the complete opposite after winning the Grand Swiss. He said he got better at chess the past few years because he stopped streaming, and that makes more common sports sense.
It's impossible for me to delete comments but it's even impossible for Vladimir to delete comments that have already been posted. If you are referring to that thread we were debating in, the original poster must have deleted their own original comment out of Shame. Maybe they work for the website, or is a huge hikaru fan, and they realized that it was looking bad. Trust me she did the right thing because it would have looked worse for you to leave that conversation up and I wish she did. Also my friend you are the one that reports people's comments to try and cheat at a debate online like you almost certainly cheat at the game. The only reason I know this cuz the only comments that I get hidden are the ones that I try to reply to your name. Smh....
@cooloutac what?! I report you?? I was wondering why you said that first time. This is absolutely wrong and absurd. I wouldn't report you even if you were rude, which you weren't. I believe in freedom of speech which means everyone can speak even if they are wrong! And of course it is the host who is quite able to delete messages and I bet he does.
@cooloutac what a gisgusting thing it is to accuse me of reporting to cheat in a debate! I have my full right to say that a liar expects everyone else to lie, and you know the rest ;) I had no idea there is a way to do this in a debate and I am able to win in an argument by reasoning, like I did to you. By the way, the original poster was him, not her. Funny how you are wrong each time you make an assumption.
Накамуре стоит обратится к своим корням и вспомнить как карается в Японии такое неподобающее обращение к старшим, которое этот американец позволяет себе. Я уже молчу про то, что Крамник Чемпион Мира и Легендарный Грандмастер !
Great video showing the fundamental mathematical error made by Hikaru and co leading to an incorrect calculated probability for his streak. The error was they averaged his opponents ratings and calculated the probability of him not losing one game (streak of 1) and they confused that number with his probability to score a streak of 46 games in a row which branches each single game having a new chance to win or draw/lose adding or cutting his streak respectively. The streak probability would be the chance of him winning a single game powered by the length of the streak. To illustrate let us assume that Hikaru has a 0.9 (9 out of 10) chance of winning a game against an opponent that has the average rating of his opponents during that 46 games winning streak, for Hikaru win his first game his chances are 0.9 to streak 2 games the chances for him are 0.81 (0.9 x 0.9) he need to win his first game (0.9 chance) and then his second (0.9 chance) and to streak 10 games his chances are 0.348. That means if Hikaru tried to adopt the average rated opponent 100 times he is expected to succeed 35 times, but his chance to win any single game against his opponent is still 0.9. For the 46 games winning streak, using Hikaru's 89% chance of winning a single match the expected result is 0.0047. Meaning that if Hikaru played 200 times 46 games in a row he is expected to win them all only in a single run of the 200. The run is impressive and interesting because such run is mathematically considered rare to achieve. Hikaru is a great player, but I never got why he has made such a big of a deal about Vlad's post beside content creation. Vlad never accused him of anything he only found the streak very rare to achieve according to the math and asked for an examination. The way Hikaru and chesscom took Vlad post is illogical to say the least and not say not very innocent in the means of exploiting the situation to create some random content showing no class or respect toward the other being in the right or the wrong.
@@mgoogyi Yes. I guess you didn't understand both Vlad in the three part video and my comment. Of course such streak is achievable. You didn't get in my previous comment the meaning of if Hikaru played 200 times 46 games he is expected to win 46 in a row only once. That means in 200 x 46 = 9200 games played he is expected (according to Elo' s expected score which is their probability of winning plus half their probability of drawing) to score 46 in a row only once (it can happen during your 5000 consecutive games this is not the anomaly). Hikaru has almost 30000 blitz games, scoring such a run is in fact very likely (according to the total number of blitz games played), but dude is throwing such rare streaks left and right with even more improbable performances 3600+. Hikaru had 16 improbable performances for the 30000 games since last year in less than 4000 games played against a 2900 chesscom rating average opponent. The anomaly lies in multiple improbable performances not a single rare run.
Владимир Борисович, судя по дате выкладки видео, Вы это сделали намеренно, чтобы доставить Хикару дискомфорт на турнире претендентов. Но ладно бы ваше видео было бы правдивым. Однако оно - лишь компот из фраз и цифр, которые вы просто рандомно брали и подтасовывали под свои доводы.
"randomly took" hahah. right.... 4 or 5 almost 50 game streaks in almost as little as 30 days is outrageous enough. But when the website doesn't even address it that is just wild and obscene. I think there is a reason Erik Allebest said FIDE cut all ties with them, online chess is unmarketable and they can't get any sponsors for tournaments and he will probably be leaving the company in 5 years. HIkaru is part of the problem. I don't know who suggested him to do speedruns or farm players for ratings for entertainment as acceptable, but its turning chess into another cheating kids game like every other e-sport. IMO, him and ALireza don't even deserve to be in the candidates. They are terrible for the integrity of the sport. It would be like Hans winning the candidates. Maybe the streaming community will be happy if HIkaru wins, but I think we should start picking out gravestones for chess if he does.
@@crashbandicoot5941 google my name, look at my profile details. Grow up and have a mind of your own. Your comment is nothing more than a public concession to every mature person reading.
@@crashbandicoot5941 grow up. did you even look up my name or check my profile details? smh.... get out of this bubbled echo chamber you live in kiddo.
reporting peoples comments is what a cheater would do bud. At least google my name or look up my profile details. smh.... Learn to think for yourself because otherwise, you are the real "bot"...
Я думаю, что будущее шахмат будет все больше и больше смещаться в сторону компьютерного мира, и люди будут все больше и больше играть на bot-engines. люди будут искать адверсорал аттаки в engins, найти недостатки в движке, а не в самих людях. Но это очень футуристический взгляд.
I will write it in english, Im half-russian, half-german. So what I meant by this, is that stockfish is basically now a neural network. There is a way to attack an neural network, by an "Adversarial Attack". There is even a paper about AlphaZero losing most of the game to another angine, because of an adversarial attack
millions play online e-sports. But they are unmarketable as professional sports unfortunately because of too many people who act unsporting so they have no fans that suppor them. And what you are describing is not what we call a competitive sport bud.
Summary: Hikarus thinks 0.87 to the 41st power = 0.87 In fact 0.87 to the 41st power = 0.0033 In other words, Hikaru's win rate was either a one chance in 300 extreme fluke. Or something more seriously suspect.
Владимир Борисович, огромное спасибо за ваши видео. Всегда приятно послушать рассуждения умного и образованного человека. Не останавливайтесь творить благие дела во благо шахмат.
I don't understand what is the end goal of this? You think it's possible to ban someone like Hikaru for cheating without any physical evidence? Unrealistic. But it's a good way to grow your TH-cam channel though 🤔
![โอ้มายก๊อด! ไม่ถึงชั่วโมงหมดเป็นแสน 😅 [cc] แดนแพทตี้ SS2 | EP.42 |](http://i.ytimg.com/vi/OH2ShXfO6tA/mqdefault.jpg)
Since Nakamura havent explained presicely due to low understanding of the subject I suspect, what was he calculating, I ASSUME he (his guys) were calculating MATH PROBABILITY of this overperformance. Because that is the only thing matters in this test. Now start to doubt if he meant this or just EXPECTED score, which DOESNT HAVE ANY RELEVANCE actually in this particular case calculating the probability of some result. Not sure he or his " data scientists" know themselves what did they examine in fact, but either way it has nothing to do with professional statistical and mathematical research
💯
@@anatolyl what makes you think we didn't do that? And btw it is more complicated than you write here,there are nuances, just wanted to show how amateurish and ridiculous HIS calculations are. Ours will publish anytime AFTER chess com or its hero presents their
@@VBKramnik show it then, if you DID do it lmao.
@@dominiks5068 I WILL DO WHAT AND WHEN I DECIDE TO, NOT YOU ☝
Just publish full data (game number - game result - player's rating) of his series in last year (7 or 8 long series over 800 games, as far as I remember), and do a small program that computes it's probablilty based on Monte-Carlo simulation. I could do this in few hours if I had the data and exact probability model for win/loss/draw based on rating difference. Then everyone will understand this is impossible, because you will get 10^-10 or 10^-20 probability (in the later case, you won't even get the event happening in the simulation process). And you don't have to be world top statistician to do it, or understand the program (which will be probably 20 lines of code). This will be absolute, 100% proof for everyone who knows math.
You are my biggest inspiration. Please never quit your content, is nice for the people that know to appreciate it. Don't focus only on the haters, you have a lot of fans as well!
Good evening Vladimir. I appreciate the intelligent and logical points that you make. They are not based on emotion or some baseless number crunching and (possibly more importantly) you do not ever try to undermine your opposition immediately by discrediting them in some way as the first course of action. I will look forward to the next parts in your series. Thank you for taking the time to film your clips.
4:01 "I don't make these movements you know I don't try to look well to speak like this "you know guys I'm very pleasant man you know you like me you see I'm handsome you know" no sorry uh that is not my way of doing it".. haaaaaaaahaha.
Не по уведомлению, а по зову сердца ❤❤❤
Hello Mr. Kramnik. It was really interesting video, thank you for your job and investigation. I am on the same page with you, because I am from Ukraine and understanding your views about the difference between "blaming" and "assumption/suspection", but according to American culture I think that is why Hikaru didn't catch it well. Or, as you said, its only deliberately manipulation. I wish you good luck in this topic and hope that truth will be coming!
P.S. Hello from Lviv :)
Thank you very much for your commitment Mr. Kramnik.
It is very honorable that you are speaking out publicly on this subject.
ENGLISH, thanks God.
Хикару думал что история так просто закончится, но нет. Крамник только начал.
41:02 "don't please tell me or Nepo what should we do. You are not our daddy". Владимир Борисович, гениально! 👍😂😂😂
I love Kramnik. He has add so much to the chess lingo latley: the procedure, interesting, its all clear now, the untouchables.
Finally an English video!! Thank you
А я всё думал, какой мне фильм на ночь сегодня посмотреть🤣🤣🤣
Следим вместе с Вами ❤
Будете ли Вы комментировать финиш Турнира Претендентов?
I feel Levitov’s advice behind media manipulation deck of arguments :) go VB!
"not our daddy" 🤣
С одной стороны интересно наблюдать, как Владимир Борисович объясняет элементарные вещи по этике, шахматам и статистике. С другой - не покидает ощущение, что всё это лишено смысла. В современном мире в моде супергерои, подавляющему большинству плевать, какая там вероятность. Если Хикару выдаст 100 побед подряд в классику - этот мир будет аплодировать, а не сомневаться. Чесском уверенно на этом зарабатывает. Хочешь быть как Хикару - оформи подписку. Сейчас его за уши тянут к победе в турнире претендентов, которая на 99% сделает его чемпионом мира. Доходы чесскома от этого вырастут в разы. Что бы там Эрик не говорил про неважность денег, именно они являются главным мотивом авторов вселенной Марвела и стримов Накамуры.
Согласен полностью,кроме того,что бессмысленное занятие :). С небольшими шансами на успех,согласен,но иначе обречены на поражение точно, а так еще посмотрим
All honest, rational chess players support what you are doing. Thank you, VK!
Interesting analysis.. i noticed when he posted the original video that he was lying about knowing anything about data science
Последние годы замечаю что многие гроссмейстеры относятся к Хикару неоднозначно, у многих он заблокирован, многие не хотят с ним иметь никаких дел. Наверное не просто так.
I'll look at the story tomorrow and then comment. I find his perspectives very interesting. Hopefully we'll still see him playing chess.
everything is clear to me now
🙏 Back to English 🙏
Thank you for the mathematical explanations. I'am just wondering if there is a more theoretical approach than Monte-Carlo simulations to evaluate the streak percentage. When trying to implement the algorithm I was also stucked to compute the probability of a drawn game knowing the winning probability distributionn as a function of elo distribution. Still a long way to go be be up to date at math :-).
It isnt trivial,precise calcilations here
Right now Hikaru is manipulating everyone by saying that Ian stole his preparations just before the game. Such a bad person.
"Common sense" is the idea that a thought need not be justified with reason or data. That is not scientific.
Владимир Борисович, спасибо за Вашу деятельность и борьбу с "особо честными" игроками виртуального пространства! К сожалению, таких становится все больше и больше в шахматном мире на самых разных уровнях, от любителей до гроссов.
Владимир, сколько бы собак не лаяло про то, что вы уже дед и не понимаете в шахматах, мы все равно будем вас поддерживать! Спасибо за ваш труд!
"Do not believe blindly anyone" including the author.
Agree, that is what I mentioned there
HIkaru clearly in the wrong.
I commend Vladimir's patience in dealing with this thoroughly repugnant person. Hikaru Nakamura is an individual utterly devoid of class. And that's being kind to him.
Хикару еще не понял куда он попал.
Good stuff!
Wow it takes a lot of courage for anyone to tell the truth about the most toxic person in chess (Nakamura). Respect Big Vlad for fighting for the beautiful game of chess!
Впервые в жизни пожалел, что не учил английский)))
Subtitle
He is top 2 blitz players in the world (2874 elo).
Love your game commentary champ. I have a hard time with the idea that Hikaru is cheating. He makes money from sponsors and streaming not prize winnings, what's the motive?
Этому сериалу по борьбе с читерством с Владимиром Борисовичем в главной роли как борца с читерами нужен экшн и новые герои как положительные так и отрицательные
Напомнило анекдот
Обьявление
" Ищем мужчину импотента на роль отрицательного героя в порнофильме" 😊
Пока только плохиши всплывают: Пешкач, Гальченко и др.
I think there is something wrong with the cited article (or its interpretation) at around the 1hr 13 minute mark. If one looks closely at the table, one sees that for each rating difference N, the probabilities pf the stronger player winning and the weaker player winning add up to 1 - but this would imply the probability of making a draw is 0! For a simple sanity check, the table predicts that you have a 50% chance of winning if your opponent is rated the same as you. This is obviously not true.
I think the table must be reinterpreted as the chances of NOT LOSING as opposed to winning.
Could be, it is not clear there, but in any case, if using Gliko ratings and system as we should of course here, the win probability in a single game is around 84 percent in this streak. Which doesn't make it same but lower and lower probability of the performance WITH EACH consecutive win. In general, expected performance is irrelevant here but winning probability ONLY, so Hikarus great examination is just ABSURD
@@VBKramnik Sure, I agree the Gliko is the more relevant measure here. I was only making the point that with FIDE ratings, the expected score of 89% sounds about right; I'm not commenting on its implications to the streak.
@@vikramnadig794 I think the way those percentages are meant is the probability of a decisive result being in favour of one player or the other. Draws aren't counted, as in the case of a draw the players' ratings are adjusted accordingly a little closer to one another. So if we have a rating difference of 300, it means that a decisive result will be in favour of the low rated player 15.1% of the time, and the decisive result will be in favour of the high rated player 84.9% of the time. It's not saying the probability of making a draw is 0, it's just about win/loss outcomes. Consider that this is just about rating _difference_ but not rating _level._ So the numbers work out for a 800 vs 500 rated player, or a 1800 vs 1500, or 2800 vs 2500. All have a 300 point difference, but obviously different chances of drawing, as high level play tends towards a draw much more often than low level play. The Elo system isn't about how often a draw occurs, it's about who wins with rating differentials. Just my understanding, I could be wrong.
Mr. Kramnik. Hikaru Nakamura has arguably the 2 most important games of his career tomorrow and the day after tomorrow. Choosing to publish this video at this time is not a classy move by you.
LIke the Video HIkaru made against Kramnik? Why should he care about the career of someone who has ruined the respectability of chess over the years and turned into a game only cheating kids respect? Its the weekend, he didn't schedule the tournament. Nothing to do with him.
Unlike Hikaru who said that Ian stole his preparation? Why is Hikaru lying again before an important game for Ian????
@@kohnr3381 wait what? LMAO. was this on twitter or something?
@@kohnr3381 I just looked it up. He did that before his game at the WC against Ding. TO be honest I didn't watch it because I was pissed Magnus wasn't in it. But this is definitely no different and seems like Karma.
And what's the Hikaru doing waiting Ian as opponent in pretendent tournament?? Argue Ian fair play? Or it's just fantasy?
Владим Борисыч, раньше я тебя хейтил, но уважал, а теперь просто уважаю. Даже если ты и в чем-то не прав, но ведь ты же веришь в то, что ты делаешь! Успехов и удачи в этой неравной битве. Надеюсь, что как можно людей перестанут быть твоими хейтерами, как я.
Kramnik daddy come to save online chess and punish cheaters, good job
Jokes aside, can we all agree that Vladimir is quite handsome?
Definitely he is
Why do you, as a man, care about the beauty of a great grandmaster?
@@honeycatcher9565 you female?
@@chrepuhon no 🙄
@@MADDPOL watch the video and you will understand
Yes!! Kramnik in English. Cheers Big Man.
If Nakamura has found a way to flag them because of his speed, then this is not chess rating, but some kind of mouse competition rating and there he may have such results? Also maybe he drinks some red bull and then he has these streaks. So not easy to prove.
He wasn't flagging there,check games
@@VBKramnik OK. GM Kramnik, it is great honour that you are making these videos for us.
Заснул под видео и первые 20 минут слушал голос Крамника прямо во сне
confirmation bias is very annoying lol never realised there were so many people on the other side supporting kramnik but i will watch the video with an open mind
That is because you live in the websites streamer social media bubble and echo chamber. Ask yourself why 96% of active titled players don't even play in online tournaments. Or why almost none of the million people playing even watch online tournaments as fans. E-sports have been dead and unmarketable in general for over 30 years because society can't respect the anonymity and no game company has ever fully applied the same principles we apply in real life to keep games honest and fair. Thats all Kramnik is asking for, and its pretty outrageous how this community of mostly children and corrupt streamers are trying to ostracize him. FIDE has already cut ties with the website, so they can stop begging for partnerships. And the CEO Erik Allebest has already said he can't get any sponsors and will probably be leaving the company in 5 years. Just the fact HIkaru sees nothing wrong with farming noobs for rating or smurfing for streaming entertainment, speaks to his immoral and unsporting character. He said himself he stopped playing chess24 events in the past because of their anti cheating measures...
233 comments, it's not that many people, really just a small group of autists.
Владимир Борисович, недавно Ханс Ниманн сказал, что Вы были с ним в тренировочном лагере. Что думаете про Ханса сейчас? Мнение поменялось?
Успехов в борьбе с читингом!
Ну мы несколько дней поработали, деталей не раскрываю никогда, этический кодекс
rank #92 OTB. rank #3 online and winning titled tuesdays like Magnus or HIkaru. right....
1:13:55 This is a mistake. 80.83% is the EXPECTED SCORE. Draws ARE COUNTED just like wins and losses are.
Could be, it is not clear there, but in any case, if using Gliko ratings and system as we should of course here, the win probability in a single game is around 84 percent in this streak. Which doesn't make it same but lower and lower probability of the performance WITH EACH consecutive win. In general, expected performance is irrelevant here but winning probability ONLY, so Hikarus great examination is just ABSURD
@@VBKramnik I think it is pretty clear. At the bottom of the website it says: "Note: This is often the way this table is used (probability of win), but it is actually more accurate to say that the percentage is your *expected score* against an opponent."
Expected performance (or same thing, expected score) is RELEVANT for all the calculations. If the expected score of a player against some opposition is 75% and the actual score is 99/100 that player is overperforming, this is obvious.
With the utmost respect, I think you are misusing the term "winning probability" when you actually mean "expected score". Please check this with your team.
No because we are talking about a winning streak therefore every "not win" finishing it. So in this case WINNING PROBABILITY is decisive. What is the point of measuring expected score here? He made 45 out 45 and a draw. Since the formula of calculating probability of 45,5 from 46 is MUCH MORE COMPLICATED, simpler to calculate 45 wins and after make certain adjustment. A bit less presice (doesnt affect much the conclusion) but less chance of a mistake and much faster :).
Your expected score is nothing else than result against average rating oposition, so who cares if its 39 or 39,5 if you are examining 45? Expected score here has no relevance simply and it tells nothing in itself about the probability. It is a consequence of your and opponents rating,like the probability itself, the BASE is a winning probability
When an opponent in argument makes ad hominem attacks, such as the data scientist comments that Hikaru made, I consider that an automatic disqualification of their position.
Владимир, спасибо что продолжаете освещение этих проблем. Всё больше людей понимают, что это не параноя, личные обиды или что-то подобное, а систематическая проблема. Ну и содействующие лица в ней все как на подбор...
Но хотелось представить вам взгляд со стороны на это видео. Обсуждения "чужих" методик в длинном видео с присутствием небольшого языкового барьера, это корм для ваших оппонентов. Накамура как всегда может сказать "Я не аналитик данных + я не особо понял что он имел ввиду". Лучше не комментировать его чушь, а спрашивать его то, от чего не отвертеться. В этом и сила анализа данных, даже если играть в адвоката дъявола, вывод не меняется и очевиден всем. Ну и обвинения Накамуры тяжело делать, даже эти его стрики я проверил, и игра там стандартная по качеству + формат для его сильных сторон создан. Там нужно нормы подстраивать, делать базисы для дисперсии качества игры на каком-то отрезке партий и т.д.
А вот спросить его, почему он выигрывает 50 подряд + столько лет в топе, а ни одна из его серий не превосходит по качеству игры Пешкача и Гальченко. Как Пешкач имеет 14.28 сантипешек в равных позициях и зевает -0.5 в 7.08%, а Хикару в своих стриках не показывал лучше чем 20 сантипешек и 11% частоты зевков? На такое уже "Крамник крейзи" не ответишь, ведь надо будет объяснить как ФМ лучше его в разы. Ну и почему в бане Крамник, а не эти "мастера".
Ещё один хороший вопрос касается вашей игры. Как так получается, что у вас игра не проседает по качеству относительно других ГМов, одни из лучших показателей в равных позициях, но разбивка позиций по частоте странная. Как будто вам искусственно порог достижения результата поднимают и хорошая игра не гарантирует стандартные % по win/draw/lose позициям. Тут уже не скажешь что кто-то крейзи, т.к. игра с Пешкачами и Гальченко такие аномалии вызывает.
Я постораюсь прикрепить скриншоты в следующем посте, чтобы было наглядно. Но думаю, что это гораздо более действенный способ противодействия их чуши. Я для себя сделал исследование по игрокам кого вы подозревали и легко убедился в вашей правоте и то что интуиция вас не подвела. Вот пусть Накамура, вместо обзывательств и обид расскажет всем, как "ФМ 2400 на ничью ОТБ" в онлайне играет на порядок лучше его в 40+ стриках побед. И чем занимается его любимый знаток анализа данных чесском, если это всё можно одному на ПК проверить за неделю. Тогда их докапывания до вас будут бесполезными, т.к. вы никак на эти данные не влияете и от вашего мнения или гипотезы они не изменятся.
Изаиняюсь за длинный пост и "советы", но очень хотелось бы увидеть как вы вскроете некомпетентность чесскома и тд, спрятанных за наглой уверенностью в своей правоте или вообще в злом умысле.
GM V. Kraminik in the documentary final gambit communism in check GM V. Anand said the documentary with a spectacular phrase that nowadays this generation plays chess for fun. I'm 45 years old and have been part of the chess community since I was a child. The times of chess and cultural change will never return. 🙄🤔
Датасаентисты ческома постарались🤣 🤣
Было бы интересно посмотреть дебаты Грищука с Крамником об онлайн читерстве, Александр приводил логичные контраргументы
Hikaru has always used dirty tricks (you can remember the cases with Aronyan or Nepo), this is his element, manipulation, deceiving people, cheap posing, pathetic insults
I will always be a Kramnik fan.
Let's do the procedure 😅
You said they emphasized the cheating accusation because of PR tactics, but you have to understand that's also just great for content/views on youtube (clickbait).
Vova, you are best!
Never blindly believe anyone
Все таки сила советского образования видна. Основательность. Логика. Научность. Зуммерам на понять
Ну, с учётом того, как ты написал наше поколение.. 🤡
@@chrepuhon ты зуммер что ли?
Было бы корректно сравнить рейтинг оппонентов Магнуса на кубке мира и тех с кем Хикару играл во время Страйка , так как перфоманс легче показать с более слабыми игроками
Нет,во первых, рейтинг перформанс учитывает любую рейтинговую разницу, и достигнуть некий рейтинг перформанс не легче или сложней с разными соперниками
Втопое, не станем действовать,как Хикару и смешивать классику и блиц. Хоть перформанс Карлсена на Кубке Мира,конечно, тоже на порядок ниже
@@VBKramnik , мне казалось что вы сами это утверждали , тогда сорри)
Мне кажется,Хикару должен как Ханс начать учить русский язык.Даже необязательно много учить,можно начать с одной известной фразы: "Молчание - золото"
Ну а так,возможно,прикольный видеоролик,хоть и долгий,но осознаю,что для понимания всей "картины мира" пояснения необходимы,что и выливается в хронометраж.
Спасибо!
We love u champion kramnik
Thank you
41:05 - You are not our daddy - Vladimir Kramnik 2024
Still waiting for anyone of the top 100 players except for Vladimir to openly state that there is the 'anomaly problem'. Modern players are too concentrated on the possible money problems than on the fair play.
MVL, Fabiano, Dubov, Nepo have spoken up about it. Even hikaru sid before he got sued and was focus of suspicions himself. Most titled players just don't play in tournaments online. It's not taken seriously.
Почему Каспаров не борется с читингом? У него же своя платформа игровая, вроде.
Владимир Борисович решил подкрутить этот турнир претендентов :)
Связи никакой, материал подготовил и выпустил,когда был готов
Другой разговор,если бы поприличней был персонаж, тогда подождал бы до понедельника. А чувства этого товарища,как он воспримет и тд, мне безразличны. Но не подстраивал специальным образом, много чести слишком 😊
Как всегда отличное видео, Владимир Борисович. Подскажите, планируете ли в дальнейшем стримить "Cheating tuesday"?
Мне не разрешают играть,спасибо Матвею 😊
he is the reason I watched my first titled tuesday, after playing on the website for almost 3 years. lmao. I also refuse to play on the website though. Got banned for similar reasons as Vlad "community abuse", basically complaining about fair play measures. refused my 2nd chance, made another account with the same name just to close it myself and now I just play on lichess ever since. Quality of matchups have gone down past year but games I still feel are more consistently competitive there compared to com. They take it a little more seriously but at my level only so much they can do with all the anonymity.
NICE thumbnail
Here are some key data points from the results of statistical analysis:
• At an Elo rating of 3100, the probability is approximately 0.000191%
• At an Elo rating of 3300, the probability increases to about 2.25%
• At an Elo rating of 3400, the probability reaches about 15.98% (significant improvement)
So, to achieve a "realistic" probability of around 1% to 5% for scoring 45.5 out of 46 games, a player needs an Elo rating in the vicinity of 3300 to 3350.
@@mgoogyi Here are the probabilities for achieving exact streaks of wins, given a win probability per game of 0.9:
Probability of a 30-game winning streak: about 4.24%
Probability of a 40-game winning streak: about 1.48%
Probability of a 50-game winning streak: about 0.52%
These probabilities reflect the chance of achieving exactly those streaks within any specific set of consecutive games. However, to understand the likelihood of observing at least one such streak over the course of 5000 games, we'd need to consider the entire sequence and calculate how the odds increase as more opportunities for streaks arise.
The complete calculation would involve a more complex analysis, involving simulation to account for the overlapping probabilities of streaks throughout the 5000 games. But these figures give a basic sense of how likely it is to achieve such streaks purely by chance, assuming a high constant win rate. The probabilities are non-negligible, so seeing multiple streaks in a large number of games could indeed happen naturally, depending on the player's consistent performance and the strength of their opposition.
@@mgoogyi Here are the probabilities for achieving exact streaks of wins, given a win probability per game of 0.9:
Probability of a 30-game winning streak: about 4.24%
Probability of a 40-game winning streak: about 1.48%
Probability of a 50-game winning streak: about 0.52%
1:23:10 Dear Kramnik, I think you can make a video series "Every Russian Schoolboy Know About Maths" :) As an American HN should not mass with an ex-soviet "comrade", as HN says, on these subjects, ha ha.
I heard once from great mathematician Vladimir Arnold that American pupils sum rational numbers like (2/3)+(3/5)=5/8. Of corse a joke.
Thanks Vlad for being Honest...ANother interesting topic is how in a world, females with a -2200 fide performs like 90% precision in games bullet TT and Titled Tuesday...I guess is for those 100 bucks prize or in order to get games against some strong titled players
which female is this? lol
GM Vladimir Kramnik's interventions are always very interesting. Tomorrow I'll watch the new episode. I'm excited to see what he presents to us.
👏🏻
I write comments supporting Kramnik (without abusing) - Teacher deletes comments
People write aginst Kramnik - Teacher keeps comments and even replies
Not Fair 😥
I never deleted your comment 🤷♂
@@VBKramnik I dont know why but my comments are getting automatically deleted.. tried multiple times
Great Vlad, the cheaters' impaler every tuesday!
Hikaru as great as he is, cannot hide he was cheating online...
здравствуйте, мне нравится ваш канал. Я уважаю и люблю вас. Пожалуйста, продолжайте делать такие интересныи видео.
So many promises, so disappointing delivery. btw where are the numbers and equations?
Probability speaks for itself...
Владимир Борисович, хорошая с Вашей стороны затравочка перед завтрашней партией - японца совесть замучит, если она у него есть; но корона чесскомовская на голове есть точно.
P.S. Оскорблять оппонента в матерной форме в ситуации, когда у тебя нет весомых аргументов супротив, да ещё и экс-чемпиона мира, - это стыд и позор, хотя он этого и не осознает, по всей видимости.
А кто оскорблял Крамника с матом?
@@user-db3mx8zi8jЛично Хикару Накамура
@@user-db3mx8zi8j , Накамура публично на одном из своих стримов прямым текстом на английском матом послал Крамника по известному направлению.
Хикару просто осёл
also voiced obscenities against Kramnik as well. Its very strange how overly defensive he is on the subject. Then again we can assume why...
А можно краткую версию на русском, хотя бы в двух абзацах?
Накамура будет покаран
Hikaru is boss
Hikaru lost twice to Vidit in the candidates yet is 3600 online lol
Ладно, это правда умора когда Хикару тупо взял среднее и сказал, что вероятность стрика такая же как и в одной игре.
Но на самом деле интересно глянуть на большой выборке количество таких стриков и провести аккуратный стат тест. Хотя и так вроде очевидно, что это оверперформанс, причем маловероятный.
Смотрите твиттер человека kaganskymike
This should have waited until after candidates.
I don't even think Hikaru or Alireza should even be there. Two cheats imo. Terrible for the integrity of the sport.
No.
Ну что, Накамурушка, самурайчик наш, не изведал ты ещё силушки нашей богатырской? Сможешь ли увильнуть от правды пылающей?
Да восстанет справедливость от богатыря нашего, да сгинет во веки веков ложь хитрого самурайчика.
Владимир Борисович, при всём уважении ваши доводы тоже вилами на воде писаны. То, что вы сделали называется в науке cherry-picking нельзя рассматривать 46 партий и удивляться малой вероятности заработать 45.5 очков в отрыве от остальных игр. Всем очевидно 20 раз увидеть подряд орла жутко малая вероятность, но если смотреть на серию из 1000 испытаний то 20 раз подряд орёл уже не так маловероятен.
Мало того, фразы в стиле "я никого не обвиняю, просто обращаю внимание" очень некрасиво звучат. К примеру, можно накопать различной притянутой за уши или просто выдуманной статистики "во всех городах где Крамник выступал в дни выступлений поднимался уровень ограблений" , и в конце добавить "я конечно не намекаю, что Крамник по вечерам обворовывал старушек в перерывах между играми, но хочу обратить внимание на эту связь". Ваш изначальный твит выглядит именно так.
So he tried to spin it that you were questioning why his online rating is so high compared to FIDE rating? when that doesn't have anything to do with anything? Shows what he thinks of his fans too and maybe they actually do indeed believe that lol. I always found it extremely strange that he claims online chess, where he farms noobs for rating and smurfs for streaming entertainment, made him a better chess player. Because that simply goes against Human nature. Vidit on the other hand said the complete opposite after winning the Grand Swiss. He said he got better at chess the past few years because he stopped streaming, and that makes more common sports sense.
dont you think deleting comments that express an opposite opinion looks like faking massive support?
@@user-jg8js6ew9u you actually believe I can delete comments? Lol
It's impossible for me to delete comments but it's even impossible for Vladimir to delete comments that have already been posted. If you are referring to that thread we were debating in, the original poster must have deleted their own original comment out of Shame. Maybe they work for the website, or is a huge hikaru fan, and they realized that it was looking bad. Trust me she did the right thing because it would have looked worse for you to leave that conversation up and I wish she did. Also my friend you are the one that reports people's comments to try and cheat at a debate online like you almost certainly cheat at the game. The only reason I know this cuz the only comments that I get hidden are the ones that I try to reply to your name. Smh....
@cooloutac what?! I report you?? I was wondering why you said that first time. This is absolutely wrong and absurd. I wouldn't report you even if you were rude, which you weren't. I believe in freedom of speech which means everyone can speak even if they are wrong! And of course it is the host who is quite able to delete messages and I bet he does.
@cooloutac what a gisgusting thing it is to accuse me of reporting to cheat in a debate! I have my full right to say that a liar expects everyone else to lie, and you know the rest ;) I had no idea there is a way to do this in a debate and I am able to win in an argument by reasoning, like I did to you. By the way, the original poster was him, not her. Funny how you are wrong each time you make an assumption.
Какое великолепное чувство юмора у ВБ
Накамуре стоит обратится к своим корням и вспомнить как карается в Японии такое неподобающее обращение к старшим, которое этот американец позволяет себе.
Я уже молчу про то, что Крамник Чемпион Мира и Легендарный Грандмастер !
Great video showing the fundamental mathematical error made by Hikaru and co leading to an incorrect calculated probability for his streak.
The error was they averaged his opponents ratings and calculated the probability of him not losing one game (streak of 1) and they confused that number with his probability to score a streak of 46 games in a row which branches each single game having a new chance to win or draw/lose adding or cutting his streak respectively. The streak probability would be the chance of him winning a single game powered by the length of the streak.
To illustrate let us assume that Hikaru has a 0.9 (9 out of 10) chance of winning a game against an opponent that has the average rating of his opponents during that 46 games winning streak, for Hikaru win his first game his chances are 0.9 to streak 2 games the chances for him are 0.81 (0.9 x 0.9) he need to win his first game (0.9 chance) and then his second (0.9 chance) and to streak 10 games his chances are 0.348. That means if Hikaru tried to adopt the average rated opponent 100 times he is expected to succeed 35 times, but his chance to win any single game against his opponent is still 0.9.
For the 46 games winning streak, using Hikaru's 89% chance of winning a single match the expected result is 0.0047. Meaning that if Hikaru played 200 times 46 games in a row he is expected to win them all only in a single run of the 200. The run is impressive and interesting because such run is mathematically considered rare to achieve.
Hikaru is a great player, but I never got why he has made such a big of a deal about Vlad's post beside content creation. Vlad never accused him of anything he only found the streak very rare to achieve according to the math and asked for an examination. The way Hikaru and chesscom took Vlad post is illogical to say the least and not say not very innocent in the means of exploiting the situation to create some random content showing no class or respect toward the other being in the right or the wrong.
@@mgoogyi Yes. I guess you didn't understand both Vlad in the three part video and my comment. Of course such streak is achievable. You didn't get in my previous comment the meaning of if Hikaru played 200 times 46 games he is expected to win 46 in a row only once. That means in 200 x 46 = 9200 games played he is expected (according to Elo' s expected score which is their probability of winning plus half their probability of drawing) to score 46 in a row only once (it can happen during your 5000 consecutive games this is not the anomaly). Hikaru has almost 30000 blitz games, scoring such a run is in fact very likely (according to the total number of blitz games played), but dude is throwing such rare streaks left and right with even more improbable performances 3600+. Hikaru had 16 improbable performances for the 30000 games since last year in less than 4000 games played against a 2900 chesscom rating average opponent. The anomaly lies in multiple improbable performances not a single rare run.
Владимир Борисович, судя по дате выкладки видео, Вы это сделали намеренно, чтобы доставить Хикару дискомфорт на турнире претендентов.
Но ладно бы ваше видео было бы правдивым. Однако оно - лишь компот из фраз и цифр, которые вы просто рандомно брали и подтасовывали под свои доводы.
"randomly took" hahah. right.... 4 or 5 almost 50 game streaks in almost as little as 30 days is outrageous enough. But when the website doesn't even address it that is just wild and obscene. I think there is a reason Erik Allebest said FIDE cut all ties with them, online chess is unmarketable and they can't get any sponsors for tournaments and he will probably be leaving the company in 5 years. HIkaru is part of the problem. I don't know who suggested him to do speedruns or farm players for ratings for entertainment as acceptable, but its turning chess into another cheating kids game like every other e-sport. IMO, him and ALireza don't even deserve to be in the candidates. They are terrible for the integrity of the sport. It would be like Hans winning the candidates. Maybe the streaming community will be happy if HIkaru wins, but I think we should start picking out gravestones for chess if he does.
@@cooloutac mr. Kramnik, its your bot account?
@@crashbandicoot5941 google my name, look at my profile details. Grow up and have a mind of your own. Your comment is nothing more than a public concession to every mature person reading.
@@crashbandicoot5941 grow up. did you even look up my name or check my profile details? smh.... get out of this bubbled echo chamber you live in kiddo.
reporting peoples comments is what a cheater would do bud. At least google my name or look up my profile details. smh.... Learn to think for yourself because otherwise, you are the real "bot"...
Я думаю, что будущее шахмат будет все больше и больше смещаться в сторону компьютерного мира, и люди будут все больше и больше играть на bot-engines. люди будут искать адверсорал аттаки в engins, найти недостатки в движке, а не в самих людях. Но это очень футуристический взгляд.
I will write it in english, Im half-russian, half-german. So what I meant by this, is that stockfish is basically now a neural network. There is a way to attack an neural network, by an "Adversarial Attack". There is even a paper about AlphaZero losing most of the game to another angine, because of an adversarial attack
millions play online e-sports. But they are unmarketable as professional sports unfortunately because of too many people who act unsporting so they have no fans that suppor them. And what you are describing is not what we call a competitive sport bud.
Можно создать программу, которая одновременно считает и оценку позиции, и сложность нахождения правильного ответа
Summary: Hikarus thinks 0.87 to the 41st power = 0.87
In fact 0.87 to the 41st power = 0.0033
In other words, Hikaru's win rate was either a one chance in 300 extreme fluke. Or something more seriously suspect.
Владимир Борисович, вы молодец!
Хикару ведёт себя очень высокомерно, человек- редиска, да ещё и читер. Давно пора было вывести его на чистую воду!
Владимир Борисович, огромное спасибо за ваши видео. Всегда приятно послушать рассуждения умного и образованного человека. Не останавливайтесь творить благие дела во благо шахмат.
I don't understand what is the end goal of this? You think it's possible to ban someone like Hikaru for cheating without any physical evidence? Unrealistic. But it's a good way to grow your TH-cam channel though 🤔
Anyone if (IF ☝️)found guilty must be banned (at minimum). HIkaru,Anand or me, shouldn matter the name
🍿🍿🍿😎