Full access at just Rs 3/day

Journalism of Courage

How Magnus Carlsen would need to cheat just once in a game of chess to be invincible

While revealing Hans Niemann cheated much more than he publicly admitted, a new report by chess.com explains the methods of modern-day chess cheating.

Magnus Carlsen watching Hans Niemann at the 2022 edition of the Sinquefield Cup. (Photo: chessphotoshop.com)

An internal report by chess.com, the sport’s most popular digital platform, has revealed that US Grandmaster Hans Niemann cheated much more than he has publicly admitted. Last month, Niemann, 19, had confessed to using illegal means in only a few “random games” after world champion Magnus Carlsen had hinted that he was a cheater.

The report, besides nailing Niemann’s lie, also underlines the overwhelming advantage those assisted by “chess engines” enjoy over their rule-abiding opponents. With apps and software programmed to play like Grandmasters with an unreal rating of 3500-plus and think close to 40 moves in advance, the superiority of Machine over Man has multiplied many folds in recent times.

Also in Explained |Is it easy to cheat at chess? Apparently yes. Here’s how

In an alarming observation that casts a shadow on over-the-board chess too, the report says that cheaters playing at elite level can always get the better of their rivals by indulging in “selective cheating”. As Carlsen is quoted in the report as saying: “I would have just needed to cheat one or two times during the match … That is all I would need in order to be almost invincible.”

Quoting the report to understand modern-day chess cheating:

Subscriber Only Stories

What’s the verdict on Niemann?

“Our investigation has concluded that he did, however, cheat much more than he has publicly admitted to, including in many prize events, at least 25 streamed games, and 100-plus rated games on chess.com, as recently as when he was 17 years old. Hans’ online and OTB behaviours may be completely different, and that should be taken into consideration. We have shared our findings with FIDE and will cooperate with any investigation or requests they pursue.”

Do cheaters take help from chess engines for all their moves?

Not always. The report says that there are two types of cheaters. “Common types of cheating examples range from ‘every move is an engine move’ (i.e., where every moved played was the top move recommended by a chess engine) to ‘we don’t have enough evidence to close’ (i.e., where the player’s moves are unusually sophisticated, but still within realistic bounds of statistical possibility).”

What’s the common modus operandi?

“Some, often newer, players use a chess engine like Stockfish to decide every move they make. This form of cheating is obvious and easy to detect. Other players, especially those that play at Hans’ level, are much more sophisticated, and engage in ‘selective cheating,’ using a chess engine to give advice only in key moments, and often intentionally making sub-par moves to mask their engine use. At the higher echelons of competitive chess, many games are won or lost in a critical moment, and having any sort of assistance during those moments can turn the outcome of the game. Indeed, many top players have expressed that using a chess engine in just a handful of key moments can add hundreds of Elo in strength.”


Are chess engines impossible to beat?

“Human chess and computer chess are different, even at the highest levels. The best humans play at an Elo rating of 2800. ‘Stockfish,’ the most powerful chess engine, has an estimated rating of more than 3500. In a theoretical match between World Champion Magnus Carlsen and Stockfish, we estimate that it is most likely that Magnus Carlsen would lose every single game — no wins and no draws.”

How are chess engines programmed?

“Most chess engines use neural nets which have been trained on millions of top-level chess games to capture the deepest of chess strategic understanding. They also have nearly infallible tactical calculation, as they can look more than 40-plus moves deep into the position and calculate potential outcomes.”

Magnus Carlsen and Hans Hiemann. (Express Illustration: Suvajit Dey)

Why is ‘being human’ a fallibility in chess?

“Humans rely on past experiences, heuristics and guidelines, pattern recognition, and a fallible ability to calculate moves in the future where they cannot see the live position. This leads to humans dismissing potentially winning moves because they look to be bad based on past experience or general chess principles. But a computer is not constrained by these rules and will often make moves a human — even an elite player — would immediately reject. Similarly, even the best chess-playing humans often make moves that seem sound, but are bad due to a series of calculations which show it to be losing.”

How do top players see ‘selective cheating’?

International Master Levy Rozman says, “Top players [only] need to cheat three times a game. Literally. Top players know so much about the position that if you even insinuate that they might be better or worse, they might find the right move.”

World champion Carlsen too is on the same page. “Had I started cheating in a clever manner, I am convinced no one would notice. I would have just needed to cheat one or two times during the match, and I would not even need to be given moves, just the answer on which move was way better, or here there is a possibility of winning, and here you need to be more careful. That is all I would need in order to be almost invincible, which does frighten me.”

First published on: 06-10-2022 at 07:11:03 am
Next Story

A tale of two Sena rallies: Trading fireworks, Uddhav Thackeray guns for BJP, Eknath Shinde claims Bal Thackeray legacy

Next Story