You calculated everything - or so you thought. Then the blunder: "how could I have missed that?" Rarely because you ran out of computing time; almost always because the brain made the call with a bias, not with a calculation error. Cognitive science has mapped these drifts; five of them stick especially well to the chessboard.

Bias #1: the Einstellung effect (the first-idea problem)

The Einstellung effect is probably the most studied cognitive bias in chess. It refers to the tendency to cling to the first solution that comes to mind, even when a better one exists.

The term comes from German and means "set" or "fixed mental attitude." It was studied directly in chess by Merim Bilalić and colleagues in a famous series of experiments published in 2008.

In their protocol, chess problems contained two solutions: a "familiar" solution (a standard mate the players knew well) and an "optimal" solution (a faster, more elegant one). Expert players, capable of finding the optimal solution in other contexts, systematically missed the better solution when the familiar one was available, even after being explicitly warned that a better solution existed.

Brain imaging (fMRI) showed that in these conditions, brain regions linked to inhibition of the first response were in hyperactivation, and that despite that, the familiar solution kept "invading" thought.

How it shows up in a game: You see a tactical combination that looks strong. You analyze it for several minutes. The combination works but is not the best option. A better idea was available from the start, but your brain, "fixed" on the first idea, blocked the exploration of alternatives. You play the correct-but-suboptimal combination, or worse, a variant calculated from the first idea with an error.

How to counter it: After finding a good option, force yourself explicitly to look for a better one before committing to calculation. The mental formula "is there something better?" as a systematic question before any deep calculation. That question opens the "inhibitory breach" needed to explore alternatives.

Bias #2: confirmation bias (seeing what you want to see)

Confirmation bias is the tendency to seek out, interpret and remember information in a way that confirms pre-existing beliefs, while ignoring contradictory information.

Daniel Kahneman, in Thinking, Fast and Slow, describes it as one of the most universal cognitive biases and among the most resistant to correction. It affects novices and experts alike in their domain.

How it shows up in a game: You decide that the opponent is preparing a kingside attack. You start analyzing moves through that filter. Moves that confirm your hypothesis are noticed and remembered. Moves that point to a completely different strategy (a central attack, a minority plan) are under-weighted or ignored. You play against a threat the opponent was not preparing, leaving their real intention unanswered.

This bias is amplified in positions where you have invested time in an analysis. The more you have calculated an idea, the harder it is to abandon it even faced with contrary evidence.

How to counter it: Practice "reverse" analysis. After forming a hypothesis about your opponent's intention, actively look for evidence that it is wrong. "What would show me I'm wrong?" is a defensive question against confirmation bias. Force yourself to find at least one argument against your own plan before playing it.

Bias #3: anchoring (the first moves trap you in an evaluation)

Anchoring bias is the tendency to give excessive weight to the first piece of information received (the "anchor") when making decisions, even when later information should change the evaluation.

Kahneman and Tversky demonstrated the effect in many contexts. A number presented first influences subsequent estimates even when it is clearly arbitrary. Adjustment from the anchor is systematically insufficient.

How it shows up in a game: You evaluate the position after move 15 as slightly favorable to you (+0.3 in engine terms). By move 25, after several exchanges, the nature of the position has changed: it is now slightly unfavorable (-0.5). But your evaluation is "anchored" on the positive assessment from move 15. You underestimate the change in the position's nature, play with optimism in a position that calls for caution, and blunder because you have not updated your evaluation correctly.

Opening anchoring is a special case. If you play a variation considered "equal" in theory, you tend to evaluate the position as equal even when precise moves have created a real imbalance. The theoretical anchor resists positional updating.

How to counter it: Train yourself to periodically "wipe" your evaluation and rebuild it from the current position, with no reference to the previous assessment. Ask yourself: "If I were seeing this position for the first time, without knowing how I got here, how would I evaluate it?" That question forces an evaluation without an anchor.

Bias #4: overconfidence (after a run of good moves)

Overconfidence bias is the tendency to overestimate one's abilities, the quality of one's information and the accuracy of one's predictions. It is documented in practically every decision domain studied, from doctors to engineers to financiers.

In chess specifically, overconfidence has a particular trigger: streaks of good moves. After playing well for 10 consecutive moves, confirmed by an objectively better position, the brain enters a state of heightened confidence. That state reduces vigilance and calculation depth on the following moves.

How it shows up in a game: You played a beautiful combination, won a piece, and your position is clearly superior. You are in a state of "strategic euphoria." You play the next move quickly, without deep calculation, because you feel "in the flow." But that next move hides a counter-move you did not see. The position flips. The blunder lands precisely in the most favorable position, because that is where you were least vigilant.

This bias is amplified when you are a spectator of your own good form: "I'm playing well today, so my calculations are reliable" is a false reasoning chain.

How to counter it: Keep a constant level of vigilance regardless of how you evaluate the position. Winning positions need as much attention as losing ones. The simple rule: never play a "routine" move in a superior position. Every move is the most important of the game.

Bias #5: the availability heuristic (you fear what you remember easily)

The availability heuristic describes the tendency to overestimate the probability of easily recalled events, usually because they are recent or emotionally salient.

Kahneman and Tversky showed that people judge the frequency and probability of events based on how easily examples come to mind, rather than on an objective statistical assessment.

How it shows up in a game: You recently lost a game to a knight sacrifice on h7. That defeat is emotionally salient. In your next game, you obsessively watch for h7 sacrifices even when they are not in the air at all, because the recent example is very "available" in memory. While you watch that imaginary threat, you miss a completely different one you would normally have seen.

The availability bias can also work in reverse: if you have never suffered a certain type of combination, you tend to underestimate it because it is not available in your emotional memory, even if it is theoretically known.

How to counter it: Regularly practice varied problem types, not just focusing on your recent weak points, but covering a broad spectrum systematically. Varied tactical training calibrates the "availability" of patterns on a more balanced sample rather than on recent memories. (These biases also explain why blunders always seem to fall at the worst moment: Murphy's law in chess analyzes that phenomenon.)

The meta-bias: believing you are immune

A final bias deserves a mention, even though it is not specific to chess: the bias blind spot, the tendency to recognize cognitive biases in others while believing oneself immune.

Pronin, Lin and Ross showed that individuals systematically perceive themselves as less subject to cognitive biases than the average of their peers. That overestimation of one's own rationality is itself a bias.

For the chess player, the consequence is important: knowing these five biases is not enough to be immune. Theoretical knowledge and in-game behavior are two different things. Explicit behavioral protocols (the systematic questions mentioned in each section) are needed for theoretical knowledge to translate into different behavior at the board.

Chess thinking is not just calculation training. It is the training of vigilance over your own cognitive processes. That is what psychologists call metacognition: thinking about your own thinking. And it may be the most valuable skill chess can develop.

After reading: pick one bias that has cost you a recent game; over the next three sessions, ask the same control question before every important move (e.g., "what other idea am I dismissing?" for Einstellung).

Key takeaways

  • Cognitive biases are systematic deviations from rational reasoning, documented by psychology research
  • They are not a sign of stupidity: they affect everyone, including experts
  • In chess, they show up in identifiable and predictable ways
  • The first step in countering them is recognizing them in real time

Sources and references

  • Bilalić, M., McLeod, P., & Gobet, F. Why Good Thoughts Block Better Ones: The Mechanism of the Pernicious Einstellung (Set) Effect. Cognition, 108(3), 652-661, 2008. (Foundational study on the Einstellung effect in chess with brain imaging.)
  • Kahneman, D. Thinking, Fast and Slow. Farrar, Straus and Giroux, 2011. (The complete synthesis of cognitive biases, including anchoring, overconfidence and availability.)
  • Kahneman, D., & Tversky, A. Judgment under Uncertainty: Heuristics and Biases. Science, 185(4157), 1124-1131, 1974. (The founding paper on heuristics and cognitive biases.)
  • Pronin, E., Lin, D. Y., & Ross, L. The Bias Blind Spot: Perceptions of Bias in Self Versus Others. Personality and Social Psychology Bulletin, 28(3), 369-381, 2002. (The blind-spot bias and the tendency to see oneself as less biased than others.)
  • Gobet, F., & Simon, H. A. Expert Chess Memory: Revisiting the Chunking Hypothesis. Memory, 6(3), 225-255, 1998. (Expert memory and pattern recognition in chess, context for understanding fixation biases.)