The Science Behind Candidate Moves
When you look at a chess position, your brain generates a small number of candidate moves from the hundreds of legal options. How? The answer involves pattern recognition, chunking theory, and a form of intuition that neuroscience is only now beginning to understand. Understanding the science can make you a better player.
Chunking: how experts see the board
Herbert Simon and William Chase demonstrated in the 1970s that chess masters perceive positions not as individual pieces but as "chunks", familiar groupings of 4–6 pieces that function as a unit. A kingside castled position with a fianchettoed bishop is one chunk. A pinned knight on f6 with a bishop on g5 is another. These chunks activate associated candidate moves automatically.
See also: What Chess Does to a Champion's Mind for more on the neuroscience of chess expertise.
Pattern-triggered generation
When you see a position, your long-term memory pattern-matches against stored positions. Each match triggers a candidate move (or a plan). A GM with 50,000+ stored patterns generates better candidates faster, not because they think harder, but because they have more patterns to match against. This is why puzzle solving is so effective: each puzzle adds patterns to your library.
The role of intuition
Chess "intuition" is not mystical, it is rapid, unconscious pattern matching. When a GM "feels" that a move is right before calculating it, their brain has matched the position against a stored pattern and generated a candidate below the threshold of conscious awareness. Intuition is earned through thousands of hours of study and play.
How to generate better candidates
- Solve puzzles daily to build your pattern library.
- Study master games to see candidate moves in context.
- Use the candidate-moves method to systematise your conscious generation process.
- Post-game analysis: after each game, check what candidates you considered and which the engine found, the gap reveals missing patterns.