Candidate Move Selection
This article is the advanced companion to our Candidate Moves introduction. While that guide covered the basics, this one dives deeper into selection, the art of narrowing multiple candidates to the single best choice through systematic comparison and evaluation.
Generating candidates
The first step is generating good candidates.
- Forcing moves: Checks, captures, and threats, always consider these first.
- Strategic moves: Improving piece placement, creating weaknesses, controlling key squares.
- Prophylactic moves: Preventing the opponent's plans. 4. "Strange" moves: Always include at least one non-obvious option, many brilliancies come from moves that look wrong at first glance.
The comparison method
Don't evaluate each candidate in isolation, compare them against each other. For each pair of candidates, ask: "Which resulting position would I rather play?" This comparative approach is faster and more reliable than trying to assign absolute evaluations to each candidate. Kotov's "tree of analysis" method works best for forcing sequences; the comparison method works best for positional decisions.
Common selection errors
- Falling in love with the first candidate. The first move you consider gets unfair weight, deliberately consider alternatives.
- Ignoring quiet moves. The best move is often a quiet, improving move, not the flashy tactical option.
- Too few candidates. One candidate is not a "choice", generate at least three.
- Too many candidates. More than five candidates wastes time, narrow quickly.
Practice
In your games, deliberately pause before each move and name your candidate moves (mentally or on paper). After the game, check with an engine: was the actual best move among your candidates? If not, why did you miss it? This exercise rapidly improves candidate generation. See also: Finding the Best Move.