Candidate moves: When you see a good move, look for a better one


Table of Contents

Top image: Online ChessQuiz by Laszlo Moldovan (Schaakacademie Apeldoorn)
Bottom image: IM Mark Dvoretsky and GM Artur Yusupov, Apeldoorn 2010 (photo team captain Karel van Delft)

Candidate moves:
When you see a good move, look for a better one

Chessable science team research paper

Authors: Karel van Delft, Sipke Ernst, Dries Wedda

When you see a good move, look for a better one. Look for candidate moves, the moves that deserve attention when you decide about how to continue in a position. But, how to choose them and is this trainable? A workgroup of the Chessable science team did research. One conclusion is that choosing candidate moves is multifaceted and about a concept as well as methods. It is about knowledge and trainable skills. Pattern recognition plays a crucial role in recognizing possibilities and limitations when evaluating positions.

The research consisted of interviewing grandmasters, reviewing chess literature, cataloguing psychological insights, and an online test.

The workgroup consisted of Karel van Delft (coordinator), GM Johan Hellsten, Dr. CM Can Kabadayi, Laszlo Moldovan, Dries Wedda, Dr. WIM Alexey Root, Dr. Benjamin Balas, GM of chess composition Yochanan Afek, and GM Sipke Ernst. Consultants were GM Artur Yusupov and IM Merijn van Delft.


The research is introduced by an interview with GM Artur Yusupov. The theme of candidate moves is extremely important for chess players, he says. For beginners the concept is not obvious at all. In his books he describes candidate moves not only as a concept but also as methods. Coming back to the initial position after analysing variations is extremely important in the process of finding a solution. It might change your evaluation and even trigger you to check for new candidate moves. The concept of candidate moves involves aspects of searching strategies and decision making. Research about candidate moves is useful, Yusupov states. “It could lead to methodological advice for training.”
Pattern recognition, seeing quickly what pieces can do, and synthetic qualities are important.
A trainer should encourage students to look for alternative and active possibilities in a position and to look from the perspective of the opponent.
Mental flexibility is important because the situation on the board can be complex and there are different methods to come to a decision.
“Many people think for five seconds and know the answer. One poet said: you have two ears because you have to hear both sides. That is what candidate moves are about.”


The Russian GM Alexander Kotov was the founder of systematic thinking about candidate moves with his book ‘Think like a grandmaster’. He advised examining each branch of the variations only once (the so-called analysis tree). This approach was criticized later by other authors like GM John Nunn, who describes how there is interaction between findings in variations.
Other authors came up with other insights and practical advice, such as: Start with the simplest variation (GM Michal Krasenkow), analyse and make decisions with techniques such as comparison and elimination (IM Mark Dvoretsky), be practical (GM Johan Hellsten), don’t follow immediately your intuition (GM Sam Shankland), and use backwards thinking (GM Stefan Kindermann).

Other authors investigated how chess players think, analyse, evaluate, and make decisions like FM Amatzia Avni (The Grandmaster’s Mind) and GM Andrew Soltis (The wisest things ever said about chess). IM Willy Hendriks (Move First, Think Later) stated: There is no order in the way we look at the board. IM Merijn van Delft emphasized that with developing skills verbalization is an important phase in development. About training GM Josh Friedel says beginners are not considering enough options, and it is important for a trainer to look from the perspective of the trainee.


Scientific research is about description, explanation, predicting, and influencing phenomena.
In this research paper we reflect on what psychology and science can contribute regarding developing thinking skills over the board and training methods. Dr. Benjamin Balas discusses cognitive aspects of perception. He explains that chess expertise leads to a specialized visual vocabulary for summarizing board configurations. Because of cognitive biases, this specialized visual vocabulary may not always lead to better performance.
Although there was not prior scientific research on the application of candidate moves, research has been done about relevant cognitive issues as pattern recognition, perception, heuristics, reasoning, problem solving, decision making, searching strategies, calculation, cognitive biases, memory, intuition, and metacognition. There is also research about skill development, expertise development, and deliberate practice.
Besides cognitive issues, also mental aspects are of influence such as self-knowledge, self-management, mind-set, task-oriented attitude, mental skills and competences, productive thinking routines, self-talk, stress regulation, resilience, motivation, and concentration. These aspects can be related to social aspects like learning by observing, role models, and social interaction.
Physical aspects play a role in how chess players perform. Sometimes coincidence plays a role.


For an online study with 207 chess players, who tried to solve 10 provided positions, candidate moves were defined as the moves that deserve attention. The hypothesis was that chess players make better decisions when prompted to consider candidate moves. When taking the online ‘ChessQuiz’ (created by Laszlo Moldovan), participants experienced one of two conditions: prompted to consider candidate moves or asked to find the best move. The research group did not find statistical evidence that prompting participants to use candidate moves improves performance.

Possible reasons are discussed. One potential limitation has to do with the nature of our control group. In the control group, participants may have known about candidate moves and may have used this method to some degree.

Participants in the experimental group may not have been applying the candidate move method as we intended. Though the use of candidate moves may be beneficial if one has some training in how to apply it, it is possible that any participants in our experimental group who had not used this method before found it confusing or cumbersome, weakening their performance.

Taken together, these observations lead us to wonder if our explaining in just a few sentences what the candidate move method is may have not been sufficient to improve performance in our task.


GM of chess composition Yochanan Afek states endgame studies are very rich in candidate moves and make good training material. This is why top trainers use them regularly.
The computer era is reflected in developments in the way chess players think. Thinking over the board has become more concrete, young Dutch grandmasters Liam Vrolijk and Thomas Beerdsen say.


GM Alexander Kotov stated much could be won if chess training would be based more on scientific insights. IM Mark Dvoretsky advised developing more techniques based on psychological insights. Some cognitive issues are well researched, others less so. Based on psychological and other scientific insights, combined with best sport practices (not only in chess), it seems possible to develop more techniques to improve chess performance. That’s why the Chessable science team made a start with the development of ‘thinking tools’, such as using an analysis questionnaire, diary, creative tools, etc. These thinking tools will be part of a new Chessable course about candidate moves.

About the authors:

Karel van Delft is psychologist and the science project manager of Chessable. He is the author of the book ‘Chess For Educators’ and co-author of the book ‘Developing Chess Talent’. He runs Schaakacademie Apeldoorn (Chess Academy Apeldoorn, in The Netherlands.

GM Sipke Ernst is a FIDE chess trainer in The Netherlands. He is author of the Chessable course Seriously Shock the Semi-Slav.

Dries Wedda studies Artificial Intelligence at the Rijksuniversiteit Groningen. He is an online international chess trainer from The Netherlands.

Correspondence: [email protected]

Link to the full report.

Was this helpful? Share it with a friend :)

4.9 with 3.65K user reviews

Check them on individual course pages