Why Chess Coaches Should Review
More Than the Engine Line.
The engine review tells a coach what the strongest move was. It does not tell the coach why this player did not see it on this day. Five categories of error every post-game review should distinguish.
The post-game review used to be the structural core of competitive chess training. The coach would sit with the student, the game would be reconstructed, candidate moves would be debated, and the missed moves would be located through conversation. The conversation itself was the lesson. The coach was not just delivering the answer. The coach was reading the player.
The engine has changed what that looks like. The strongest move is now retrievable in seconds. The evaluation graph is undeniable. A reasonable instinct is to centre the review on the engine line, namely “here is what was best, you played this instead, let us study the difference”. This is correct as far as it goes. It is also incomplete in a way that compounds across hundreds of training games and produces players who know the answers without knowing themselves at the board.
This article is about what the engine review cannot tell a coach, and what the coach can do about it.
What the engine review does well
Three things, namely it settles the technical verdict in seconds, it removes the need for the coach to be the strongest player in the room (a useful equaliser at higher levels), and it produces undeniable evaluation data the player and coach can both anchor to. None of these is small. None of these existed in this form even fifteen years ago.
The engine review is the cleanest piece of feedback in any competitive practice. The accuracy is settled. The depth is enormous. The cost is zero. A coach who refuses to use the engine is leaving the strongest available technical instrument on the table.
What the engine review cannot tell you
The engine review can tell you that the player missed Nf6 on move 23. It cannot tell you which kind of error that was, namely whether the player did not know the pattern, knew it but miscalculated, knew it and calculated correctly but had no time left to verify, knew it and had time but was thinking about the standings, or knew it and had time and attention but was in their sixth round of the day and physically empty.
Five different errors. Identical engine output. Five different training responses.
A coaching program that responds to all five with the same intervention, namely “study tactics in this opening”, is going to produce strong results for one of the five and modest to negative results for the other four. The player who failed on stamina will be over-trained tactically when what they needed was a stamina protocol. The player who failed on attention will be drilled on a pattern they already knew. The player who failed on time management will study lines they would have found if they had two more minutes on the clock.
Knowledge.
The player did not know the pattern. The move was not in their repertoire of recognised positions. Training response: pattern study, namely tactical motifs, structural plans, endgame technique appropriate to the rating band.
Diagnostic question: “If I gave you the position with unlimited time and no stress, would you find this move?”
Calculation.
The player saw the candidate move, started calculating, and miscounted somewhere in the variation. The pattern was known. The arithmetic broke. Training response: visualisation training, candidate-move discipline, the player verbalising the line out loud during training.
Diagnostic question: “Walk me through what you calculated. Where did you stop?”
Time.
The player saw the candidate, calculated correctly, and ran out of time before they trusted the result. They played a different move because the clock forced a commitment. Training response: time-management protocols, namely move-budgeting per opening phase, blitz training to compress decision cycles.
Diagnostic question: “How much time did you have when you decided? Would another two minutes have changed the move?”
Attention.
The player had the knowledge, the time, and the calculation ability, but their attention was elsewhere, namely on the standings, on the opponent’s reputation, on a previous round. The position was not the most expensive thing in their head. Training response: pre-round attentional protocols, recognising drift mid-game, returning to position-only thinking through deliberate cues.
Diagnostic question: “What were you thinking about right before you played the move?”
Stamina.
The player had everything except the physical and cognitive resources to find the move at this point in this day. Late round of a long tournament. Insufficient sleep. The error was not chess. The error was the body. Training response: sport-science protocols, namely sleep and nutrition discipline during tournaments, physical conditioning, scheduled energy management.
Diagnostic question: “What round was this? When did you last eat? When did you last sleep properly?”
How to use the taxonomy in a session
The five categories are most useful when applied in order during the post-game review, namely starting with the most basic and moving toward the most contextual. Many coaches reverse this order out of instinct, namely they ask “were you tired” first because it feels generous. Reversing the order produces softer conversations and worse training data.
Start at category 01. If the player did not know the pattern, the rest of the diagnostic is irrelevant for this position. Only if the player knew the pattern do you move to category 02. Only if calculation was clean do you move to category 03. The structure protects against false attributions, particularly the natural drift toward attributing every error to “psychology” or “stamina” because those are the most flattering explanations to a developing player.
The result is a single named error per position with a specific training response attached. Not a vague “we will work on it”. A diagnosis and a treatment, the same way medical reasoning works.
Why this matters for academies specifically
Academies are the place where the coaching review architecture either compounds or fragments across many students. A single coach with five students can hold the five-category taxonomy in their head and apply it case by case. An academy with thirty students across multiple coaches needs the framework written down, applied consistently, and reviewed across the cohort. Otherwise students get five different verbal frameworks from five different coaches, and the developmental architecture is incoherent.
The argument I am making in The Human Move is that the coaching review is the place where the engine era either lifts the coach or replaces them. If the review is centred on the engine line, the coach becomes a redundant intermediary. If the review is centred on the five-category diagnosis, the coach is doing what only a coach can do, namely reading the player.
If you remember one thing
Five categories of error look identical on the engine graph. Knowledge, calculation, time, attention, stamina. The engine cannot distinguish between them. The coach can. The post-game review either does that work or it does not. If it does, the coach is irreplaceable. If it does not, the engine is enough.
The training architecture for chess in the post-engine era.
The Human Move takes the five-error taxonomy and extends it into a full chapter on coaching architecture, namely how the post-game review needs to evolve so that the coach does what only the coach can do.
Kerim Demirkol is a competitive chess player, swimmer, triathlete, and Certified Fitness Trainer and Instructor. He is the author of the Lessons from the Board essay series and the book The Human Move.
Discover more from Kerim Demirkol
Subscribe to get the latest posts sent to your email.
Leave a Reply