How to Analyze a Chess Position: A Step-by-Step Checklist
To analyze a chess position, run a five-point checklist before you calculate: king safety, material, piece activity, pawn structure, and immediate threats. Only then list your candidate moves and calculate the two or three that matter.
The five-point evaluation checklist
Before you calculate a single line, take a static reading of the position. Strong players do this almost instantly; you build the reflex by doing it deliberately. Run these five in order:
| # | Check | Ask yourself |
|---|---|---|
| 1 | King safety | Whose king is more exposed? Are there open lines toward either king? |
| 2 | Material | Is the material equal? If not, does the side with less have compensation? |
| 3 | Piece activity | Whose pieces are doing more? Any bad bishop, dominant knight, or passive rook? |
| 4 | Pawn structure | Weak squares, isolated or doubled pawns, passed pawns, pawn majorities? |
| 5 | Threats | What does my opponent threaten right now, and what do I threaten? |
King safety comes first on purpose: a material or structural edge means nothing if your king is about to be mated. Only after this static read do you start calculating.
The order is not arbitrary. It runs from the concerns that can end the game immediately down to the ones that shape the long game. King safety and threats are urgent — miss them and nothing else matters. Material, activity, and structure are the slower-burning factors that decide who wins if nobody gets mated. Run them in this sequence every time and you stop the classic amateur error: falling in love with a beautiful plan while your opponent is one move from checkmating you.
The same checklist is the backbone of reviewing your finished games, not just live positions. When you go back over a loss, running these five questions at each critical moment shows you exactly which one you neglected — and it is usually the same one, game after game. That is the bridge between analyzing a single position and finding your recurring weakness; see how to find your chess weaknesses.
Candidate moves: Kotov's method
Alexander Kotov, in Think Like a Grandmaster, gave the most quoted advice on calculation: at each critical moment, first identify your candidate moves — the two, three, or four moves worth considering — then calculate each one once, cleanly, without hopping back and forth. He called the wandering-mind alternative "checking the same variation five times and never the important one."
In practice this means: pause, name your candidates out loud in your head, then take them one at a time. The discipline matters more than the depth. Amateurs lose more from calculating the wrong move deeply than from calculating the right move shallowly. This is exactly the calculation habit that reviewing your own games trains fastest.
A practical rule for finding candidates: look at every check, capture, and threat first — for both sides. Forcing moves narrow the position and are easy to miss when you are focused on your own idea. If a candidate move is a check or a capture, it deserves at least a glance even if it looks ugly, because forcing moves are where tactics live. Only once you have scanned the forcing options should you consider the quiet, positional candidates. This one habit catches most of the tactics club players hang.
Silman's imbalances: turning a read into a plan
Jeremy Silman reframed positional understanding around imbalances — the concrete differences between the two sides that a plan can exploit. His list overlaps the checklist above: minor-piece quality (bishop vs knight), pawn structure, space, material, king safety, and control of key files or squares.
The insight is that a plan should flow from an imbalance, not from a vague wish. If you own the only good knight and your opponent has a bad bishop, your plan is to keep the position closed and trade off their good pieces. If you have a queenside pawn majority, your plan is to march it. Read the imbalance, and the plan writes itself — which beats staring at the board hoping for inspiration.
Silman's practical advice was to make a short list of every imbalance in the position, favorable and unfavorable, before choosing a plan. Then you play toward the side of the board where your imbalances are strongest. This also tells you what to avoid: if the only imbalance in your favor is a space advantage, trading pieces relieves your opponent's cramp and throws your one edge away. Knowing which imbalance you own keeps you from making the trade that helps your opponent — a mistake even strong club players make routinely.
A worked example, in words
Picture a middlegame from a Queen's Gambit Declined. Both kings have castled short and are safe — king safety is roughly equal, so it does not decide anything yet. Material is level. Now the imbalances: you have a knight on an outpost at e5 that cannot be kicked by a pawn, while your opponent's light-squared bishop is hemmed in behind its own pawn chain — a classic bad bishop. On piece activity, you are clearly better.
The structure confirms the plan: the closed center means the bad bishop stays bad. So your candidate moves are the ones that keep the position closed and pile pressure on a fixed weakness — not the ones that open lines and free the enemy bishop. You calculate the two moves that improve your worst-placed piece and leave theirs stuck. Notice you reached a plan without a single engine line: the checklist and the imbalances did the work. The engine's job comes next — to confirm the concrete tactics hold.
Now flip the same position and imagine the tempting-but-wrong path. You see a chance to open the center with a pawn break, calculate a flashy line, and feel good about it. But the static read already warned you: opening the center hands your opponent's bad bishop the diagonals it was missing, trading your one real advantage for a few forcing moves that fizzle. This is the payoff of evaluating before you calculate. The checklist stops you from calculating your way into a worse position — the single most common way club players talk themselves out of a good game.
When to trust the engine, when to trust yourself
An engine and your own judgment answer different questions, and confusing them is a common way to stop improving.
- Trust the engine to check whether a concrete tactical line actually works — it calculates flawlessly, and it has topped computer rating lists for years. If you think a sacrifice is sound, the engine is the referee.
- Trust your own judgment for the plan, the understanding, and the "why." The engine will tell you the best move; it will not, on its own, teach you the idea you can reuse next game.
- Do not memorize engine moves you do not understand. A move you cannot explain is a move you will not find again over the board.
The strongest way to improve is to evaluate the position yourself first, commit to a move and a reason, and only then let the engine grade you. The gap between your reasoning and the engine's verdict is the lesson. A chess analysis app that classifies your mistakes can turn those gaps into a study plan instead of a pile of evals.
Frequently Asked Questions
How do you evaluate a chess position?
Run a static checklist before calculating: king safety first, then material, piece activity, pawn structure, and immediate threats. This tells you who stands better and why. Then identify your candidate moves and calculate the two or three that matter, one at a time. The static read gives you the plan; the calculation confirms it works. Doing the read before you calculate is what separates purposeful analysis from moving pieces around in your head.
What are candidate moves in chess?
Candidate moves are the handful of moves genuinely worth considering in a given position — usually two to four. The term comes from Alexander Kotov, who advised identifying all your candidates first, then calculating each exactly once without jumping between them. The point is discipline: most amateurs calculate the wrong move in great depth and never seriously look at the move that mattered. Naming your candidates before you calculate fixes that.
Should I use an engine to analyze positions?
Use it as a referee, not a teacher. Evaluate the position yourself first — commit to a move and a reason — then let the engine check whether your line works. Engines calculate flawlessly and have topped rating lists for years, so they are perfect for verifying tactics. But an engine move you cannot explain is one you will not reproduce over the board. Do your own thinking first; use the engine to grade it.
How do strong players analyze a position so fast?
They have internalized the same checklist you run deliberately — king safety, material, activity, structure, threats — so the static read happens almost instantly. They also recognize the position's imbalances (in Silman's sense) and know the standard plans that follow, so they are choosing between a few sensible candidate moves rather than starting from scratch. It looks like speed, but it is really pattern recognition built from analyzing thousands of positions.
