Imagine this: you find a promising move, calculate your opponent’s main reply, and reach a position you like.
Then you check it again.
You consider another move, return to your original choice, and repeat the same variation. Several minutes later, you play the move you wanted to play in the first place.
Was all that thinking necessary?
Sometimes, yes. A second look can reveal a defensive resource or correct a mistake in your calculation. But extra thinking should have a purpose. Otherwise, you are spending time without improving your decision.
My advice is to keep calculating while you have a specific, important question to answer. Once you have answered it well enough for the position and the time available, make your move.
That sounds straightforward. The difficult part is recognising what “well enough” means.
Start by asking what could change your choice.
Suppose you are considering a pawn sacrifice. Your question might be: “After the best defence, can I recover the pawn, or do I have enough activity to justify giving it up?”
If you are considering a queen exchange, it might be: “Can my opponent’s king reach the weak pawns before mine can defend them?”
If you are improving a knight, it might be: “Does moving this piece allow a central break that I currently prevent?”
Each question gives your calculation a direction. You know what you are looking for, and you can recognise when you have found it.
“I want to be completely sure” gives you no such endpoint.
Calculate your opponent’s resistance, including the replies you would prefer them not to play.
Finding an attractive continuation for yourself is only part of the work. Before committing, look for the opponent’s strongest practical objection.
Can they ignore your threat? Exchange your attacking piece? Return material? Insert a check before recapturing? Defend with a quiet move?
You can calculate a beautiful variation ten moves deep and still miss the point if your opponent has a stronger alternative on move two.
For an ordinary position, start with a small shortlist of plausible moves. Expand it if none works or if the position gives you a reason to search further. You do not need to analyse every legal move equally.
The aim is to compare realistic choices against serious resistance.
Finish the relevant sequence before evaluating it.
Suppose a capture appears to win a pawn. Before concluding that you are better, check whether your opponent has an intermediate move and whether your piece can escape afterwards.
Stopping immediately after your own attractive move can leave the most important part of the variation unexamined.
Try to reach a position where you can explain the consequences: the material balance, king safety, piece activity, and any immediate threats.
However, a quiet position is not automatically a safe stopping point.
A king-and-pawn ending may contain no checks or captures for several moves, yet a single king move can determine whether a pawn promotes. Before entering that ending, you may need to calculate the pawn race or establish who wins the opposition.
There is no useful universal answer to “How many moves ahead should I calculate?” The necessary depth depends on the decision you are making.
Notice what has changed since your previous look.
Rechecking is useful when you have found a concrete reason to doubt your analysis.
Perhaps you visualised a bishop on the wrong diagonal. Perhaps you forgot that a pawn was pinned. Perhaps you discovered a move order that changes the result.
Those are good reasons to start again.
But if you repeat the same moves, reach the same position, and make the same assessment, pause. Ask yourself what you expect another repetition to reveal.
If you cannot identify an unresolved question, consider whether it is time to commit.
In his article on time trouble, GM Noël Studer describes a rule suggested by his coach GM Markus Ragger: after thinking for more than 15 minutes, reconsider the logical move and play it if there is no concrete objection. Treat this as an example of a decision checkpoint for a long game; a rapid game requires a much earlier checkpoint.
The clock changes how much certainty you can afford.
Imagine a rapid game in which you have 12 minutes remaining. Spending four minutes on your next move uses one-third of your current time.
That may be worthwhile if you are deciding whether a sacrifice wins or loses. It is harder to justify if you are choosing between two reasonable improving moves and your analysis has revealed no meaningful difference.
Before another long think, ask:
- What remains unresolved?
- How much could it affect the position?
- Is another minute likely to help me resolve it?
- How much time will I need for the rest of the game?
An irreversible decision deserves attention: committing your pawn structure, sacrificing material, or entering a king-and-pawn ending. A routine move can also deserve attention if there is a tactical detail.
Use the features of the position to decide where to spend time.
Before moving, make one final check from your opponent’s side.
Visualise your intended move on the board and ask:
- Have I dealt with the opponent’s existing threat?
- What checks, captures, or direct threats does my move allow?
- What has my piece stopped defending?
- Is there a strong reply I have not considered?
If this reveals a concrete problem, return to calculation. Otherwise, avoid turning the final check into another complete search through every candidate.
You can practise this decision away from the board.
Choose six unfamiliar positions from annotated games, including both tactical and positional decisions. Give yourself two minutes per position. Record your move, your expected best reply, and the question you still need to answer.
Then allow three more minutes and record what changed.
These are training limits, not recommended thinking times for every game position. The purpose is to observe what additional thinking produces.
Did you discover a new defensive resource? Correct your evaluation? Find a better candidate? Or simply repeat your original line?
Afterwards, compare your reasoning with the annotations and engine analysis. Pay attention to where you stopped too early and where you continued without learning anything useful.
Over time, this gives you evidence about your own thinking habits.
At the board, you will sometimes have to act with uncertainty. Make that decision after testing your move against the opponent’s strongest replies and checking the consequences that matter.
When you decide to spend another minute, know what you are trying to find.