Draws

Should you accept a draw offer?

Updated · Sergei Tovmasian

Short answer

Play on if the position still has an extra pawn, a pawn break or pieces to manoeuvre with, and you have time on the clock. Take it if the position is genuinely dead, if you are the one who is worse, or if you are short of time against a stronger player. Below club level, declining more often is the habit that pays: those positions are still full of mistakes.

Here is the offer I see accepted most often, and it is the one that annoys me most.

8
7
6
5
4
3
2
a
b
c
d
e
f
g
h 1
White is a pawn up, the a-pawn has the whole queenside to itself and Black's h5 pawn needs watching. Rook endgames are drawish, but this one has a plan in it: a4-a5-a6 and Rd7. Play on.

That is a normal club position: White has an extra passed pawn on the wing, Black has a slightly loose kingside, both sides have a rook. Someone offers a draw because rook endings are drawish, and someone accepts because he cannot see a forced win. But nobody wins a rook ending by force at move thirty. You win it by pushing the a-pawn to a6, cutting the king off on the seventh rank, and waiting for one careless move.

Reasons to play on

An extra pawn is a reason. So is a pawn break your opponent cannot meet, a rook on the only open file, a bishop against a knight in a position with pawns on both wings, or a king that can walk somewhere useful while his cannot.

The clock counts too. If he has eight minutes and you have twenty-five, a level position is not level in any way that matters.

And there is the plain statistical point, which nobody enjoys hearing. Below about 1800, games are decided by mistakes and not by evaluations, and a position that is objectively drawn is still a position where one of you will hang something in the next twenty moves. Playing those out is the single cheapest source of rating points there is, which is half the argument in how to break a rating plateau.

Reasons to take it

Take it when the position is genuinely dead: equal material, no breaks, no files, nothing to aim at. Take it when you are worse and the offer is a gift. Take it when you are short of time against a stronger player and know you are about to be outplayed on the increment. Take it in the last round when half a point wins you the section, which is a tournament decision rather than a chess one.

And take it when you are exhausted. Playing on for two more hours in a level ending, then losing on move seventy, then losing again the next morning because you never recovered, is a well-known way to ruin an event.

The offer is information

Somebody offering you a draw is telling you something. Most of the time it means he does not like his position, or he is tired, or his clock is worrying him. So the moment the offer arrives, stop and look at the board again as if you had just sat down. What is he afraid of? That question has found me more wins than it has cost me draws.

The one I still think about is from a junior tournament. Two pawns up against a player rated three hundred points above me, thirty minutes each on the clock, and I accepted a draw because of the number next to his name. I went through it that evening and the win was there three different ways. Nobody ever offers you a draw because they think you are winning by accident.

If you want to make the decision with something better than a feeling, learn the endings the offers turn up in. Half of them are rook endgames, and the other half are positions where the offer is really a perpetual check waiting to happen. The etiquette of saying yes or no is on draw by agreement.

More questions

Is it rude to decline a draw offer?

Not at all. Say no, or just play your move, and get on with the game. Declining is an ordinary part of chess and no strong player takes it personally. The rude version is the one where you ignore the offer, think for half an hour and only then accept, or where you keep offering after being told no once.

Should a beginner accept draws?

Less often than they do. Games below club level are decided by blunders rather than by evaluation, so a level position with pieces still on the board is worth playing out: somebody will go wrong and it is a coin flip which of you it is. Take the draw when you are worse or short of time, not when you are simply unsure.

How do I know if a position is really drawn?

Count what is left. Equal material, no pawn breaks for either side, no open file to fight over and no way for either king to make progress usually means a genuine draw. If you can name one plan and one weakness to aim at, the position is not dead yet, whatever the engine would say about the evaluation.

Go deeper

Convert won positions instead of drawing them. Train the finish in the app.

← All: Draws · Learn chess