Converting a winning position

The most painful tag in the library is threw away a winning position. You did the hard part. You won a piece, or built an attack, or ground out a better endgame. Then, in a handful of moves, it was equal again, or lost. The pattern page has a number that should be printed on every scoresheet: the share of “winning” positions at club level that end without a win.

The cause is not technique in the chess-book sense. It is attention. Something in the head says this is done, and the checks stop.

Rule one: nothing changes

Everything you did to get ahead, keep doing. The hanging piece check. The back rank check. Looking at the opponent's forcing moves. The game is not decided until the opponent has no counterplay left, and the opponent knows that better than you do, because they have nothing to lose.

Rule two: trade pieces, not pawns

Material advantages grow as pieces come off, because the extra piece becomes a larger share of what is left. A bishop up with queens and rooks on is a fight. A bishop up with only kings and pawns is a formality.

The exception is pawns. Do not trade pawns when you are ahead in material, because the endgame you want is one where your extra piece can win their pawns and escort yours. Fewer pawns means fewer targets and, sometimes, a drawn ending with no way to make progress.

Rule three: kill the counterplay before you cash in

Winning players think about what they will win next. Winning players who convert think about what the opponent wants. If the opponent has one idea (a passed pawn, an open file, a knight coming to a good square), spend a move stopping it before you go pawn-hunting. Giving back a pawn, or even the exchange, to remove the last active enemy piece is almost always right when you are clearly ahead.

Rule four: no new weaknesses

The winning side tends to get ambitious with pawns. A pawn push that would be fine in an equal position is a mistake when you are winning, because you do not need it, and every pushed pawn is a square you no longer control. Play the moves that keep everything defended and improve the worst-placed piece. It is boring. It wins.

Rule five: do not rush, do not drift

Two opposite failures. Rushing is the forcing sequence that “must” work because you are winning; it works two times in three and the third time you are lost. Drifting is thirty moves of shuffling because you cannot see a plan and the opponent slowly untangles. The middle path is a short concrete plan: put the rook on the seventh, then bring the king up, then push the a-pawn. Three steps you can actually name.

How to know you are converting well

The evaluation should not move. If the number is going up, fine. If it is flat for ten moves while you trade and improve, fine. If it drops by a pawn on a move that felt natural, that move was the mistake, and it is the one worth repairing.

The tell

You are ahead, and you just played a move without asking what it hangs. Or you are ahead and the opponent's last move created a threat you have not looked at because “it does not matter”. It matters.

The habit

When you notice you are winning, say it to yourself in words: "I am winning. Now I am playing not to lose it." The sentence sounds silly. It is the switch that keeps the checks running. Then repair your squandered-advantage positions: the site puts you back at the exact move where the guard dropped.