Bishop against knight

A game on lichess.org, 2014. A knight faces a bishop, each side has six pawns, and nearly all of them are locked together. The black king has wandered to a4, and the bishop on d3 moves on light squares. White to move.

The knight reached a dark square, and the opponent's bishop stayed on the light ones. That is the whole difference between these pieces: the knight is slow but hits squares of both colours, the bishop is fast but moves on squares of one colour all game long. No bishop move could stop the mate: the dark c5-square was out of its reach from the very start of the game. In this chapter we look at when each of these features matters more.

How they differ

The bishop and the knight are minor pieces. They are usually counted as equal, three pawns each, as in the chapter on piece values. The engine rates the bishop slightly higher than the knight. But this average value says little about a particular position. In one position the knight is stronger than the bishop by almost a pawn, in another it is the other way round.

The bishop is long-range. From one edge of the board it reaches the other, and it can defend a pawn on one wing while attacking on the other. But the bishop moves on squares of one colour all game: half the board is closed to it for good. The knight moves only a short way: to cross from one wing to the other it needs three or four moves. But it hits squares of both colours, jumps over pawns and pieces, and its path cannot be blocked.

All the rules follow from this. A bishop needs open diagonals: there its range does its work. A knight needs firm squares close to the opponent: there it stands and cannot be driven away. A bishop likes pawns on both wings and play across the whole board. A knight likes all the pawns on one wing: it does not have to travel far.

There is also a subtle difference that matters in the endgame. A bishop can make a waiting move: shift along its diagonal and keep holding the same squares. A knight cannot: every move it makes changes all the squares it hits. What is more, a knight cannot lose a tempo at all: it returns to its old square only after an even number of moves. A bishop can: Bc1–e3–d2–c1 is three moves, and the bishop is back where it was.

Open or closed position

The main question is whether the position is open. In an open position there are few pawns in the centre, the diagonals are free, and the bishop hits across the whole board. In a closed position pawn chains block the diagonals, the bishop runs into pawns, and the knight jumps over them and finds squares for itself.

The character of the position is often set by the opening. In the King's Indian Defence and in the French the pawn chains stay for a long time, the centre is closed, and knights feel at home there. In the open lines of the Sicilian Defence, or in openings where the central pawns are traded early, the diagonals are free and bishops soon become stronger than knights. When you choose an opening, you also choose which pieces will be stronger for you.

A position does not stand still. Pawns get traded, and a closed position can open within a few moves. So the side with the bishop tries to open the game: trade pawns, free the diagonals. The side with the knight, on the contrary, keeps the pawns in place and closes the lines. If you have a knight against a bishop, put your pawns where they block the bishop's path, and do not hurry to trade them.

The good and the bad bishop

A good bishop is a bishop whose own central pawns stand on squares of the other colour. The pawns do not get in its way, and together they hold squares of both colours: the bishop its own, the pawns the others. A bad bishop is the opposite: its pawns stand on squares of its colour, block its diagonals, and the bishop turns almost into a pawn that defends other pawns.

A bad bishop is not necessarily a weak piece. It defends its pawns, and if it stands in front of its own pawn chain, it can be quite active too. But in an endgame against a knight a bad bishop often loses: its pawns on squares of its colour become targets, and the bishop itself is tied to their defence.

What can you do with a bad bishop? There are three ways. You can bring it out in front of your pawn chain, so that it stands ahead of the chain and hits squares in the opponent's camp. You can trade it for a good enemy piece, especially for the opponent's good bishop. And in time you can change the pawns that get in its way, for example by undermining the chain with a pawn. The worst is to leave the bad bishop standing behind its pawns and hope the game never reaches an endgame.

The engine knows about this too. On the board below Black has the bishop on c8 against the white knight on f3. Black's pawns on d5 and e6 stand on light squares, and the bishop is light-squared too. In material the bishop is worth more than the knight, but the “Bad bishop” and “Mobility” rows turn the evaluation in White's favour.

Hence the rule we already saw in the chapter on weak squares: put your pawns on squares of the colour opposite to your bishop. And if you have a knight against a bishop, put your pawns on squares of the colour of the opponent's bishop: then it cannot attack them, and your knight will get to any square. Against a bad bishop, firm squares of the other colour become a home for the knight.

When the knight is stronger

A knight is strongest on a firm square, especially on an outpost in the centre or in the opponent's camp. From there it hits eight squares and is not afraid of pawns. A knight is good in closed positions, where pawns block the bishop's diagonals, and in endgames where all the pawns are on one wing: it does not need to travel far.

Another strength of the knight is forks. A knight's move cannot be blocked, and a knight attacking a bishop is especially unpleasant: the knight always attacks from a square of the other colour, so the bishop cannot take it in return. So in endgames against a knight, watch whether your king and bishop stand one knight's jump away from the same square.

Against a knight it helps to remember its weaknesses too. A knight on the edge of the board hits few squares, and in the corner only two. A knight crosses from wing to wing slowly. And a knight can be deprived of squares: if every square it could go to is guarded by pawns, it stands idle, like a bad bishop.

A classic endgame win is a good knight against a bad bishop. The weaker side's pawns stand on squares of its bishop's colour, the knight occupies a firm square of the other colour, and the stronger side's king walks in over squares the bishop cannot defend. The bishop can only move back and forth, and the pawns become prey one after another. Such endgames are won even without an extra pawn.

When the bishop is stronger

A bishop is stronger in open positions and in endgames with pawns on both wings. It helps its own passed pawn from afar and at the same time guards the opponent's pawns on the other wing. A knight cannot keep up in such an endgame: while it runs to one wing, a pawn goes through on the other.

A knight has one more difficulty: the rook's pawn. A knight trying to stop a pawn on the a- or h-file stands at the edge of the board and hits few squares. A bishop, on the contrary, holds the promotion square from anywhere on its diagonal. So an edge passed pawn is an especially unpleasant opponent for a knight.

Two bishops are stronger still. The bishop pair hits squares of both colours and has none of the single bishop's main weakness. In an open position two bishops against a bishop and knight, or against two knights, are a noticeable advantage. Wilhelm Steinitz, the first world champion, already valued it: he was one of the first to explain why two bishops are stronger than a bishop and knight in an open position. The engine also gives the bishop pair a separate term, about half a pawn.

Opposite-coloured bishops

A special case is opposite-coloured bishops: one side has a light-squared bishop, the other a dark-squared one. They never meet: each holds squares the other cannot even attack. The consequences differ in the middlegame and in the endgame.

In the endgame opposite-coloured bishops pull strongly towards a draw. The defending side's bishop stands on a square in front of the opponent's pawn, a square of its own colour, and it cannot be driven away from there: the opponent's bishop moves on other squares. Often even two extra pawns do not win. The engine knows this: in an endgame with opposite-coloured bishops it scales the evaluation down sharply.

White has an extra pawn and is almost a pawn ahead in material. But the “Opposite-coloured bishops” row takes away more than half of that edge, and the total is about a third of a pawn. The black bishop holds the light squares in front of White's pawns, and it is hard for White to break through them.

In the middlegame it is the other way round. Opposite-coloured bishops help the attacker. If your bishop hits the squares around the opponent's king, their bishop cannot defend them: it moves on the other colour. So in a middlegame with opposite-coloured bishops the initiative decides: whoever starts an attack on the squares of their bishop's colour first often wins. The defending side in such a position should trade as many other pieces as possible, to survive into a drawn endgame.

So when the opponent offers an exchange after which opposite-coloured bishops remain, look at the stage of the game. If an endgame lies ahead and you are a pawn down, such an exchange is your salvation. If the queens and rooks are on the board and your bishop is aimed at the opponent's king, such an exchange can be the start of an attack.

Which piece to keep

In every game you have to decide what to trade: the bishop or the knight. Before trading, ask yourself a few questions. Will the position open or stay closed? Which squares do the pawns stand on, yours and the opponent's? Does the knight have firm squares? Will there be pawns on both wings? The answers usually show clearly which piece will be stronger twenty moves later.

A good rule: keep the piece that will be better in the endgame if the position is heading for an endgame, and the one that attacks better if an attack lies ahead. And do not forget the opponent's bad bishop: trading your good bishop for their bad one is usually a favour to the opponent.

What's next

Opposite-coloured bishops showed how much the initiative matters in an attack. Attacking the king is an art of its own: what conditions are needed, how many pieces, when to sacrifice. That is the subject of the next chapter.