A bishop hemmed in by its own pawns is one of the saddest pieces in chess, and knowing how to create one, keep it bad, and then attack where it cannot help is a complete plan in itself. This short exhibition game is the idea in its purest form. Black's light-squared bishop was boxed in behind pawns fixed on light squares, and Capablanca made sure it stayed that way, stopping every attempt to trade it off. The bishop drifted to the corner square a8 and simply watched the rest of the game. With his opponent effectively a piece short wherever it mattered, Capablanca threw his forces at the black king and mated, giving back material along the way because the stranded bishop could never come to the rescue. Take one idea from this game: when your opponent has a bad bishop, keep it bad, and attack on the side of the board where it can never arrive.
A bishop is bad when its own pawns are fixed on its color, blocking its diagonals. Black's pawns on d5 and e6 buried his light-squared bishop.
Do not let the opponent trade their bad piece for a good one. Capablanca played a4 to stop the bishop escaping to b5, locking it in for good.
A dead piece on one side means you are effectively a piece up on the other. Capablanca attacked the king while the bishop sat useless in the corner.
Watch the lesson, then play the game through on the board below. The board stays with you as you scroll. Make each move yourself while you read. You'll remember a plan you played with your own hands far longer than one you only watched.
Black's light-squared bishop was headed for trouble from early on: his central pawns on d5 and e6 sat on light squares, the same color the bishop travels on, giving it poor long-term prospects. Sensing it, Black tried to swap it off, rerouting his queen to prepare Bb5, but Capablanca saw the plan and played 12.a4 to take the b5 square away, keeping the bad bishop locked on the board. From there the bishop had nowhere useful to go; it ended up on a8, in the far corner, where it watched the rest of the game without making a single useful move. That left Black effectively a piece short everywhere else, and Capablanca used the extra force where it counted. He activated his own bishop with 19.Bb2, brought his knights to the kingside, and crashed through with 22.Bxf6 and a sacrifice on f7, mating the black king. He even gave material back during the attack, because it did not matter: the piece that might have defended was stranded in the corner. One bad bishop, kept bad, decided the whole game.
Capablanca set up a solid Stone Wall formation with pawns on d4, e3, f4 and c3, a closed structure that keeps the position strategic rather than sharp. Black answered passively, locking his own central pawns on light squares and leaving his light-squared bishop with no good diagonals. When Black tried to free it by rerouting toward b5, the real contest of the game appeared: could Black trade his bad bishop for a good one, or would Capablanca keep it buried? The lesson begins at move 12, with Capablanca deciding exactly that.
After 11...Qe8 Black has a clear idea: the queen is heading for b5 by way of the bishop, preparing Bb5 to trade off the bad light-squared bishop for White's good one. That trade would solve Black's main problem. Ask the key question: how does White keep the bad bishop bad? Look for the quiet pawn move that takes away the escape square, before you read on.
Black is preparing Bb5 to trade off his bad light-squared bishop. How does Capablanca keep that bishop buried?
A bad bishop is only a problem if it stays on the board. Find the quiet move that denies the b5 square and keeps Black's worst piece locked in, and picture where that bishop will end up with nowhere to go. Think, then reveal.
Capablanca's plan has two halves, and the first makes the second possible. First, he keeps the bad bishop bad: 12.a4 controls b5 and stops a clean Bb5 trade, and when Black tries queenside play with b4, Capablanca simply recaptures and answers 19.Bb2, turning his own bishop into a monster on the long diagonal while Black's bishop retreats all the way to a8. Now the second half begins. With Black's bishop a permanent spectator, White attacks the king with a material edge that does not show on the scoreboard but is very real where it matters. The knights come to g5 and g4, and 22.Bxf6 smashes the pawns in front of the king; after 22...gxf6 the sacrifices rain down, 23.Nh6+ and 24.Nhxf7, dragging the king into the open. The queen joins with 25.Qh5 and 26.Qh6+, and 27.Nxg5 finishes the net. Throughout, Black's extra material on the board is meaningless, because the piece that could defend is stuck in the corner. That is the whole method: create a bad piece, keep it bad, and attack where it cannot reach.
It is White to move, and the groundwork is done. Black's light-squared bishop is buried on a8, taking no part in the game, so around the black king White has, in effect, an extra piece. Capablanca's own bishop rakes the long diagonal, both knights sit ready on g5 and g4, and the queen can swing across. This is the moment to turn a positional advantage into a direct attack. Find the breakthrough that tears open the black king.
White to play. With Black's bishop stranded on a8, how does Capablanca break open the king?
22.Bxf6, smashing the king's shelter. Capablanca gives up his strong bishop to wreck the pawns around the black king, because with Black's light-squared bishop dead on a8 he has the attackers to follow up and the defender cannot. After 22...gxf6 the black king is stripped bare, and the knights pour in: 23.Nh6+ drives the king out, 24.Nhxf7 keeps up the storm, and the queen arrives with 25.Qh5 and 26.Qh6+ before 27.Nxg5 seals the mating net. Notice the engine of the whole attack is not a single tactic but the imbalance underneath it: Black is defending with one fewer piece than White is attacking with, because that bishop on a8 might as well not exist. In your own games, when your opponent has a piece that is permanently out of play, treat the rest of the board as if you are a piece up, and attack where the numbers favor you. A sacrifice that opens the king is sound when you simply have more force at the target than the defender can bring.
Look at the position after 24.Nhxf7. Capablanca has sacrificed his bishop and is pouring knights into the black king, and the remarkable thing is that he is not even winning material in the usual sense, at this moment the count is close. It does not matter at all. Count the pieces attacking the king and the pieces defending it: White has his queen and both knights swarming, while Black's rook, queen and bishop are scattered and, crucially, the bishop on a8 is a complete bystander. The attack has the numbers, and the defense is short exactly one piece, the bad one.
The finish is clean: 24...Qe8 25.Qh5 fxg5 26.Qh6+ Kg8 27.Nxg5, and Black resigned, the attack decisive. Here is the detail worth savoring: at the final position Black is actually ahead on material, and he is lost anyway, his king overrun, because his light-squared bishop on a8 never took a single part in the defense. That is the power of a bad bishop fully exploited. The win was not a combination that appeared from nowhere; it was the harvest of a plan begun on move 12, to keep one enemy piece useless and then attack where it could never help. Take that method to your own games, and bad bishops will start winning you points.
The rule: a bishop is bad when its own pawns are fixed on the same color, blocking its diagonals and leaving it with no scope. In this game: Black's pawns on d5 and e6 buried his light-squared bishop. In your games: check which bishop your pawn structure is helping and which it is hurting.
The rule: stop your opponent from trading their bad bishop for your good one; the imbalance is only worth something while it is on the board. In this game: 12.a4 controlled b5 and denied the trade. In your games: deny the squares where the bad piece could be exchanged.
The rule: if the opponent has a piece permanently out of play, you are effectively a piece up everywhere else. In this game: the a8 bishop meant Black defended the king a piece short. In your games: play on the side of the board where the useless piece cannot reach.
The rule: a sacrifice that opens the king is sound when you have more attackers at the target than the defender can bring. In this game: 22.Bxf6 worked because the a8 bishop could not defend. In your games: count attackers and defenders before you strike.
The rule: what matters is the balance of force at the point of action, not the total count on the board. In this game: White mated while behind on material. In your games: judge each part of the board on its own, especially around the king.
A bishop is called bad when its own pawns are fixed on the same color of square that the bishop travels on, so those pawns block its diagonals and leave it with little or nothing to do. In this game Black's central pawns sat on d5 and e6, both light squares, and his light-squared bishop was stuck behind them with no open lines. A bad bishop is not always a disaster, but it is a long-term weakness, because it defends poorly and attacks nothing, and if it cannot be traded off it can turn into a permanent spectator, exactly as Black's bishop did on a8.
To keep Black's bad bishop bad. Black had just played Qe8 with the idea of bringing the bishop to b5, where it could be traded for White's good light-squared bishop and Black's main weakness would disappear. By playing a4, Capablanca took away the b5 square and shut the door on that trade, so the bad bishop had to stay on the board with nowhere useful to go. A bad piece is only an advantage for you while your opponent is stuck with it, so preventing its exchange is often the single most important move.
Because the sacrifice opens the enemy king and White has more than enough force to follow up, while Black does not have enough to defend. The key is the imbalance underneath: Black's light-squared bishop was dead on a8, so around the king White was effectively attacking with an extra piece. When you have more attackers at the target than the defender can bring, giving up material to tear open the king is not a gamble but simple arithmetic. The strong bishop had done its job; what mattered now was opening lines for the knights and queen.
Because material on the board means little if a piece cannot take part. At the final position Black was actually ahead on the count, but his light-squared bishop on a8 had never made a useful move and could not help defend the king. So where the game was decided, around the king, Black was effectively a piece down, and White's queen and two knights overwhelmed a defense that was short one piece, with no good way to meet the threats. The lesson is that force is local: what counts is how many pieces each side has at the point of action, not the total on the board.
Looking at your pawns and bishops together, and asking which pieces your structure is helping and which it is burying. Many players fix their pawns on the wrong color and turn a bishop into a long-term liability without noticing, or they let an opponent off the hook by allowing a bad bishop to be traded. Train yourself to spot bad bishops on both sides, to keep the opponent's bad and improve your own, and to steer the game toward the part of the board where their worst piece cannot help. Do that, and whole games will fall into place around a single imbalance.
Six kingside attacks, the reward for attacking where the defender is short a piece, rising in difficulty. In each one the attacking side has the force in place, just as Capablanca did with the black bishop stranded on a8; find the breakthrough and play it out. Drag the pieces; the board answers for the defence.
Spotting bad bishops, keeping them bad, and attacking where the opponent's worst piece cannot help is the kind of positional understanding that turns a quiet opening into a winning attack. In a 2-hour Strategy Assessment I go through your games personally and build you a roadmap for your chess, not a textbook student's.
Book Your $130 Strategy Assessment 2 hours via Zoom · Worldwide · Limited spots