When the kings castle on opposite sides, the game becomes a race and the rules change. Manoeuvring is a luxury; every move that does not advance a pawn toward the enemy king is a move given away. Capablanca had already removed Black's compensation for his doubled pawns by trading a bishop, and when Janowski answered the opposite-side castling with a slow knight move instead of a pawn storm, the race was effectively over. Take one idea into your games, with opposite castling, push the pawns in front of their king and count tempi, not beauty.
In a castling race the pawns open the lines and the pieces follow. Janowski's 10...Nh6 was a developing move in a position that had no time for one.
Black's doubled c-pawns were paid for with the bishop pair. 7.Be3 invites the trade, and once a bishop goes the weakness remains without the payment.
Even Capablanca loses time here with 9.Qe1. In a race a slow move is not a small error, it is a move handed to the opponent.
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.
The Exchange Ruy Lopez gives Black doubled queen-side pawns in return for the bishop pair, and Capablanca's whole opening scheme is designed to collect the debt without paying the interest. With 7.Be3 he offers the trade of dark-squared bishops, and after 7...Bxe3 8.fxe3 he has three things at once: one of Black's bishops is gone, the f-file is open toward the king-side, and the doubled e-pawns now control d4. When Janowski castles long, the position turns into the purest kind of race, White's pawns go at the black king on the queen-side and Black's should go at the white king on the king-side. What decides it is not evaluation but speed. Capablanca is not perfect here, 9.Qe1 is slow, but Janowski answers with 10...Nh6, a piece move in a position that demanded pawns. By the time the black attack is ready to begin, 15.b5 has already opened the b-file. In your games, when the kings go to different wings, start pushing immediately.
Capablanca met the Spanish with 4.Bxc6 dxc6 and then chose 5.Nc3, a line he had discussed with Alekhine, who used it in the same tournament against Emanuel Lasker with a fine position. The strategic thread is simple. Black holds the bishop pair as compensation for a damaged structure, so White trades a bishop off at the first opportunity. After 7.Be3 Bxe3 8.fxe3 the compensation is halved, the f-file is open, and the doubled e-pawns are doing useful work covering d4. Janowski then castled queen-side, boldly and characteristically, and the race began.
Black has just castled long while White is castled short. Material is level and both sides have exactly the same instruction: attack the king in front of you with pawns. The only question that matters now is who does it faster. Stop and assess before you read on.
The kings have castled on opposite wings. What should each side do, and what is the one thing neither can afford?
Ask which pawns can advance toward the enemy king without exposing your own. Then ask what a quiet piece move costs in a race. Think, then reveal.
Capablanca runs the race in three stages. First he storms, 11.b4 and 12.a4 push directly at the king with no preparation beyond what is needed, because in this structure the pawns are the attack. Second he breaks, 15.b5 opens the b-file at the moment Black is least ready, and after 15...cxb5 16.axb5 a5 the black king's cover is permanently compromised. Third he invades, the knight comes to d5, the rooks arrive on the open files, and 27.c6+ finishes the job by dragging the king into the open. Janowski meanwhile has spent his moves on Nh6, Nf7, f6 and Ng5, all reasonable-looking, none of them a pawn moving toward the white king. In your games, judge a castling race by a single number, how many moves each side has spent advancing pawns at the enemy king.
White has advanced on the queen-side and Black has spent his moves repositioning knights. The pawns are as far forward as they can go without a break. This is the moment the race is won or lost.
White to play. The pawns are advanced and the black king is behind them. Which move opens the position?
15.b5, the break the whole attack was built for. After 15...cxb5 16.axb5 a5 the b-file is open, the black king has lost the pawn cover it was relying on, and every white piece has a route in. This is the moment to note what made it possible: Black's doubled c-pawns, created back on move four, are exactly what let the b-pawn break open the position so cheaply. The structural decision and the attacking decision are the same decision, made ten moves apart. In your games, when you storm a castled king, look for the pawn break that opens a file rather than the one that simply gains space, and time it for the moment their counterattack is still a move or two away.
With the queen-side open the invasion is quick. The knight reaches d5, the heavy pieces occupy the c- and d-files, and 27.c6+ is the final blow, a pawn check that pulls the king onto squares where the rooks and queen are already waiting. Black's king-side counterattack never produced a single threat.
The finish is short. 28.cxd7 Qxd7 29.d5 and the passed pawns and the exposed king decide together, with 31.Qc6 ending matters. The whole game is one lesson repeated: in an opposite-castling race the pawns are the attack, and the side that starts first usually finishes first. Carry it into your games, when the kings split, put a pawn in motion on the very next move.
The rule: the pawns in front of the enemy king are your attack, so start them at once. In this game: White pushed b4 and a4 while Black played knight moves. In your games: when the kings split, advance a pawn on the next move.
The rule: in a race, your own cover must stay intact. In this game: White's king-side pawns hardly moved all game. In your games: attack with the pawns far from your king.
The rule: when the opponent has a weakness paid for by an asset, trade the asset off. In this game: 7.Be3 removed a bishop and left the doubled pawns unpaid for. In your games: exchange the piece that justifies their structure.
The rule: in an attack, the pawn break that opens a line beats the one that merely advances. In this game: 15.b5 opened the b-file straight to the king. In your games: aim the break where your rooks can follow.
The rule: in sharp opposite-castling positions, tempo counts more than refinement. In this game: 10...Nh6 lost the race, and even 9.Qe1 was slow. In your games: ask whether each move brings the attack closer, and if not, find one that does.
Because a castled king is protected by pawns, and only pawns can remove them cheaply. Pieces alone cannot open a file. In an opposite-castling race the pawns do the opening and the pieces do the invading, in that order.
Nothing, except the position it was played in. It develops a piece and improves nothing about Black's attack. In a race that is a tempo handed over, and here it was the tempo that let 15.b5 arrive before any black threat existed.
Because Black's bishop pair was the compensation for his doubled c-pawns. Trading a bishop keeps the structural weakness and removes the payment for it. It also opened the f-file and let the doubled e-pawns cover d4, so one move achieved three things.
They make the break cheap. With pawns on c7 and c6, White's b4-b5 hits a structure that cannot hold, and after the exchanges the b-file opens straight onto the king. The opening weakness and the middlegame attack are the same idea at different stages.
No, and it is worth being honest about it. 9.Qe1 is slow in a position that punishes slowness, and Black could have taken the initiative with an immediate pawn storm. Capablanca won because his opponent was slower still, which is exactly how most races are decided.
Six queen-side attacking exercises, rising in difficulty, the operation that won this game. In each you have the winning side, so open the lines and get at the king. Drag the pieces, the board plays the defence for you.
Opposite-side castling positions are decided by a handful of tempi, and most club players lose them by making sensible moves at the wrong moment. In a 2-hour Strategy Assessment I go through your own games, count the tempi you gave away in sharp positions, and hand you a plan built for your chess, not a hypothetical student's.
Book Your $130 Strategy Assessment 2 hours via Zoom · Worldwide · Limited spots