In 1901 a twelve-year-old Capablanca faced Juan Corzo, the champion of Cuba, who kept choosing sharp gambits to punish the boy''s lack of opening theory. In this game Corzo grabbed a knight on f7 to drag the black king out, but the move that mattered was his own: forced to play Kf2 and then Kg1, Corzo gave up the right to castle, and his king sat stranded for the rest of the game. Capablanca did what the position demanded. He stopped counting material, pointed every piece at the stuck king, and would not let it escape. He even let his opponent grab a rook, because a rook on h8 does not defend a king on g1. Take one idea from this above all others: when the enemy king cannot castle, the king itself becomes the target, and a lead you can measure in pawns is worth less than an attack you cannot measure at all.
Once Corzo was forced to play Kf2 and Kg1, his king was stuck in the open for good. Capablanca aimed every piece at it and never looked back.
Capablanca gave back a piece and let Corzo take a rook. A unit that is not helping defend the king is not worth defending yourself.
Every move was a check or a direct threat, so Corzo never got a free move to untangle. The king was chased from g1 to e1 and resigned in a mating net.
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.
Corzo played the Vienna Gambit and then the sharp knight sacrifice 7.Nxf7, dragging the black king to f7 to profit from the boy''s missing opening theory. But Capablanca answered 8...d5 and 9...Qe7+, and after 10.Kf2 g3+ 11.Kg1 it was the white king, not the black one, that was stranded. From there the lesson is pure attack. Capablanca traded the queens with 12...Qc5 and 13...Qb6, which looks strange in an attacking game, yet his pieces kept the initiative without them. He lifted a rook with 16...Ra4, invaded on the second rank with 20...Rd2, and when Corzo won a whole rook with 22.Bxh8 Capablanca did not flinch, because that rook was on the far side of the board doing nothing. He broke the position open with 22...f3, planted the knight with 23...Nf4, and finished with 24...Rg2+ 25.Kf1 Rf2+ 26.Ke1 Nd3+, a mating net the stranded king could not survive. A whole game won on one idea: the uncastled king is the target, and material is only a tool to get at it.
Capablanca learned chess at four by watching his father, had almost no book knowledge, and loved studying endgames. Corzo, far more experienced, chose the Vienna Gambit (1.e4 e5 2.Nc3 Nc6 3.f4) and then the Hamppe-Allgaier line with 6.Ng5 and 7.Nxf7, a sharp sacrifice designed to make theory decide the game. Capablanca invented his replies over the board. The key point came not from a book but from his own analysis: an earlier game in the match had reached a similar position and been drawn, and when Capablanca heard that Corzo thought Kf2 saved it, he studied the line himself, disagreed, and steered straight back into it to prove his own judgement. The lesson begins once the smoke clears on move 14 and the white king is stuck for good.
After 14...axb6 the queens are gone and the material is level. It looks like a quiet position, the kind where an attack should have fizzled out. The one permanent feature is the white king on g1, boxed in by its own rook and bishop and unable to castle to safety, ever. The question is whether Capablanca can keep hunting it with rooks and minor pieces alone. Stop and assess before you read on.
The queens are off and the material is nearly level. Where is Black''s advantage, and how does an attack survive into a near-endgame?
Look at the white king on g1 and ask whether it can ever reach safety. Picture a black rook on the second rank, a knight planted near the king, and the light squares around it that White''s dark bishop can never guard. Study the board, think, then reveal.
Capablanca converts the attack without queens, which is the hard part to believe. He opens a path for the rook with 16...Ra4, trades a pair of minor pieces to reach the white king with 17...Bxd4+ and 18...Rxd4, and swings the rook to the second rank with 20...Rd2. Corzo tries to distract him by winning material: 21.Bh5+ picks up a knight and 22.Bxh8 grabs a whole rook. Capablanca ignores both, because the captured units were never part of the defence. He plays 22...f3 to rip open the g-file and f-file, lands the knight on f4 with 23...Nf4, and then calculates the finish: 24...Rg2+ 25.Kf1 Rf2+ 26.Ke1 Nd3+, driving the king into a net it cannot escape, and Corzo resigns. Material was against Capablanca at the end, and it did not matter in the least.
This is the moment the attack pays off. Corzo has just played 24.Be5, having won a rook on the last move, and on material he is ahead. But his king is on g1 with a black rook on d2 and a black knight on f4 breathing down its neck. It is Black to move, and there is a forcing line that ends the game. The whole point of the attack is hidden in the next three moves.
Black to play. How does Capablanca finish the hunt, and why is being down material irrelevant?
24...Rg2+, and the king is driven to its death. The rook cannot be captured, because the knight on f4 guards g2, so Kf1 is the only legal move: 25.Kf1 Rf2+ 26.Ke1 Nd3+. Now the knight hits the king and guards the escape squares at the same time, the rook on f2 seals the second rank, and wherever the king goes the net closes, so Corzo resigned. Notice what Capablanca spent to get here. He is behind on material, yet the white king has no defenders that matter, because a bishop on e5 and a rook on h1 cannot race back to shield a king being chased to e1. In your games, when the enemy king is stranded and cannot castle, stop counting material and start counting threats. The attack is only over when the king is safe, and a king that cannot castle is rarely safe.
Look at the picture after 26.Ke1. Capablanca has chased the white king from g1 to f1 to e1 with a string of checks, each one forcing, none of them refusable. The rook sits on f2, the knight is about to land on d3, and the black bishop on c8 eyes the light squares the white army can never cover. On material Capablanca is actually behind, but not one white piece is doing anything to defend the king. This is what an attack on a stranded king looks like at its end: total on the squares that matter, and it does not matter anywhere else.
26...Nd3+ is the final nail. The knight checks the king, covers the squares it would use to run, and leaves Corzo with no answer to the threats that follow, so he resigned. A twelve-year-old had just mated the champion of Cuba with rooks and minor pieces, no queens on the board, and behind on material. Carry the real lesson into your games: when your opponent''s king cannot castle, make it your only target, keep every move forcing, and do not let a count of the pieces talk you out of the attack. The king is worth more than any rook.
The rule: a king that has lost the right to castle, or sits in the centre, is a permanent weakness you can attack for the whole game. In this game: after Kf2 and Kg1 Corzo''s king never found safety. In your games: when you take away the enemy''s castling, switch your plans to a direct attack on the king.
The rule: in a live attack the value of a piece is how much it helps mate or defend, not its point count. In this game: Capablanca gave back a piece and a rook and still won. In your games: judge a sacrifice by whether the attack keeps rolling, not by the material it costs.
The rule: checks and direct threats stop your opponent from untangling or counter-attacking. In this game: the king was chased from g1 to e1 with checks and never got a free move. In your games: look for the forcing move first, so the defender is always reacting.
The rule: a bishop or rook on the wrong part of the board is a spectator; aim at the squares your opponent cannot cover. In this game: White''s dark bishop could never guard the light squares around g1. In your games: steer the attack onto the colour or file the enemy pieces cannot reach.
The rule: you do not need the queen on the board to mate a stranded king; rooks and minor pieces are enough if the king is exposed. In this game: the queens came off on move 14 and the attack rolled on. In your games: do not assume a queen trade ends the attack when the enemy king is still unsafe.
Because it cannot reach the safety of a castled position behind a wall of pawns, so it stays exposed in the open where every enemy piece can take aim. Once Corzo was forced to play Kf2 and Kg1 his king was stuck in the corner behind its own men, unable to castle for the rest of the game, and that single permanent weakness was enough for Capablanca to build a winning attack around.
Because the pieces Corzo captured were not defending his king. A rook sitting on h8 and a bishop parked on e5 do nothing to stop a mating net on g1, so their point value was irrelevant to the only question that mattered: can the white king survive? Capablanca spent material the way you spend money on something you need, and he bought a mating attack with it.
The queen is the strongest attacker, but it is not the only one. When the enemy king is stranded and short of defenders, rooks and minor pieces working together can weave a mating net on their own. Here Capablanca traded queens on move 14 and still mated with a rook, a knight and a bishop, because the white king had no safe squares and no help.
It is a forcing check the king cannot answer by capturing, because the knight on f4 already guards g2, so 25.Kf1 is the only legal move. After 25.Kf1 Rf2+ 26.Ke1 Nd3+ the king is driven into a net where the knight both checks and covers its escape. The move works because every reply is forced, which is the whole art of finishing an attack on an exposed king.
Counting material in the middle of an attack. Capablanca shows that when the enemy king cannot castle, the right currency is threats, not pawns. Train yourself to ask first whether the enemy king is safe, and if it is not, to pursue it with forcing moves and to spend material freely, because the king is worth more than any piece you give up to reach it.
Six kingside attacks, rising in difficulty. In each one you have the attacking side; find the blow that breaks through and play it out. Drag the pieces; the board plays the defence for you.
Knowing when a king is too exposed to survive, when to spend material for an attack, and how to keep every move forcing are the skills that win the games other players agree to draw. In a 2-hour Strategy Assessment I go through your games personally and hand you a roadmap built for your chess, not a hypothetical student''s.
Book Your $130 Strategy Assessment 2 hours via Zoom · Worldwide · Limited spots