Day 3of 14, Master the Fundamentals with Capablanca

The Uncastled King

Janowski vs Capablanca · Havana 1913 · The Uncastled King

Capablanca lost this game to David Janowski for one reason, and it is the most basic reason in chess. He did not castle. On move eight, with everything ready, he played his knight to f8 instead of tucking his king away, and from that moment every problem in the position traced back to a king stuck in the centre. Janowski opened lines, checked, and harvested. Take one idea into your games, while the queens are on the board an exposed king outweighs every other consideration, so castle first and philosophize later.

Three ideas you'll take from this endgame
01

Castle, then think

On move eight Capablanca had a free hand and played 8...Nf8, leaving the king in the centre. Every later difficulty grew from that single omission.

02

Queens change the maths

With queens on, king exposure outweighs structure, material and rook activity. Capablanca's positional assessments would have been right in a queenless ending, and were wrong here.

03

One open line is enough

Janowski needed a single line to reach the king. Checks, captures and threats then combine on their own, and the h4 break tore the last shelter away.

Study Mode: Watch & Play Along

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 Lesson
♞ Play Through the Game: The Board Stays With You
The Big Idea

He never castled, and never recovered

This is the first of Capablanca's published losses that turns on a beginner's rule rather than a subtlety, and that is exactly why it teaches. From an irregular queen-pawn setup Black is solid but slightly behind in development, and the correct reply to that is simple, finish developing and get the king to safety. Instead 8...Nf8 kept the king in the centre, and after 13.Bxe7 Capablanca had to recapture with the king itself. From there Janowski had a permanent target. He traded into positions where the black king could not hide, opened the h-file with 24.h4, and every check gained time for the next one. In your games, treat castling as a move you make early and without negotiation while the queens are on. A king in the centre does not simply feel risky, it changes the value of every other feature on the board.

How the Endgame Arose

An irregular setup, and time lost

Capablanca answered 1.d4 with a flexible King's-Indian-flavoured system, solid enough but slow, and the slowness cost him. By move eight White had castled and had both rooks ready to come to the centre, while Black still had a knight on d7, a bishop on c8 and a king on e1. Capablanca himself was uncertain about the whole system in his notes. The immediate question is not whether the opening is sound, it is what to do with the tempo he has, and the answer the position demands is short: castle.

Stop and Assess

White has castled. What should Black do now?

After 8.O-O the material is level and nothing is yet wrong with Black's position. But White is fully developed and castled, and Black is not. This is the moment the game is decided, long before any tactic appears. Stop and assess before you read on.

Material is level and White has just castled. What is the one move Black must make, and what happens if he delays?

Count the developed pieces and look at the two kings. Which side is ready for the game to open? Think, then reveal.

Janowski vs Capablanca, after 8.O-O, White is castled and developed while Black's king is still in the centre.
After 8.O-O, material is level, but White is castled and Black is not
Reveal the assessment

White's method (Janowski)

  • Open the position. The better-developed side opens lines, and 11.dxe5 and 12.Ne4 start the process.
  • Fix the king in the centre. 13.Bxe7 forces 13...Kxe7, and castling is gone for good.
  • Break the shelter. 24.h4 rips open the last cover and every check gains time.

Black's problem (Capablanca)

  • 8...Nf8 instead of castling. The one move that decides the game, played without a threat to answer.
  • A king that cannot hide. After 13...Kxe7 the king is a permanent target on an open board.
  • Assessments from the wrong game. Capablanca judged the position as if the queens were already off.
The Plan

Open the lines, fix the king, then collect

Janowski wins in three simple stages, and none of them requires brilliance. First he opens the game, 11.dxe5 dxe5 and 12.Ne4, because the better-developed side wants lines. Second he fixes the king, 13.Bxe7 leaves Black no choice but 13...Kxe7, and from that moment there is no castling and no shelter. Third he attacks the shelter that remains, 24.h4 opens the h-file, 25.Qxh4+ starts a sequence of checks that gain time, and 28.Qxg7 collects. The pattern is worth memorising because it is the standard punishment for an uncastled king, open a line, check, and let the checks buy the next move. In your games, when your opponent leaves the king in the centre, do not look for a knockout, look for a line to open.

The Critical Moment: Think First

Janowski has just played 13.Bxe7. It looks like an ordinary exchange of bishops, and it is the decisive moment of the game. Look at Black's possible recaptures before you read on.

Black to play. Janowski has just captured on e7. What is forced here, and why does it decide the game?

Janowski vs Capablanca, after 13.Bxe7, Black must recapture with the king.
After 13.Bxe7, only 13...Kxe7 is possible, and Black's king is fixed in the centre for good
Reveal the move and the why

13...Kxe7 is forced, and with it the game. The queen on a5 cannot reach e7 and no knight covers the square, so the king itself must take, and castling disappears permanently. This is the hidden cost of 8...Nf8, a move that looked merely slow. From here Janowski does not need a combination, only lines: he trades toward positions where the king has no shelter, plays 24.h4 to open the h-file, and lets the checks do the work. In your games, before you make a quiet developing move, ask which recaptures you are leaving yourself. A forced recapture with the king is usually the moment a game is lost.

♞ Follow the Finish: The Board Returns
Castle, Then Think

Open the h-file, and let the checks work

With the black king marooned on f6, Janowski cracks the last shelter. 24.h4 opens the h-file by force, 24...gxh4 25.Qxh4+ begins a run of checks that gain time with every move, and by 28.Qxg7 White has collected material and kept the initiative. None of it is deep. It is simply what an exposed king costs.

Janowski vs Capablanca, after 24.h4, the h-file is about to open against the black king.
24.h4: the break that opens the last line to the uncastled king

The rest is conversion. Janowski keeps the checks coming, wins the g-pawn, and pushes his central majority while Black's king still has nowhere to stand. By move 36 the position is resignable. Carry the lesson into your games in its simplest form, castle early, and when your opponent does not, open a file toward the king and start checking.

Janowski vs Capablanca, the final position, White is winning with an exposed black king.
The finish: material and the exposed king decide, and Capablanca resigns
Principles to Take to Your Own Board

This game is over. Your games are where it matters.

1 · Castle early, without negotiating.

The rule: while the queens are on, get the king to safety before you do anything optional. In this game: 8...Nf8 instead of castling lost the game on move eight. In your games: if there is no concrete reason to delay, castle.

2 · With queens on, king safety outranks everything.

The rule: an exposed king outweighs structure, material and activity while queens remain. In this game: Capablanca's evaluations were endgame evaluations applied to a middlegame. In your games: judge king safety first, then everything else.

3 · The better-developed side opens the position.

The rule: if you are ahead in development, open lines; if behind, keep it closed. In this game: Janowski opened with 11.dxe5 and 12.Ne4 precisely because he was ready. In your games: match the pace of the position to your development.

4 · Watch which recaptures you leave.

The rule: a quiet move that forces you to recapture with the king is not quiet. In this game: 13.Bxe7 left only 13...Kxe7. In your games: check your recaptures before you commit to a slow move.

5 · One open line is enough.

The rule: against a king without shelter you need a single line, then checks buy the rest. In this game: 24.h4 opened the h-file and the checks did the work. In your games: open the file nearest the enemy king and start with a check.

Test Yourself

Five questions before you go

1 · Where exactly did Capablanca go wrong?

On move eight. With White castled and fully developed, 8...Nf8 left the king in the centre when nothing forced it. Everything afterwards, the forced 13...Kxe7, the open h-file, the run of checks, follows from that single omission.

2 · Why is castling more urgent than developing a knight?

Because while the queens are on, king safety changes the value of every other feature. A developed knight is worth little if the position opens against a king in the centre. Castling is the move that makes all your other pieces safe to use.

3 · Why was 13...Kxe7 forced?

Nothing else covers e7. The queen sits on a5 and no knight reaches the square, so the king must recapture. That is the practical cost of 8...Nf8, and it is why you check your recaptures before playing a slow move.

4 · How did Janowski actually break through?

By opening a line. 24.h4 forced the h-file open, 25.Qxh4+ started a series of checks that gained time, and 28.Qxg7 collected the material. Against an exposed king you do not need a combination, only a line and a check.

5 · Were Capablanca's own notes reliable here?

Not entirely. He described several positions as safe or equal, and those judgements would have been sound in a queenless ending. With queens on the board and his king in the centre they were wrong, which is the honest lesson he preserved by publishing the loss.

Sharpen Your Eye: Six Endgame Exercises

Now you convert the edge

Six knight endings, rising in difficulty. In each you have the winning side, so bring the knight and king together and convert. Drag the pieces, the board plays the defence for you.

Castle Early. Then Play Chess.

Knowing the rule is not the same as obeying it under pressure, and most club games are still decided by a king left in the centre a few moves too long. In a 2-hour Strategy Assessment I go through your own games, show you the exact moments your king safety slips, 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
Next, Game 4: Capablanca vs Znosko-Borovsky, and the Rule of Three in calculation.