Doubled pawns · For club players · 18 min read
Doubled Pawns: How Fischer Froze Spassky's c-Pawns, Reykjavik 1972
In game 5 of the 1972 match Fischer gave Spassky doubled pawns and the bishop pair, then shut the position. Stockfish 19 shows where the doubled pawns really turned bad, and it was not move 7.
- outpost
- target
- Fixed queenside pawns
- pawn chain or line of force
In short
- Doubled pawns are two pawns of the same colour on one file; they cannot protect each other and are slow to advance.
- In game 5 of the 1972 World Championship, Fischer gave Spassky doubled c-pawns on move 6, then closed the centre so White's bishops had nothing to do.
- Stockfish 19 says the doubled pawns cost White almost nothing at first: the damage came from 13.fxe5, which froze the structure, and from the lonely a-pawn the doubling left behind.
- Fischer's edge never grew beyond about a pawn until 27.Qc2, a one-move blunder that 27...Bxa4 punished at once.
How we know what we say
- Verified From the record: Stockfish's numbers, the players' own notes or interviews, or a cited source.
- Inferred Read from the moves and the position: a reasoned reconstruction, not the player's words.
- Speculative A guess about mood, intent or practical motives: plausible, not proven.
Part 1The setting: Reykjavik 1972, Fischer a point behind
The 1972 World Championship in Reykjavik matched the champion, Boris Spassky of the Soviet Union, against the challenger, Bobby Fischer of the United States. Fischer lost game 1 and forfeited game 2 by not turning up, won game 3 and drew game 4. Game 5, played on 20 July 1972, began with Spassky leading 2½–1½. [2][1] Verified
| Game | White | Result | Match score |
|---|---|---|---|
| 1 | Spassky | 1–0 | 1–0 |
| 2 | Fischer | forfeit, 1–0 to Spassky | 2–0 |
| 3 | Spassky | 0–1 | 2–1 |
| 4 | Fischer | ½–½ | 2½–1½ |
| 5 | Spassky | 0–1 | 2½–2½ |
On the July 1972 rating list Fischer stood at 2785 and Spassky at 2660. The game is short, 27 moves, and ends with one tactical blow, but the reason the blow existed was set up twenty moves earlier, when Fischer gave up a bishop and handed White doubled pawns. [1][2] Verified
Part 2Are doubled pawns bad? The idea at a glance
Doubled pawns are two pawns of the same colour on the same file, created when a pawn captures sideways. They are a weakness when the position is closed and slow, because they cannot guard each other and the front one blocks the back one. They are a fair price when the capture gave something back: the bishop pair, an open file, or extra pawns in the centre, and when the position stays open enough to use it.
- plan
- key square
- Room for the centre
- outpost
- target
- Fixed targets
- pawn chain or line of force
One structure, two verdicts. On the left the doubled pawns belong to a mobile centre that can open lines for the bishops; on the right every pawn is fixed and the bishops have nothing to attack.
The side with the doubled pawns wants open lines and pawn breaks before the weaknesses can be attacked. The side against them wants to fix the pawns, close the position so the bishops have nothing to bite on, and bring knights and the other bishop to the targets. One detail the word "doubled" hides: when the b-pawn captures on c3, the a-pawn loses its only neighbour and becomes isolated too.
Part 3The opening: a bishop given up, a centre locked
Fischer chose the Nimzo-Indian Defence and, after 4.Nf3 c5 5.e3 Nc6 6.Bd3, took on c3 at once with 6...Bxc3+, the plan of the German grandmaster Robert Hübner. Robert Byrne, annotating the game in 1972, explained the point: White's knight on f3 is worse placed than it would be on e2, and the plan offers doubled pawns without waiting for a2-a3. [1] Verified
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Nf3 c5 5.e3 Nc6 6.Bd3 Bxc3+ 7.bxc3 d6 8.e4 e5 9.d5 Ne7 10.Nh4 h6 11.f4 Ng6 12.Nxg6 fxg6 13.fxe5 dxe5 −0.27about equal
The opening as played, ending on Stockfish's verdict after 13...dxe5: a small edge for Black from a structure that looked good for White. Verified
With 7...d6 and 8...e5 Black put the pawns on dark squares, the colour of the bishop just given away, and blocked the e4 pawn in front of White's bishop on d3. After 9.d5 the centre was closed. Stockfish's first choice for Black at move 8 is the game move 8...e5, and after 7.bxc3 it still gives White a small edge: by the engine's count the doubled pawns themselves cost nothing yet. [3] Verified
- outpost
- target
- key square
- Pawns on light squares
A first glance might give the impression that White stands well here, with the two Bishops and a protected passed pawn.
Byrne added that the bishop on d3 had a purely passive job, and that Black, manoeuvring against the fixed pawns, had the upper hand. Sagar Shah, annotating for ChessBase, listed White's assets and asked how Fischer won in spite of them. [1] Verified
Part 4The plans: open lines against fixed targets
By move 17 the shape of the fight is clear. White's assets need open lines, and only the f-file and the b-file are open. Black wants everything shut: trade the rooks on the f-file, then bring the knight to f4 and the bishop and queen against a4, c4 and e4, three pawns that only White's pieces can guard.
- plan
- target
- The b-file
- plan
- outpost
- Pawns to guard
The same position after 16...a5, read for each side. White needs open files to use the bishop pair; Black needs the files closed or emptied of rooks, so that only the minor pieces and the fixed pawns are left.
Shah put Black's idea in plain terms: once all the rooks are traded, the weaknesses on c4, a4 and e4 will be attacked by two black minor pieces while only one white bishop can defend them. That is exactly what happened on moves 24 and 25. [1] Verified
Part 5The critical moments
Three decisions shaped the structure before the decisive one. Try each position before opening the answers.
Critical moment 1 of 4 · Move 11, Black to play
White pushes f4: open the position or keep it shut?
- threat
- idea not played
- key square
Spassky has played 11.f4 to open lines for the bishops. If the f-pawn and e-pawn trade, White's bishops and rooks come to life. How does Black meet the push?
Think it through on the boardWeigh the candidates
| Move | Stockfish | Verdict |
|---|---|---|
| 11...exf4 12.Bxf4 g5 13.e5 | +1.13White clearly better | Opens the f-file and the c1-h6 diagonal for White's bishop: Stockfish gives White more than a pawn. Byrne's line shows the danger, 12.Bxf4 g5 13.e5 tearing Black's position open. Verified |
| 11...Bg4 12.Qd2 exf4 13.Qxf4 | +0.15about equal | Stockfish's second choice, close to level. Shah's line 12.Qd2 exf4 13.Qxf4 still opens the f-file, which is the kind of position White wants. Verified |
| 11...O-O | +1.49White clearly better | Natural, but it does nothing about fxe5 and leaves the knight on e7 passive: Stockfish gives White well over a pawn. Verified |
What Fischer played
11...Ng60.00about equal
Fischer played 11...Ng6, attacking the f4 pawn and the knight on h4 at once. Byrne called it a weird reply that stopped White cold.
Stockfish's first choice too. Depth 31.
What it revealed
Fischer's first concern was the structure, not development: the knight move forced White to trade knights or release the tension, and both kept the centre closed. Black accepted doubled g-pawns after 12.Nxg6 fxg6 to keep it that way.
- Verified Byrne showed that 12.Nf5 Bxf5 13.exf5 Nxf4 and 12.fxe5 Nxh4 both favour Black.[1]
Critical moment 2 of 4 · Move 13, White to play
Take on e5, or keep the tension?
- idea not played
- key square
White's f-pawn and Black's e-pawn face each other. White has the bishop pair and doubled c-pawns. Should White resolve the tension now or keep it?
Think it through on the boardWeigh the candidates
| Move | Stockfish | Verdict |
|---|---|---|
| 13.O-O 13...b6 14.a4 a5 15.h3 O-O 16.g4 | 0.00about equal | Stockfish's choice: castle and keep the f-pawn's options (f5, or fxe5 at a better moment). The position stays level. Verified |
| 13.a4 | 0.00about equal | Just as good by Stockfish's count: fix the a-pawn first and keep the tension in the centre for another move. Verified |
| 13.f5 | −0.29about equal | Gains kingside space but hands Black the e5 square for good; Stockfish rates it no better than the game move, about a third of a pawn worse than castling. Verified |
What Spassky played
13.fxe5−0.24about equal
Spassky took on e5 with 13.fxe5, and after 13...dxe5 the centre was fixed for the rest of the game.
Stockfish preferred 13.O-O (0.00); the game move gave up about 0.2 pawns. Depth 30.
What it revealed
The capture released White's only lever. No white pawn can now challenge Black's e5 pawn, only g2-g3 could ever contest the f4 square, and White's c-, d- and e-pawns are frozen on light squares, where the bishop on d3 can only guard them.
- Verified Stockfish puts 13.O-O about a third of a pawn ahead of 13.fxe5. Small in number, but it is the moment the structure became permanent.
- Verified Shah notes that later players in this line preferred 13.O-O and pushing forward rather than taking on e5.[1]
- idea not played
- key square
- outpost
- key square
- Frozen centre
The road not taken and the road taken. Keeping the f-pawn on f4 keeps the structure flexible, which the side with the bishop pair needs; capturing makes it permanent, which the other side wants.
Critical moment 3 of 4 · Move 16, Black to play
White's a-pawn advances: let it come or stop it?
- threat
- target
Spassky has played 16.a4, threatening to push the pawn one square further and open a second queenside file for the rooks. How should Black handle the queenside?
Think it through on the boardWeigh the candidates
| Move | Stockfish | Verdict |
|---|---|---|
| 16...Ne8 17.Rxf8+ Kxf8 18.a5 | +0.26about equal | About a quarter of a pawn worse than the game move by Stockfish's count. Shah's line shows the problem: after 17.Rxf8+ Kxf8 18.a5 the queenside opens for White's pieces. Verified |
| 16...Ng4 17.Bd2 | 0.00about equal | Active-looking, but after 17.Bd2 the knight has gained nothing and the queenside is still open; Stockfish calls the position level, a quarter of a pawn worse for Black than the game. Verified |
| 16...Bd7 | 0.00about equal | Developing first lets White decide the queenside structure; Stockfish rates it level, again a quarter of a pawn below the game move. Verified |
What Fischer played
16...a5−0.29about equal
Fischer answered with 16...a5, stopping the a-pawn for good and leaving the b6 pawn backward on a half-open file.
Stockfish's first choice too. Depth 28.
What it revealed
Fischer accepted a weakness of Black's own on b6 to keep the queenside closed and to fix White's a-pawn on a light square, the colour of Black's remaining bishop. Eleven moves later that pawn decided the game.
- Verified Stockfish's first choice is the game move 16...a5; every alternative it was asked about lets White back to equality.
- Verified Shah called it a superb move fixing the a4 pawn on the light square, and said Fischer understood the importance of keeping the position closed.[1]
- Inferred The a4 pawn is the hidden cost of the doubled pawns: after 7.bxc3 there is no white b-pawn, so no pawn can ever defend a4.
Then came the slow part. Spassky swung a rook from b2 to f2 with 19.Rbf2, a move Byrne called a nothing move, though Byrne saw no clear way forward for White anyway. Fischer took the f4 square for good with 20...g5, brought the queen to g6, traded both pairs of rooks on the f-file and planted the knight on f4. [1] Verified
Part 6The turning point: 27.Qc2
By move 27 every black piece is better than its white counterpart, but the position is not lost. Byrne wondered whether the advantage was enough to win by force: Fridrik Olafsson thought so, Byrne was not sure. [1] Verified
Critical moment 4 of 4 · Move 27, White to play
Where does the queen guard best?
- target
- outpost
- White's back rank
Black's knight is on f4, the queen on g6 eyes e4 and g2, and the bishop on d7 eyes a4. White's queen on d3 guards e4 and the bishop on d1 guards a4. Find a queen move that keeps both pawns safe.
Think it through on the boardWeigh the candidates
| Move | Stockfish | Verdict |
|---|---|---|
| 27.Qe3 27...Qe8 28.Qf3 Kg8 29.h4 | −0.97Black clearly better | One of Stockfish's top choices. The queen guards e4 from e3 without blocking the bishop's diagonal to a4: Black is about a pawn better but has no immediate blow. Verified |
| 27.Qb1 | −1.09Black clearly better | Byrne's and Shah's suggestion. Stockfish scores it about a pawn for Black, a shade better than Qe3 in its own search, because the bishop on d1 still guards a4. Verified |
| 27.Qd2 | −5.24Black completely winning | Leaves e4 unguarded: 27...Qxe4 wins a pawn and the black queen goes deep into White's position. Stockfish gives Black more than five pawns. Verified |
What Spassky played
27.Qc2−7.98Black completely winning
Spassky played 27.Qc2. Fischer replied 27...Bxa4, and Spassky resigned without making another move.
Stockfish preferred 27.Qe3 (−0.98); the game move gave up about 7.0 pawns. Depth 26.
What it revealed
On c2 the queen guards both a4 and e4, but it also stands between the bishop on d1 and a4, so the bishop no longer defends the pawn. One piece guarding two pawns is overloaded: after 28.Qxa4 Qxe4 Black threatens mate on g2 and on e1 at the same time.
What Stockfish adds Stockfish's verdict jumps from Black being better by about a pawn (its first choice is 27.Qe3) to Black winning by about eight pawns. The doubled pawns did not lose the game by themselves; the overloaded queen did.
- Verified Byrne called it Spassky's biggest blunder of the match and gave 28.Qxa4 Qxe4 29.Kf2 Nd3+ 30.Kg3 Qh4+ 31.Kf3 Qf4+ 32.Ke2 Nc1 mate; declining with 28.Qb1 Bxd1 29.Qxd1 Qxe4 leaves White two pawns down.[1]
- Speculative Spassky had used much more of the clock (2 hours 9 minutes after 27.Qc2, against Fischer's 1 hour 18 minutes after the reply, by the published times), and the long passive defence may have worn down Spassky's vigilance.
28.Qxa4 Qxe4 29.Kf2 Nd3+ 30.Kg3 Qh4+ 31.Kf3 Qf4+ 32.Ke2 Nc1# 0–1Black completely winning
Byrne's line if White takes the bishop: the queen lands on e4 with two mating threats, and the king is chased to e2 and mated by the knight. Verified
The final blow was tactical, but it hit the two weaknesses the strategy had created. The a4 pawn had no pawn to defend it after 7.bxc3 and was fixed on a light square by 16...a5; the e4 pawn was fixed by 13.fxe5. The overload worked only because neither pawn could move. Inferred
Part 7What Spassky could have done
White had two windows. The strategic one was move 13, where keeping the f-pawn on f4 kept the structure fluid (the pair above). The practical one was move 27, where a queen move that kept guarding e4 without blocking the bishop's diagonal to a4 would have held for now.
- key square
- outpost
- pawn chain or line of force
27.Qe3 Qe8 28.Qf3 Kg8 29.h4 −1.28Black clearly better
Stockfish's defence: the queen shuffles between e3 and f3, and White tries h4 to loosen Black's kingside pawns. Verified
After 27.Qb1, Shah judged Black clearly better and perhaps close to winning: the black king walks to c7 to guard b6, the bishop goes to a6 and the knight to d6, and White can only wait. Stockfish's number is more modest, about a pawn, but the direction is the same. [1] Verified
Part 8A different angle: the doubled pawns were not the losing move
The usual summary is that Fischer gave Spassky doubled pawns and the bishop pair, then proved the knights better in a closed position. Checked move by move with Stockfish 19, the story is more precise. After 7.bxc3 and 8.e4 the engine still gives White a small edge. The damage began when the position stopped being fluid, and that was a White decision, not a Black one. Verified
| Move | Played by | Stockfish's choice | What it cost |
|---|---|---|---|
| 7.bxc3 (the doubled pawns) | Spassky | 7.bxc3 | nothing: White still slightly better |
| 13.fxe5 | Spassky | 13.O-O or 13.a4 | about a third of a pawn |
| 19.Rbf2 | Spassky | 19.Bd2 or 19.h3 | almost half a pawn |
| 20.Bc2 | Spassky | 20.Bd2 | about a fifth of a pawn |
| 27.Qc2 | Spassky | 27.Qe3 or 27.Qb1 | about seven pawns |
The second correction is about size. After 26...Nf4 Fischer's edge was real but, by the engine's measure, about a pawn, and Byrne doubted it was enough to win by force. Strategies against doubled pawns rarely win by themselves; they make the defender be perfect for a long time, and Spassky's one lapse came on move 27. [1] Verified
Compare game 6 of the same match, where Fischer as White chose the open, flexible side of the structure battle (pawn breaks in game 6). In both games Fischer decided early whether the position should be open or closed and played every move to keep it that way. For a bishop shut in by its own pawns, see good knight against bad bishop. Inferred
Part 9How to play with and against doubled pawns
Doubled pawns turn up in many club openings: the Nimzo-Indian, the Exchange Ruy Lopez, the Rossolimo Sicilian, and any line where a bishop takes a knight on c3, c6, f3 or f6. The recipe works from either side.
Playing against doubled pawns
- Close the centrePut your pawns where they block the opponent's breaks, ideally on the colour of the bishop you gave up, as Fischer did with ...d6 and ...e5.
- Fix the doubled pawnsMake the front pawn unable to advance: a pawn of yours on c5 against c4, or a piece on the square in front of it.
- Find the lonely neighbourCheck the pawn next to the doubled pair. If the capture removed its neighbour, it is isolated too, and often easier to win than the doubled pawns themselves.
- Trade the rooks, keep the minor piecesRooks use open files. With them gone, the side with the fixed pawns has fewer ways to make play, and your knight (ideally on an outpost like f4 here) and bishop do the attacking.
- With doubled pawns, keep a lever (an f-pawn break, a c-pawn push) as long as possible.
- With doubled pawns, open lines for the bishop pair before the opponent closes them.
- Against doubled pawns, close the position first and attack the pawns second.
- Before each capture, ask whom a permanent structure helps.
Your turnPractise it
Set each position up, decide your plan before you look, then play it out against Stockfish to see whether it holds.
Black to play after 26.Bd1: the rooks are gone. Where does the knight on h5 belong?
Hint
Look for a square no white pawn can ever attack.
A good answer
26...Nf4, as Fischer played and Stockfish's first choice. The knight is guarded by the g5 pawn (chasing it with g2-g3 would loosen White's king), eyes g2, d3 and e2, and supports the queen against e4.
Black to play after 27.Qc2: find the move that wins.
Hint
Which white piece is guarding two pawns at once, and what is the queen now blocking?
A good answer
27...Bxa4. The queen on c2 blocks the bishop on d1 from defending a4, so only the queen guards both a4 and e4. If 28.Qxa4 Qxe4 Black threatens mate on g2 and e1; anything else loses at least two pawns. Stockfish rates Black as winning.
Black to play after 27...Bxa4 28.Qxa4: how does Black punish the capture?
Hint
Which pawn did the white queen stop guarding when it took on a4?
A good answer
28...Qxe4, threatening Qxg2 mate and Qxe1 mate at once. Only 29.Kf2 (or giving up the queen with 29.Qe8+) parries both for a move, and Byrne's line 29.Kf2 Nd3+ 30.Kg3 Qh4+ 31.Kf3 Qf4+ 32.Ke2 Nc1 is mate.
White to play after 18...Rb8: Spassky played 19.Rbf2. Find a quieter move that keeps White's position more flexible.
Hint
Keep the rook where it ties Black to b6, and give the king some air or the bishop a better post.
A good answer
19.Bd2 or 19.h3. Stockfish rates both almost half a pawn better than 19.Rbf2: the rook stays on b2 against b6, and h3 also takes the g4 square from Black's knight.
Part 10What to take from this game
- Doubled pawns are a price, not a verdict. They are bad when the compensation (bishops, centre, files) is switched off.
- Fluid helps the owner, fixed helps the attacker. 13.fxe5 cost little by the numbers but made every weakness permanent.
- Count the lonely neighbour. The a-pawn left isolated by bxc3 was the pawn that fell.
- Trade the pieces that use open lines. With the rooks off, Black's minor pieces outnumbered White's defenders.
- Overload is the strategic weakness turned tactical. One queen guarding two fixed pawns lost both the pawns and the game.
Questions readers ask
Are doubled pawns bad in chess?
Not always. Doubled pawns cannot protect each other and are slow to advance, so they are a weakness in closed, slow positions. They are fine, even good, when the capture that created them gave something back, such as the bishop pair, an open file or more central pawns, and the position stays open enough to use it.
How do you play against doubled pawns?
Fix them first, then attack them. Block the front pawn so it cannot advance, close the centre so the opponent's bishops have no open lines, trade the rooks, and bring knights and your remaining bishop to the fixed pawns. Fischer's game 5 against Spassky in 1972 is the classic model.
Why did Spassky lose game 5 of the 1972 World Championship?
Spassky blundered with 27.Qc2, which blocked the bishop's defence of the a4 pawn and left the queen guarding both a4 and e4. Fischer's 27...Bxa4 won at once. The ground was laid by Fischer's closed structure and by 13.fxe5, which fixed White's pawns; Stockfish says Black was only about a pawn better before the blunder.
What is the Huebner Variation of the Nimzo-Indian?
It is the line 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 c5 5.Bd3 Nc6 6.Nf3 Bxc3+ 7.bxc3 d6, reached in game 5 by a different move order. Black gives up the bishop at once to double White's c-pawns, then sets the pawns on dark squares with ...e5 to close the centre and blunt White's bishop pair.
Sources and method
The game score was checked against two independent sources. Every position, line and candidate in this article was replayed for legality, and every engine number was produced by Stockfish 19 WASM Multithreaded for this article (depth 19 to 40, deepest at the critical moments), not copied from elsewhere. What the players thought is reconstructed from the moves, their published notes and later analysis; it is labelled as verified, inferred or speculative, and nothing is presented as a quotation unless a source is cited for it.
- Frederic Friedel, "Bobby Fischer in Iceland – 45 years ago (5)", ChessBase, 2017: the game score with clock times, Robert Byrne's notes (Chess Life & Review, September 1972) and Sagar Shah's notes
- World Chess Championship 1972: the game scores and the match results (Wikipedia)
- NM_Vanessa, "Bobby Fischer Ties Match", Chess.com, 2022: the 50-year anniversary report on game 5
Read next
Fischer's 20.e4 is the most praised move of Reykjavik 1972, but Stockfish 19 says it only asked a question. Learn when a pawn break wins and when it merely invites a mistake.
Attacking the Base of a Pawn Chain: Pragg v Carlsen, Norway 2026Praggnanandhaa broke Carlsen's French Advance chain at its base and won a pawn by move 11. Stockfish says it was level for twenty more moves: what the base attack really won was time.
The Maróczy Bind: Karpov v Kavalek 1974, Checked by StockfishKarpov clamped d5, traded queens and squeezed. Kavalek blamed a time-pressure slip on move 37. Stockfish 19 finds the biggest swing nine moves earlier, in a position Kavalek thought had a drawing flavour.
Turn reading into skill
Strategy sticks when you use it. Practise real positions with Stockfish's verdict at once, or play the plans out from any diagram above.