Weak Colour Complex: How Petrosian Took Mecking's Dark Squares

Part 1The setting: a former champion and a teenage star

In January 1971 the Hoogovens tournament in Wijk aan Zee brought Soviet grandmasters together with the best of the West. Viktor Korchnoi won it with 10 out of 15. Tigran Petrosian, world champion from 1963 until losing the title to Boris Spassky in 1969, played in the field, and so did Henrique Mecking of Brazil, a teenager who would reach number three in the world in 1978. ^[4][5][6] ✓ Verified

Petrosian was known for a style built on prophylaxis, Aron Nimzowitsch's idea of stopping the opponent's plan before making your own, and for a defence so hard to break that the nickname was Iron Tigran. Mecking was known for lively piece play. Petrosian won their games in 1969, 1971 and 1972. ^[1][3] ✓ Verified

Petrosian's judgement of the young Brazilian, quoted by Mark Dvoretsky and Artur Yusupov, was that Mecking did not understand the importance of weak and strong squares: lost in 1969 on the light squares, a year later on the dark squares, and again on the dark squares in San Antonio. ^[3] ✓ Verified

Part 2What is a weak colour complex?

A weak colour complex is a group of squares of one colour, light or dark, that no pawn of yours can ever guard. It appears when your pawns stand on one colour and your bishop for the other colour is gone. The squares become permanent homes for enemy pieces: a knight, a queen or, in the ending, the king.

Two things make the weakness. The first is the pawns: a pawn guards the two squares diagonally in front of it, and once the pawns that could guard a square have advanced past it or been exchanged, nothing can ever guard it again. The second is the missing bishop: as long as a bishop of that colour is on the board, it can cover the holes. Take it away, and they are open.

The left board is only a small edge for White because there is little material left to use the holes. The weakness grows with the pieces that can reach it, and above all with the king in an ending, when no enemy pawn can drive it away. ◐ Inferred

Part 3The opening: how the e5 pawn fixed Black's structure

Petrosian chose a quiet system, the Torre Attack (1.d4, 2.Nf3 and 3.Bg5), with c3 and Nbd2: a solid set-up that keeps options rather than claiming an edge. Black's ...c5 and ...d5 gave a Queen's Gambit-like centre. After 6...cxd4 7.exd4 the c-pawn was gone, and with it Black's only pawn that could ever guard d6.

The key moment came at move 11. White's knight jumped to e5 and Mecking took it, so the d4 pawn recaptured onto e5. That pawn stands on a dark square, cramps Black's kingside and takes d6 and f6 away from Black's pieces. Two moves later White's bishop sat on d4 behind it, and on move 15 Black answered the queen's attack on g7 with ...g6, which left f6 with no pawn cover at all. ◐ Inferred

None of these moves is a blunder by Stockfish's count: the evaluation drifts from close to level to about a pawn and a quarter for White over the first fifteen moves. The damage is structural, and it is permanent, because pawns cannot go back. ✓ Verified

Part 4The plans: trade the guard, or close the door

From move 16 to move 27 both sides manoeuvre. White's plan has three steps: keep the pawn on e5, trade the dark-squared bishops at the right moment, and invade the holes with the queen and later the king. Black's counterplay lives on the light squares: a knight on e4, a bishop on the long diagonal, perhaps ...f5 to stake out the light squares in front of the e5 pawn. ◐ Inferred

Stockfish's best lines for Black in this phase keep coming back to ...Ne4, and earlier to ...Bc5 with a trade of bishops on d4. Petrosian avoided that trade: 21.Be3 and 23.Bc1 pulled the bishop back from d4 rather than let Black exchange it on Black's terms. ◐ Inferred

Part 5The critical moments

The evaluation climbed in steps, not in one jump. Try each position before opening the answer.

Part 6The turning point: the queen trade on f6

By move 32 the dark-squared bishops were gone and White's queen stood on f4. Black's king on g7 had no pawn able to guard f6 or h6. White had a choice between keeping the queens to attack the king and trading them to win the ending.

The rest is a clinic in using a colour complex. White's king walked g2, f3, e3, d4 and on to e5 and d6; the knight came to e5 by way of f3; b4 and b5 fixed Black's queenside pawns on dark squares; and Black's bishop on e8 never found a useful square. ^[1][2] ✓ Verified

Part 7What Mecking could have done

There was no single losing move, which is why this game teaches so well. Black's chances were in the first twenty moves, and they share one theme: trade White's dark-squared bishop while it stands on d4, on Black's terms, before White can trade it on g5 on White's terms. ◐ Inferred

A second idea is to change the structure: ...f6 to challenge the e5 pawn, or ...f5 to claim the light squares in front of it. Stockfish's choice on move 26 was 26...f5, keeping a light-square fortress on d5, e4 and f5 and staying about a pawn and a third behind. ✓ Verified

Part 8A different angle: there was no single present

Petrosian's own verdict was that Mecking handed over all the dark squares. Read alongside Stockfish, the game shows how that happens: not with one bad move but with a string of small ones, each costing between a fifth and a half of a pawn, and each making the next trade easier for White. ^[3] ✓ Verified

The obvious moment to blame, the bishop trade on g5, cost Black almost nothing by Stockfish's count. By move 30 the trade was coming one way or another; the dark squares were lost earlier, while it still looked like a normal middlegame. ✓ Verified

One detail in the record is worth a note. Petrosian's recollection places the dark-square game a year after their 1969 meeting, but the game score in both sources is from the Wijk aan Zee tournament of January 1971; there is no 1970 game between them in the archive. Memory rounds dates; the moves are the evidence. ◐ Inferred

There is a modern echo too. Engines judge which bishop guards the weak squares, not which one looks good, and they keep proposing trades that break the textbook rule about good and bad bishops. Petrosian was perhaps playing that way fifty years before the engines. ? Speculative

Part 9How to exploit a weak colour complex

You do not need Petrosian's patience to use the idea. It comes up in every opening where one side's pawns end up on one colour: the French, the Caro-Kann, the Queen's Gambit, the Torre Attack used here. Related plans are covered in good knight against bad bishop and the right exchange.

• Before each bishop trade, ask which squares each bishop guards, not which bishop is good or bad.
• As the defender, trade the attacker's best piece on your weak colour early, while it is still possible.
• As the defender, keep the king off the weak squares.
• In the ending, the king is the strongest piece in a colour complex: bring it in.

Your turnPractise it

Set each position up, decide your plan before you look, then play it out against Stockfish to see whether it holds.

Part 10What to take from this game

• Pawns make the colour. Black's pawns on light squares made the dark squares weak long before any bishop was traded.
• Trade the guard, not the bad piece. Petrosian traded the bishop that looked worse because it took Black's only defender.
• Defend early. Black's best try, 13...Bc5, offered a trade that looks like nothing; twenty moves later there was no trade left to offer.
• Swap queens into the bind. The queen trade on f6 turned an edge into a winning ending where the king used the holes.
• A bind forgives. Petrosian's slow moves in the ending cost little, because Black had no way to fight back.