Its heyday was in the decades before and after World War Two, when the Colle was a popular alternative for d4 players seeking to avoid theoretical lines in the Queen's Gambit and Indian systems.
In more modern times it fell under the shadow of the Trompowsky (1.d4 Nf6 2.Bg5) and Pseudo-Trompowsky (1.d4 d5 2.Bg5), and even more recently has played second fiddle to the London (1.d4, 2.Nf3 & 3.Bf4), the New London (1.d4 & 2.Bf4) and the Jobava-Prié (1.d4, 2.Nc3 & 3.Bf4).
But in round two of the Thuringia seniors' championships, in which I was Black against Rolf Müggenburg (1716), the Colle was played, and I answered with an anti-Colle system of my own devising, albeit one which was independently discovered by players long before I came up with it.
The game began 1.d4 d5 2.Nf3 e6 3.e3.
 |
| Position after 3.e3 |
Wikipedia says of the opening: "A major theme of the Colle System is the ambition to play a well-timed e4, where the square is defended by the bishop on d3, the knight on d2 (following Nbd2), and possibly the rook on e1 (following 0-0 and Re1)."
The Colle was developed by the Belgian Edgar Colle, and further refined by the Belgian -American George Koltanowski.
Wikipedia states: "The opening had even been referred to as the 'dreaded' Colle System."George Koltanowski, in his book The Colle System, said it offered solid development, combinations and a decent endgame, giving White 'good chances of not losing against a stronger player'."
From the diagram Black normally proceeds with 3...c5 or 3...Nf6, but there are 667 games with 3...Bd6, which is what I played.
After 4.Bd3 - this is much more popular than a Queen's Gambit transposition with 4.c4 - the main continuation is 4...f5, transposing to the Stonewall Variation of the Dutch.
But I chose 4...Nc6, which occurs in just 17 games in ChessBase's 2026 Mega database.
 |
| However, with transpositions, the diagram position appears 53 times in Mega26 |
*****
*****
*****
*****
The 'typical' Colle continuation in such setups is 5.c3, which is what was played in my game.
The move is multi-purpose: it prevents White's light-square bishop being harassed by ...Nb4; it supports the d pawn in anticipation of playing e4; it lets the light-square bishop later withdraw to c2 or b1 so as to form a bishop-queen battery on the b1-h7 diagonal.
However, Black can now be the first player to get in a central-pawn break, by playing 5...e5.
This only occurs in three of 38 games to reach the position in Mega26, but it seems perfectly sound, and has the psychological advantage of cutting across the Colle player's normal system of development.
After 6.dxe5 Nxe5 7.Nxe5 Bxe5 Black has completely equalised, according to Stockfish17.1 and Dragon1.
Indeed, the engines reckon White should eschew normal Colle development in favour of playing 5.Nc3.
Now 5...e5? drops a pawn to 6.Nxd5, but Black can 'punish' White's omission of c3 by continuing 5...Nb4.
White should probably preserve the light-square bishop with 6.Be2, when Black can play 6...c5, after which the game might continue 7.a3 Nc6 8.Nb5 Be7 9.dxc5 Bxc5 10.b4 Be7 11.Bb2 Bf6, when Dragon1 reckons White enjoys a normal slight-edge opening advantage, but Stockfish17.1 believes Black has equalised.
