Week 68: London System vs the Englund Gambit, one game and one hanging bishop

Week 68 was a single game, and a short one. On May 2, 2025, I sat down for one rapid game, met the Englund Gambit with White, and lost 8 rating points: 1260 → 1252. The game lasted four moves each. If you play the London System and are looking up London System Englund Gambit lines, this is a small, honest example of how not to handle it, plus what Stockfish recommends instead.

The week in numbers

My opponent was Manoj_2303 (1276). Both of us played badly: one blunder each in eight half-moves, and accuracies in the low forties.

Forty Englund games before this one

The Englund Gambit starts 1.d4 e5: Black offers the e-pawn on the very first move. As a London player I meet 1.d4 openings all the time, and this one had a habit of beating me. Before this game I had faced it 40 times with White and won only 18, with 22 losses. This game made it 18 wins and 23 losses. Whatever its reputation, at my level it was working.

Moment 1: declining the pawn

r n b q k b n r
p p p p . p p p
. . . . . . . .
. . . . p . . .
. . . P . . . .
. . . . . . . .
P P P . P P P P
R N B Q K B N R

I didn't take. I played 2.c3, a London-style move that supports the d4 pawn. The engine calls it a mistake: the evaluation went from +1.24 to +0.02, from an edge to nothing. In other words, refusing the pawn gave Black an equal game without having to give anything up. dxe5 takes the pawn, and the engine's line shows Black's knight coming to c6 to attack it.

Taking doesn't mean giving up the London either. I checked the position after 1.d4 e5 2.dxe5 Nc6 with the engine: the top move is 3.Nf3 (+1.46), and the London move 3.Bf4 is second (+1.00). Both are "slightly better for White". So the London player's answer to the Englund is simple: take the pawn first, then play your normal moves.

Moment 2: the bishop on h6

Two moves later came the blunder that the analysis marks as the turning point (game).

r . b . k b n r
p p p p q p p p
. . n . . . . .
. . . . p . . .
. . . P . . . .
. . P . . N . .
P P . . P P P P
R N B Q K B . R

4.Bh6 puts the bishop on a square where two Black pieces can take it: the knight on g8 and the pawn on g7. Nobody defends it. The evaluation dropped from +0.84 to −7.76, which the engine describes as "Black is completely winning". The right move was e4, taking more of the center while Black's queen sits on e7.

For a beginner, this is the simplest kind of blunder there is: a piece moved to a square where it can be captured for free. There's no deep tactic to understand. It's a missing check before letting go of the piece.

What the engine can't explain

Then something odd happened. Manoj_2303 didn't take the bishop. He played 4...exd4, and the engine scored the position at −0.02, about equal again. And right there, after my opponent's fourth move, the record says I abandoned the game, with almost nine minutes still on my clock. The loss counted the same as any other.

I can't add anything to that from the data, except the lesson it obviously contains: a blunder that isn't punished is not a lost game. My opponent had made his own blunder, and the position was level.

The lesson of the week

Two lessons, both small. Against the Englund Gambit, take the pawn: the engine prefers 2.dxe5, and you can still develop like a London player after it. And before any piece move, especially a bishop going deep into the opponent's half, check whether the target square is attacked and whether anything defends it. Bh6 failed both questions.

All games this week

Previous: week 67 · Next: week 69 · Hub: all weeks from 631 to 1385 · Want an engine plus plain-words explanation of your own short losses? See can Claude analyze chess games.