The London System is White's setup of d4, Bf4, e3, c3 (or c4), Nf3, Nbd2 and Bd3 — a structure White can aim for against almost anything Black plays. Its appeal is that you learn plans instead of variations: the same seven or eight moves produce a sound position, and the middlegame ideas repeat. Its honest limitation is that it does not fight for an opening advantage; it fights for a playable, familiar game, and well-prepared Black players have several comfortable answers.

This guide covers the full repertoire: the move orders and where they differ, the setup itself, the three real middlegame plans, and — the part most London articles skip — exactly how Black tries to make your life difficult.

The starting position and the name

The opening takes its name from the London 1922 tournament, where the d4/Bf4 setup was played. It has been used at every level since, including by Magnus Carlsen and other elite grandmasters who use it as a low-theory way to reach a fight.

Everything starts with the bishop move that defines the system:

After 1.d4 d5 2.Bf4 — the dark-squared bishop gets outside the pawn chain before e3 shuts it in. Play this position →

That single idea is the whole point. In the Queen's Gambit and most d4 openings, White plays e3 early and the c1-bishop spends the game staring at its own pawns. The London solves the problem first and builds the pawn chain afterwards. Compare it with the Colle System, which accepts a passive c1-bishop in return for a fast e3-e4 break, and you can see what White is trading.

The full setup, in the order most players reach it, is d4, Bf4, e3, Nf3, c3, Nbd2, Bd3 and short castling. The pawns on c3-d4-e3 form a chain that is very hard to break, every minor piece defends another, and there is no move Black can play that leaves White's position loose.

Move order: 2.Bf4, 2.Nf3, and the cost of committing early

There is no single "correct" London move order. Each one avoids something and allows something else, and it is worth knowing which trade you are making.

1.d4 d5 2.Bf4 is the most direct. It gets the bishop out before Black can prevent it, but it commits instantly. Black can hit the centre with 2...c5, and because the c1-bishop has left home, the b2-pawn is undefended — the root of every problem the London has.

1.d4 d5 2.Nf3 keeps things flexible for one more move and only then plays 3.Bf4. It sidesteps some immediate ...c5 tricks, but it allows Black a free move to arrange a setup you may like less, and it does nothing about ...Qb6 later. Neither order removes Black's main resource; you are choosing which version you would rather meet.

Against 1...Nf6 you have the same choice. Note one specific annoyance: after an early ...d6, Black can play ...Nh5 attacking the f4-bishop. White does not have to panic. The bishop steps to g5 or g3, and if Black spends more moves chasing it with ...h6 and ...g5, the trade on g3 gives White an open h-file for the rook — a genuine attacking asset, paid for with a doubled pawn.

The Jobava London

A modern relative, popularised by Baadur Jobava, develops the queen's knight to c3 instead of playing c3 with the pawn:

The Jobava London: 1.d4 Nf6 2.Nc3 d5 3.Bf4. White eyes e4 and Nb5, at the cost of the solid c3-pawn. Play this position →

This is a genuinely different opening, not a variation. It is sharper, more aggressive and much more theoretical, because the knight on c3 supports an early e4 and Nb5 hits c7. If the classical London feels slow to you, this is where London players usually go next — but do not expect the same "no memorisation" promise.

The main tabiya

Against Black's most natural handling — ...Nf6, ...e6, ...c5, ...Nc6, ...Bd6 — you reach the position every London player should recognise instantly:

The classical tabiya after 1.d4 d5 2.Bf4 Nf6 3.e3 e6 4.Nf3 c5 5.c3 Nc6 6.Nbd2 Bd6 7.Bg3 O-O 8.Bd3. Play this position →
1. d4 d5
2. Bf4 Nf6
3. e3 e6
4. Nf3 c5
5. c3 Nc6
6. Nbd2 Bd6
7. Bg3 O-O
8. Bd3

Two details in that sequence matter more than they look.

7.Bg3 rather than Bxd6. When Black offers the trade with ...Bd6, retreating keeps a well-placed bishop and keeps the tension. Trading hands Black the pair of bishops and, after ...Qxd6, gives Black's queen a good central square for free. There are positions where Bxd6 is right — usually when Black must recapture with a pawn and damage the structure — but as a default, step back.

Keeping the c5/d4 tension. Do not rush cxd4 or dxc5 exchanges. If Black plays ...cxd4, recapture with the c-pawn (cxd4) in most structures: it keeps a pawn on d4, opens the c-file for a rook, and leaves e3 available. Recapturing with the e-pawn is usually a concession, because it frees Black's game and blocks your own bishop's diagonal.

Plan one: the e5 outpost

The single most reliable London plan is to plant a knight on e5. The square is supported by d4 and by a knight on d2 that can follow to f3, and it is protected against ...f6 ideas because the f4/g3-bishop watches the b8-h2 diagonal.

9.O-O b6 is met by 10.Ne5 — the knight sits on the strongest square in the position. Play this position →

If Black takes with ...Nxe5, recapture with the d-pawn or the f-pawn depending on what you want. dxe5 gains space, kicks the f6-knight and opens the d-file, but concedes the d5-square. fxe5 (after f4 has been played) opens the f-file straight at f7 and builds an attacking wedge. Learn to ask which one your pieces are already pointing at — this is a standard outpost decision, not a memorised rule.

If Black leaves the knight alone, back it up with f2-f4:

The full attacking formation: pawns on c3-d4-e3-f4, knight on e5, bishops on d3 and g3, all aiming kingside. Play this position →

This is the structure the London is really about. From here the standard follow-ups are Qf3 and Rf3-h3 with the rook lift, Ndf3 supporting the outpost twice, or Qc2 combining with Bd3 on the b1-h7 diagonal — the exact battery that makes the Greek gift sacrifice Bxh7+ a live threat whenever Black's defenders wander.

The cost is real: f2-f4 loosens the king and leaves e4 and g4 as permanent holes. Play it when your pieces are ready to use the f-file, not automatically.

Plan two: the queenside squeeze

The London is not only an attacking system, and against solid Black setups the kingside plan simply will not work. Then White switches wings.

If Black has committed to ...c6 and ...d5 (a Slav-like structure), White can expand with b2-b4, a2-a4 and b4-b5, hitting c6 with fewer pawns than Black has there — the minority-attack idea familiar from Carlsbad structures. The point is not to win a pawn but to leave Black with a weak, backward c-pawn on a half-open file that your rooks then attack.

If Black has played ...c5, the plan changes shape: b2-b4 becomes a clamp, taking c5 away and gaining queenside space, usually prepared by a3 and Rb1. In both cases the c3-pawn is what makes it work — it holds d4 so the b-pawn is free to advance.

A third, quieter version: Qb3 and Rfc1 with pressure on b7 and the c-file. Against passive setups this is frequently enough, and it requires no risk at all. Knowing when to abandon the kingside dream is what separates a London player from someone who only knows the moves — see from opening to middlegame for the general skill.

How Black fights the London

Here is the part that decides whether you enjoy playing this opening. Black has three serious approaches, and none of them require deep memorisation either.

1. Early ...c5 with ...Qb6

The critical test. Black attacks d4 and, because the bishop has left c1, immediately targets b2:

1.d4 d5 2.Bf4 c5 3.e3 Qb6 — b2 hangs and d4 is hit. This is the position every London player must have an answer to. Play this position →

You have three sound approaches, and you should pick one and stick to it:

  • 4.Qc1 — the solid choice. It defends b2 and holds everything, at the price of a passive queen for a few moves. Perfectly playable; you unwind with Nf3, Be2 and O-O.
  • 4.Qb3 — offer the trade. If Black takes, axb3 gives White a healthy structure and the open a-file, and Black's queenside pressure evaporates with the queens.
  • 4.Nc3 — the gambit approach. White ignores b2 and develops. The resulting positions are genuinely sharp and require concrete study, which is the opposite of what most people choose the London for. Look at it before you play it.

What you must not do is play on autopilot. Nbd2 and Bd3 do not defend b2, and losing that pawn for nothing is the most common way London games go wrong.

2. The mirror: ...Bf5

The cleanest equalising idea is to copy you. Black develops the light-squared bishop outside the chain before playing ...e6:

1.d4 d5 2.Bf4 Nf6 3.e3 Bf5 — seen from Black's side. Black solves the problem piece and heads for a symmetrical, comfortable game. Play this position →

There is no refutation here, and you should not look for one. The practical reply is to use the fact that Black's bishop has left the queenside: Qb3 hits b7 and d5 at the same time, and c2-c4 challenges the centre immediately, since Black no longer has the light-squared bishop defending b7. Another good practical try is Bd3 offering the trade of your worse bishop for Black's good one.

3. The fianchetto setups

Black can ignore d5 entirely and play a King's Indian or Modern Defense structure with ...g6 and ...Bg7:

The London against a fianchetto: 1.d4 Nf6 2.Bf4 g6 3.e3 Bg7 4.Nf3 O-O 5.Be2 d6 6.h3, giving the bishop the h2-square and stopping ...Bg4. Play this position →

Against these setups the e5 outpost is much less effective, because the g7-bishop stares straight at it and Black meets Ne5 with ...Nfd7 or ...c5. Note the small but important prophylactic move h3: without it, ...Bg4 pins the f3-knight and ...Nh5 or ...Nd7 hits your bishop with tempo. White's play here is usually c3, Nbd2, O-O and a slow queenside expansion — a genuine game where you have no advantage but a very sound structure. If you want to understand what Black is doing, read the King's Indian Defense guide from the other side of the board.

An honest assessment

The London System is solid, not critical. That distinction matters:

  • It does not challenge Black's equality. Correctly played, Black equalises against the classical London by several routes. If you play it hoping for an opening advantage, you will be disappointed.
  • It saves enormous preparation time. You dodge the Nimzo-Indian, the Grünfeld, the Slav, the King's Indian and the Benoni with one set of ideas. For anyone who plays fewer serious games than they would like to prepare for, that is worth more than a small theoretical edge.
  • It rewards understanding, not memory. The positions are quiet and long. You win them by outplaying people, which means your middlegame and endgame matter far more than your opening file.
  • It has a specific downside: predictability. An opponent who has seen your London twice knows what is coming. Have a second setup — the Torre Attack or Trompowsky share the early-bishop philosophy and cost almost nothing extra to learn.

The five mistakes that actually lose London games are: forgetting b2 after ...Qb6; trading Bxd6 for no reason; playing f4 before the pieces are ready; recapturing exd4 instead of cxd4 and killing your own bishop; and drifting with no plan once the setup is complete. Fix those five and you have a repertoire that will serve you for years. If you are assembling a full opening file, the repertoire-building guide shows how a system opening fits alongside your Black defences, and opening principles explains why the London's move order works at all.

Practise the London on the free board

Reading about the e5 outpost is not the same as feeling it. Every diagram on this page has a Play this position link that drops the position straight onto the free board at LocalChess — no account, no sign-up, nothing to install.

Start with the main tabiya and play it out against the engine a dozen times: try the Ne5 and f4 plan once, the queenside b4-b5 plan once, and see which structures suit you. Then set up the ...Qb6 position and decide, for yourself, whether you are a Qc1 player or a Qb3 player. Once you have made that choice you have a complete White repertoire against 1...d5 and 1...Nf6 — and you can spend the study time you saved on endgames instead.