# Week 57: is the London System good for beginners? One game, honestly > Week 57 was a single rapid game: a London System loss that cost 8 points (1272 → 1264). I was two pawns up until queenside castling walked into a pin. - URL: https://thechesscoach.netlify.app/journey/week-2025-w06/ - Updated: 2026-10-22 - Author: Ignacio Soler Week 57 is the shortest week in this diary: **one rapid game, on February 6, 2025. I lost it, and my rating went from 1272 to 1264 (−8).** It was a London System, the opening I play most as White along with the Zukertort setups, so I'm using it to answer something I'm asked often: is the London System good for beginners? One game proves nothing, but this one is a very clean example of how the London can give you a great position and how you can still lose it in three moves. ## The week in numbers | | | |---|---| | Rapid games | 1 | | Wins / losses / draws | 0 / 1 / 0 | | Rating | 1272 → 1264 (−8) | | Record with White / Black | 0–1 (all with White) | | Average accuracy | 80.4% | | Blunders (mine) | 1 | | Most played openings | London System (1) | My opponent was **cmmarquezs (1298)**, and the game lasted 17 moves ([see it on Chess.com](https://www.chess.com/game/live/132576007043)). The engine's accuracy split says everything: **95.4% in my eleven opening moves, 60.8% in my six middlegame moves.** My opponent finished at 89.6%. It ended with Chess.com marking it as abandoned; after my last move the engine had me at −4.44. ## How the game went The opening was textbook London: 1.d4 d5 2.Bf4 e6 3.Nf3 c5 4.c3 Nc6 5.e3 Nf6 6.Nbd2. Then Black played 6...b5, and after 7.Bxb5 I was a pawn up, +2.19 according to the engine. A few moves later my queen took the a7 pawn too (10.Qxa7), the queens came off with 11.Qc7 Qxc7 12.Bxc7, and I had **two extra pawns** and a bishop sitting deep in Black's camp on c7. That bishop is the star of this story. On move 13, Black had just played ...cxd4 and I had to choose how to take back. ### Move 13: the wrong recapture > **Taking back with the knight also attacks Black's bishop on c6** > Position (FEN): `r3kb1r/2B2ppp/2b1pn2/3p4/3p4/2P1PN2/PP1N1PPP/R3K2R w KQk - 0 13` > Played: cxd4 — cxd4 takes back with the c-pawn > Better: Nxd4 — Nxd4 takes back with the knight and attacks the undefended bishop on c6 I took with the pawn. The engine wanted the knight: Nxd4 takes the pawn and attacks Black's bishop on c6, which has no defender, keeping +2.32. My cxd4 opened the c-file, and the engine marks it as a mistake: +2.55 became +0.76. After 13...Bb5 14.Ne5 Bb4 both black bishops were in play, and my 14.Ne5 (an inaccuracy; the engine preferred a3) had already brought the evaluation down to −0.02. ### Move 15: castling into a pin This is the moment the engine picked as the key position of the week. > **Castling queenside puts my king on the same file as my undefended bishop on c7** > Position (FEN): `r3k2r/2B2ppp/4pn2/1b1pN3/1b1P4/4P3/PP1N1PPP/R3K2R w KQk - 3 15` > Played: O-O-O — O-O-O castles queenside, king to c1 and rook to d1 > Better: f3 — f3 is a quiet pawn move that keeps the king where it is for now Castling queenside looks natural: the king goes to safety and the rook joins the game. But the king lands on c1, on the **same file** as my bishop on c7, and that bishop had no defender. Black answered 15...Rc8. The engine describes that rook move as attacking the bishop on c7 with my king behind it: a **pin**, meaning the bishop can't leave the c-file without exposing the king, so it can't run away. The evaluation went from −0.07 to −3.99. The engine's quiet 15.f3 keeps the game balanced. After that, there was no rescue. Even the engine's best try, 16.Kb1, gives up the bishop after 16...Rxc7 and stays at about −3.9. I played 16.a3 (worse, −5.29), then 17.Nb3, Black answered 17...Ba4, and that's where the game ends. ## Is the London System good for beginners? My numbers Up to and including this game, I had played **76 games that Chess.com labelled exactly "London System"**: 37 wins, 34 losses and 5 draws, with an average accuracy of 80.1%. That's slightly below my overall score with White at the time (347 wins, 310 losses and 15 draws in 672 games). Counting every opening with "London" in the name, including the Accelerated London and the East Indian London lines, it was 160 games and 82–71–7. So in my hands the London was average, not a cheat code. What it does well is exactly what this game shows: simple development and a solid position. What it doesn't do is play the middlegame for you. ## The lesson of the week **Before castling, I check what else stands on the file or diagonal my king is going to.** In this game the danger wasn't an attack on the king; it was my own undefended piece sitting on the same file. Once the king arrived on c1, a single rook move pinned and won the bishop. The same check applies to every king move: if one of my pieces ends up between an enemy rook or bishop and my king, it is no longer free to move, and a piece that can't move is a piece that can be lost. ## All games this week | Game # | Date | Colour | Opponent (rating) | Result | Opening | Accuracy | Link | |---|---|---|---|---|---|---|---| | 1351 | Feb 6 | White | cmmarquezs (1298) | Loss (abandoned) | London System | 80.4 | [game](https://www.chess.com/game/live/132576007043) | Before this: [week 56](/journey/week-2025-w05/). After this: [week 58](/journey/week-2025-w08/), three games and a queen blunder. The whole diary is on the [journey hub](/journey/). Chess Coach gives your own games the same move-by-move review, [inside Claude](/features/ai-chess-analysis/). ## FAQ ### Is the London System good for beginners? In my games it was solid but not magic. Up to this game I had 76 games that Chess.com labelled exactly 'London System': 37 wins, 34 losses and 5 draws, slightly below my overall score with White. In this game it gave me 95.4% accuracy in the opening and a two-pawn lead. ### What went wrong when you castled queenside? My king went to c1, on the same file as my own bishop on c7, which had no defender. Black answered with Rc8, pinning the bishop to my king and attacking it, and the evaluation went from −0.07 to −3.99. ### Could the bishop on c7 be saved after Rc8? Not really. Even the engine's best try, 16.Kb1, loses the bishop to 16...Rxc7, and the evaluation stays at about −3.9. ## Related - [From 631 to 1385: my chess journey in 2,117 games](https://thechesscoach.netlify.app/journey.md) - [Week 56 (Jan 2025): Pirc vs Modern, 0–3 with Black](https://thechesscoach.netlify.app/journey/week-2025-w05.md) - [Week 58 (Feb 2025): a queen hung in one move, 58% accuracy](https://thechesscoach.netlify.app/journey/week-2025-w08.md) - [Can Claude analyze chess games? Yes, with Stockfish](https://thechesscoach.netlify.app/answers/can-claude-analyze-chess-games.md)