Week 5: how to go from 800 to 1200 in chess, while sliding back to 860

Week 5 (February 5–11, 2024) went backwards: 38 games, 14 wins, 24 losses, no draws, and my rating fell from 884 to 860 (-24). In the middle of it I lost nine games in a row (#158 to #166). That's a good moment to be honest about how to go from 800 to 1200 in chess, because this week shows exactly what was keeping me in the 800s, and it wasn't my openings.

The week in numbers

The rating curve tells the story: up to 909 after game #150 on Tuesday morning, then the streak from Tuesday night to Saturday, down to 846 after game #172, and a small recovery to 860. Seventeen of my 38 games were some kind of Queen's Pawn opening. As White, the games labelled Zukertort-Chigorin (1.d4 d5 with my bishop to f4 and knight to f3, against an early ...Nc6) went 5 wins and 1 loss. As Black, the games where White answered 1.d4 d5 with 2.Nc3 (labelled Chigorin Variation) went 0 for 4.

The numbers that matter most for the 800-to-1200 question are the mistakes: 78 blunders, about two per game, 50 of them in the middlegame. In six of my losses (#138, #151, #153, #158, #170 and #175) the engine had given me at least an 80% winning chance. I wasn't losing because I didn't know theory. I was losing positions I had already won.

Where the games turned

Three moments, all from the first day of the week. Evaluations are from White's side, so a minus number is good for me when I have Black.

Game #138: attacking a rook instead of giving check

Against Chriso0989 (957) I had Black and was clearly winning (-3.39). White's rook had reached g7. I moved my queen to f8 to attack it. What I didn't check was that my own rook on e3 had no defender, and White had a queen check on g5 that also attacked it. They played it, took my rook, and the evaluation went to +3.68.

. . r q . . . .
p p . . k p R .
. . . p . . . P
. . . p . Q . .
. . . P . . . .
. . P . r . . .
P P . . . . . .
. K . . . . . .

The engine's move was a check of my own: rook to e1. Checks come first because the opponent has to answer them. I lost this game after 99 half-moves.

Game #139: not taking back the pawn

Against mannyb97 (890), Black, a Ruy Lopez, the position was dead level (0.00) after White's pawn captured on b6. The obvious move was to take it back with my a-pawn. I moved my bishop to b5 instead. The engine's punishment: the white pawn steps to b7 and attacks my rook in the corner, and White is completely winning (+6.59). My opponent didn't find that exact move, but I was worse from then on and lost.

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

Game #140: missing the pin

With White against CuddlyBee (883) I was winning (+4.93). I put my queen on g5. The engine's move was bishop to a4, attacking Black's knight on c6 with the queen on d7 behind it. That's a pin: the knight can't move without exposing the queen. After my queen move Black could answer with a bishop move that attacks my queen, and the advantage dropped to +1.09.

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

I won by checkmate after Black blundered a move later, but the pin was the clean way.

The lesson of the week

All three best moves were forcing or obvious: a check (Re1+), a recapture (axb6) and a pin (Ba4). All three of my moves were "ideas": attack the rook, develop the bishop, bring the queen closer. At 800 I was skipping the simple question that comes before any idea: is there a check, a capture or a direct attack I should play right now? If there is, I play that first, and if the opponent has one, I deal with it first.

That's my working answer to how to go from 800 to 1200 in chess: not more theory, but taking back pawns, giving the strong checks and noticing pins, consistently. The nine-game streak was nine games of doing that badly.

All games this week

Earlier: week 4, three queen mistakes. Later: week 6, when things finally went up. The full diary is on the journey page, and you can run the same review on your own games.