# Week 115: why do I suddenly keep losing in chess? Only with White

> Week 115: 1228 → 1205 in 11 rapid games. With Black I went 4–2; with White I lost all five. Why I suddenly kept losing, in three positions.

- URL: https://thechesscoach.netlify.app/journey/week-2026-w17/
- Updated: 2026-10-22
- Author: Ignacio Soler

From April 20 to 25, 2026, I played **11 rapid games and dropped from 1228 to 1205**: four wins and
seven losses. Split by colour, the week makes a lot more sense: with Black I won four of six, and with
White I lost all five. When a player asks "why do I suddenly keep losing in chess?", my first suggestion
now is to separate the games before looking for one big cause. Mine were hiding in plain sight.

## The week in numbers

| | |
|---|---|
| Games | 11 |
| Wins / losses / draws | 4 / 7 / 0 |
| Rating | 1228 → 1205 (−23) |
| Average accuracy | 81.7% |
| My blunders | 25 |
| With White | 0 wins, 5 losses |
| With Black | 4 wins, 2 losses |
| Most played | Modern Defense with 1 e4 (2), Pirc Defense, Pirc Geller System |

## What went wrong with White

My five White games were five different openings, all starting with 1.d4: a Mikenas Defense, an East Indian London, an Indian Game, a Horwitz Defense and a Zukertort. The opening
phase wasn't the problem. In four of those games my accuracy over the first 12 moves was between 85.8%
and 94.6%. In the middlegame it fell to between 59.0% and 73.6%.

And in three of the five I had been winning. Game 2055 against rahulpcf (1220) was +6.04 on move 11. In
game 2061 against noobid123 (1271) I was at +5.71 on move 13, one move later at −6.25, and I was
checkmated on move 30. In game 2058 against lebskieeeee (1260) I was +5.13 on move 29 and two moves
later −4.76. The only White game that was short and simply bad was game 2064 against Donvinoth (1228):
two blunders and it was over in 10 moves.

My queen keeps showing up in those turning points. Queen to g4 in game 2055, queen to a4 in game 2061,
queen to b2 in game 2058. The first one has a board.

### Game 2055: the queen goes pawn hunting

> **Game 2055, move 11: White is completely winning. Where does the queen go?**
> Position (FEN): `1rbqk2r/ppppbppp/2n5/2nP4/Q7/2P2NB1/PP1N1PPP/R3KB1R w KQk - 1 11`
> Played: Qg4 — Queen to g4 attacks the g7 pawn, which has no defender
> Better: Qc2 — Queen to c2, a quiet move that keeps White completely winning

My pawn on d5 was attacking Black's knight on c6, and the engine had me at +6.04. My queen on a4 had to
move, because Black's knight on c5 was attacking it. I sent it to g4 to attack the undefended pawn on
g7. The engine shows the problem: Black's best answer is a pawn to h5, which attacks my queen again and
gains time. The evaluation fell to +2.17 after my queen move, still better for me but no longer crushing.
The engine's choice, queen to c2, does nothing flashy: after Black castles, my pawn takes the knight on
c6 and I stay completely winning. I went on to make five blunders in the game
and lost with 35 seconds left ([game 2055](https://www.chess.com/game/live/167582402718)).

## Meanwhile, with Black

My Black games were the Modern and the Pirc, my two most played defences in this whole series (171 Pirc
and 166 Modern games with Black). Three of the
four wins had accuracy of 90% or higher: 95.1% against netshredder34 (1255), 94.0% against ghz55l (1205)
and 90.0% against gilder55 (1198). But even these games have a key position each.

### Game 2056: a rook that was already under attack

Against Bew93 (1099), the lowest-rated opponent of the week, I was clearly better on move 7 (−2.79).
White's queen on f3 was looking down the long diagonal at my rook on a8, because my b-pawn had already
left b7.

> **Game 2056, move 7: White's queen sees all the way to a8**
> Position (FEN): `rn1qkb1r/p1pbpp1p/5np1/1p2p3/8/1B3Q2/PPPP1PPP/RNB1K1NR b KQkq - 1 7`
> Played: a5 — Pawn to a5, which leaves the rook on a8 hanging
> Better: Nc6 — Knight to c6 blocks the diagonal and keeps Black winning

I pushed my a-pawn, and the engine's reply for White is just queen takes rook. The evaluation went from
−2.79 to +0.46. Knight to c6 would have blocked the diagonal. White did take the rook, I blundered
again with a4, and then White's queen took a pawn on a4, where my bishop on d7 could capture it. I won by
checkmate on move 21 ([game 2056](https://www.chess.com/game/live/167608279980)).

### Game 2057: the pawn break I didn't play

Against ghz55l (1205), in the Pirc Geller System, I was clearly better on move 9 (−3.29). White's knight
sat on f4.

> **Game 2057, move 9: the e-pawn can attack the knight on f4**
> Position (FEN): `rnbq1rk1/pp2ppbp/3p1np1/2p5/5N2/1P4P1/PBPP1PBP/RN1QK2R b KQ - 5 9`
> Played: Qb6 — Queen to b6, a slower move
> Better: e5 — Pawn to e5 attacks the knight on f4 and takes the center

Pawn to e5 attacks the knight and grabs the center; the engine kept Black winning after it. After my
queen move, the engine's line has White's bishop take my knight on f6, then a white knight jumps to d5
and trades itself for my bishop, leaving me with doubled f-pawns. The evaluation went to −1.28. Still better for me, and I won by checkmate on move 60 with 94.0%
accuracy ([game 2057](https://www.chess.com/game/live/167619932894)).

## The lesson of the week

When I'm winning with White, my queen should stay home. Three of my five White losses turned on a queen
move in a good position. In game 2055, the move that kept a winning
position was a plain retreat. So my rule for the next weeks: if I'm clearly better and the queen is about
to go on an adventure, I look first for the quiet move that keeps everything protected.

## All games this week

| Game # | Date | Colour | Opponent (rating) | Result | Opening | Accuracy | Link |
|---|---|---|---|---|---|---|---|
| 2055 | Apr 20 | White | rahulpcf (1220) | Loss | Queen's Pawn Opening, Mikenas Defense | 76.8% | [game](https://www.chess.com/game/live/167582402718) |
| 2056 | Apr 21 | Black | Bew93 (1099) | Win | Pirc Defense | 84.7% | [game](https://www.chess.com/game/live/167608279980) |
| 2057 | Apr 21 | Black | ghz55l (1205) | Win | Pirc Defense, Modern Defense Geller System | 94.0% | [game](https://www.chess.com/game/live/167619932894) |
| 2058 | Apr 21 | White | lebskieeeee (1260) | Loss | Indian Game, East Indian London System | 77.5% | [game](https://www.chess.com/game/live/167625124412) |
| 2059 | Apr 22 | Black | netshredder34 (1255) | Win | Modern Defense | 95.1% | [game](https://www.chess.com/game/live/167669746642) |
| 2060 | Apr 22 | Black | spicyrabbit619 (1245) | Loss | Modern Defense with 1 e4 | 86.2% | [game](https://www.chess.com/game/live/167680169006) |
| 2061 | Apr 23 | White | noobid123 (1271) | Loss | Indian Game | 69.8% | [game](https://www.chess.com/game/live/167720167216) |
| 2062 | Apr 23 | Black | gilder55 (1198) | Win | Modern Defense with 1 d4 | 90.0% | [game](https://www.chess.com/game/live/167729721148) |
| 2063 | Apr 24 | White | Qilluer14 (1220) | Loss | Queen's Pawn Opening, Horwitz Defense | 69.4% | [game](https://www.chess.com/game/live/167749061090) |
| 2064 | Apr 24 | White | Donvinoth (1228) | Loss | Queen's Pawn Opening, Zukertort Variation | 69.5% | [game](https://www.chess.com/game/live/167766350742) |
| 2065 | Apr 25 | Black | KINGTONJr (1239) | Loss | Modern Defense with 1 e4 | 86.2% | [game](https://www.chess.com/game/live/167797309036) |

Read [week 114](/journey/week-2026-w16/) for what came before and [week 116](/journey/week-2026-w18/)
for what came next, or go back to [all weeks of the journey](/journey/). Every evaluation on this page comes
from Stockfish, and you can [review your own games like this](/features/ai-chess-analysis/).

## FAQ

### Why do I suddenly keep losing in chess?

Split your results before guessing. In my week 115 the losses weren't random: I lost all five games with White and won four of six with Black. In three of the five White losses I had a winning position first.

### Is it normal to lose five games in a row with one colour?

It happened to me between April 20 and 24, 2026, at about 1220. My openings with White were fine (85.8% to 94.6% accuracy in the first 12 moves in four of the games); the middlegames were not.

### Should I stop grabbing pawns with my queen?

In game 2055 my queen went to g4 to attack a pawn when I was completely winning. The engine's best answer for Black is a pawn move that kicks the queen, and its evaluation fell from +6.04 to +2.17. A calm queen retreat kept the big advantage.

## Related

- [From 631 to 1385: my chess journey in 2,117 games](https://thechesscoach.netlify.app/journey.md)
- [Week 114 (Apr 2026): 46 blunders in 12 games, still +9](https://thechesscoach.netlify.app/journey/week-2026-w16.md)
- [Week 116 (Apr 2026): why I keep blundering pieces at 1200](https://thechesscoach.netlify.app/journey/week-2026-w18.md)
- [Free AI chess analysis of your games, explained](https://thechesscoach.netlify.app/features/ai-chess-analysis.md)