Week 46: why do I suddenly keep losing in chess? From 1159 to 1105
Week 46 went up and then fell off a cliff. Between November 18 and 22, 2024, I played 17 rapid games: I won four in a row to reach a new personal best of 1159 in game 1136, and then I lost the next seven. The week ended 6 wins, 11 losses, with my rating at 1105, down from 1140. If you've ever asked yourself "why do I suddenly keep losing in chess?", I asked it too, and this time the data had a fairly clear answer.
The week in numbers
| Games | 17 (#1127–#1143) |
| Wins / losses / draws | 6 / 11 / 0 |
| Rating | 1140 → 1105 (-35) |
| Average accuracy | 76.9 % |
| Blunders (my moves) | 46 (2.7 per game) |
| Most played openings | London System (4), Pirc Defense Classical Variation (3), Indian Game East Indian London System (2) |
Why do I suddenly keep losing in chess? Three reasons in my data
My opponents got stronger. At 1140–1159 I was paired with higher-rated players: my opponents averaged 1153 this week, against 1107 the week before and 1037 two weeks before. During the seven-game streak they averaged about 1165; the first of them, pole1380k, was rated 1201, like leon751, who had beaten me earlier in the week.
I was trying a new opening with Black. The week before I had played the Pirc Defense for the first time. This week I played it four times and won once, against Messi_ank13 (1129), the game that took me to 1159. The other three Pirc games were losses, two of them by checkmate.
I blundered more. 46 blunders in 17 games is 2.7 per game, up from 1.8. My accuracy dropped from 81.5 % to 76.9 %, and in the seven losses it averaged about 72 %. Game 1142 against mikaellaurio (1148) alone had 11 blunders of mine.
What didn't change: my White setup. The two London System wins were among my best games of the week, including game 1129, 94.8 % against Robertthorsteinsson (1163), in a game of 119 half-moves. Only three of the eleven losses were abandoned; most went to checkmate or resignation, so I can't blame quitting this time.
Three key moments
Game 1128: not taking the bishop that attacked my queen
With Black against BENELORDD (1143), a Ruy Lopez, White's bishop on g5 was attacking my queen on e7. That bishop had no defender.
Position.py even says it in my move's description: f6 "attacks the bishop, which has no defender". If it has no defender, take it. Instead, White's reply exf6 attacked my queen and my bishop on g7 at the same time, "white is completely winning" (+6.66). After Qxg5 the engine's line gets messy (13.Qxf7+ Kd8 14.Qxg7) but stays only "slightly better for white". I resigned the game a few moves later.
Game 1130: the queen move that left my bishop alone
With White against leon751 (1201), I was slightly better (+0.75). My queen on a4 was defending my bishop on b5, and I moved it to c2.
The engine's reply for Black is Nxd4: the knight takes a pawn, attacks my queen and, in position.py's words, "opens a line: the queen on the queen's side now attacks the bishop on the queen's side". After 16.cxd4 Qxb5 the bishop is gone, -1.06. The game analysis preferred Bg5; a fresh engine run also likes quiet moves such as b4 and Nb3 at around +1. None of them move the queen away from b5's defence. In the game my opponent missed it, I later blundered a bishop with 17.Bh6 (-6.54), and I was checkmated after 104 half-moves with 9 blunders.
Game 1127: a bishop that White's pawns could trap
This is a game I won, from the start of the week. With Black against Widderk2022 (1131), I was clearly better (-1.80) and retreated my bishop from c5 to b6.
The engine's reply is a5, a pawn that attacks the bishop; after 11.a5 Bc5 12.b4 a second pawn attacks it, and the line ends with White taking it, "slightly better for white" (+1.57). My own a5 first kept "clearly better for black". My opponent played 11.b4 instead, I answered a5, and I won the game.
The lesson of the week
"Suddenly" losing wasn't sudden. The opponents were more than 100 points stronger than two weeks earlier, I was learning a new opening with Black, and stronger players punished the same blind spot I always have: undefended pieces. In game 1128 I didn't take one; in game 1130 I created one; in game 1127 I put one where pawns could chase it. My rule from this week: before every move, look at my pieces that are defended only once and the opponent's pieces that aren't defended at all. Against 1100+ players, the free piece doesn't stay on the board for long.
All games this week
| Game # | Date | Colour | Opponent (rating) | Result | Opening | Accuracy | Link |
|---|---|---|---|---|---|---|---|
| 1127 | Nov 18 | Black | Widderk2022 (1131) | Win | Giuoco Piano Game Giuoco Pianissimo Variation | 75.0 % | game |
| 1128 | Nov 18 | Black | BENELORDD (1143) | Loss | Ruy Lopez Opening Cozio Defense | 80.0 % | game |
| 1129 | Nov 18 | White | Robertthorsteinsson (1163) | Win | London System | 94.8 % | game |
| 1130 | Nov 18 | White | leon751 (1201) | Loss | Queens Pawn Opening Zukertort Variation | 59.8 % | game |
| 1131 | Nov 19 | White | Nbarboutis (1184) | Loss | Indian Game East Indian London System | 85.0 % | game |
| 1132 | Nov 19 | Black | bowz_onin (1126) | Loss | Pirc Defense Classical Variation | 65.5 % | game |
| 1133 | Nov 20 | White | KDQI13 (1133) | Win | Queens Pawn Opening Zukertort Chigorin Variation | 69.6 % | game |
| 1134 | Nov 20 | Black | Theo82140 (1123) | Win | Bishops Opening | 96.1 % | game |
| 1135 | Nov 20 | White | shadowlolog (1117) | Win | London System | 92.8 % | game |
| 1136 | Nov 20 | Black | Messi_ank13 (1129) | Win | Pirc Defense Classical Variation | 82.2 % | game |
| 1137 | Nov 20 | Black | pole1380k (1201) | Loss | Pirc Defense Classical Schlechter Variation | 71.9 % | game |
| 1138 | Nov 20 | Black | Vlada_M (1178) | Loss | Pirc Defense Classical Variation | 67.8 % | game |
| 1139 | Nov 20 | White | MGTyr (1187) | Loss | London System | 80.9 % | game |
| 1140 | Nov 21 | White | BoranTimur09 (1156) | Loss | London System | 72.4 % | game |
| 1141 | Nov 21 | Black | ZozoCM (1146) | Loss | Indian Game East Indian London System | 73.1 % | game |
| 1142 | Nov 21 | White | mikaellaurio (1148) | Loss | Indian Game | 74.4 % | game |
| 1143 | Nov 22 | Black | raedon707 (1142) | Loss | Nimzowitsch Defense Declined Williams Variation | 65.2 % | game |
Previous: week 45 · Next: week 47 · The journey · Review your own games like this
Frequently asked questions
Why do I suddenly keep losing in chess?
In my week 46 the data shows three concrete reasons: stronger opponents (1165 on average during my seven-game losing streak), a new opening with Black (the Pirc, 1 win and 3 losses), and more blunders (2.7 per game against 1.8 the week before).
Is it normal to lose seven games in a row right after a personal best?
It happened to me: I reached 1159 in game 1136 and then lost games 1137 to 1143, falling to 1105. Seven losses in a row also happened to me in week 37, around 1000.
What was the most common mistake in the losing streak?
Undefended pieces. In game 1128 I didn't take an undefended bishop that was attacking my queen, and in game 1130 my queen move left my own bishop without a defender.