How to avoid blundering in chess: what 91 blunders in one week taught me
June 3 to June 8, 2024 was my busiest week so far in this stretch: 29 rapid games, 11 wins, 17 losses and 1 draw, and the rating went from 975 to 956. The number that defines it, though, is 91. That is how many blunders Stockfish found in my moves, a bit over three per game. This page is my attempt to learn how to avoid blundering in chess from my own worst week of blundering, starting with the most painful draw I've had: a stalemate with a queen up.
The week in numbers
| Games | 29 (24 at 15 min + 10 s, 5 at 10 min) |
| Wins / losses / draws | 11 / 17 / 1 |
| Rating | 975 → 956 (−19) |
| Average accuracy | 76.4 % |
| Blunders (Stockfish) | 91 (my opponents: 81) |
| Most played openings | Horwitz Defense (3), Indian Game (3), Zukertort Chigorin (3), Queen's Pawn (3), Accelerated London (3) |
My opponents were on average rated 996, so this was not a week of strong opposition. With White I went 6–9, with Black 5–8–1. Monday, June 3 alone was ten games (4 wins, 5 losses, 1 draw). Of my 134 mistakes and blunders, 75 came in the middlegame, 31 in the opening and 28 in the endgame, my worst endgame count in weeks. And there were nine games (#495, #496, #498, #499, #502, #505, #506, #509 and #513) where I had at least an 80 % chance of winning and didn't win.
Two of those deserve a sentence each. In #513 against 90stx I was +8.71 at move 44 and played 44.Qf4, attacking a rook; the queen could simply be taken, and two moves later another blunder turned it into a lost position. And in #495 against Rugbyyyy92 I reached a queen against a bare king. Stockfish saw mate in three, starting with 68...Qb4. I played 68...Kc6 instead, and White's king on a6 had no legal move while not being in check. That is stalemate, and stalemate is a draw.
Three blunders, three kinds
The double attack I didn't look for
The same game #495, much earlier. I was Black in the Center Game and already winning, -6.31, with my bishop on c5 aimed at f2.
8...d5 opens the centre and isn't terrible: the evaluation stayed clearly in my favour at -2.02. But the engine's 8...Qf6 would have kept the whole advantage. From f6 the queen attacks the rook on a1 along the long diagonal and the f2 pawn next to White's king, and nothing defends either. That's a double attack: one move, two threats, and the opponent can only answer one. I didn't look for it because I didn't ask what my queen could attack. The game ended in that stalemate, 60 moves later (game on Chess.com).
The mate threat with a hole in it
Game #496, White against StrawHat_Mike47 (1004), a 10-minute game. I was clearly better, +2.18.
13.Bh6 looks wonderful. It threatens Qxg7 checkmate, and it also attacks the g7 pawn with the rook on f8 behind it. The problem is that Black has one quiet defence: 13...Qf6 covers g7 and attacks my bishop on h6, which has no defender. After that, Stockfish rates the game about equal (+2.18 became +0.14). The calmer 13.Bd6 attacks the rook on f8 and keeps my edge. A threat of mate is only as good as the opponent's worst defence; I didn't check their best one (game on Chess.com).
Grabbing a pawn while a knight hangs
Game #497, Black against marivet88 (997), a 15+10 game. White was a little better, +0.93, and my knight on h5 was attacked by the bishop on e2 along the open diagonal.
13...Qb6 goes after the b2 pawn, which has no defender. Meanwhile White just takes my knight: 14.Bxh5, and the evaluation jumps from +0.93 to +6.46. Stockfish's 13...Ng3 saves the knight and attacks the bishop that was attacking it. This kind of blunder is the easiest one to prevent: I was counting my opponent's loose pawns and forgot to count my own loose pieces (game on Chess.com).
The lesson of the week
Every blunder above comes from looking at only one side of the board. In #497 I saw what I could take, not what I had left hanging. In #496 I saw my threat, not their defence. In #495 I didn't look for my queen's attacks at all, and 60 moves later I didn't check whether my king move left the opponent any legal moves. So my practical rule for how to avoid blundering in chess is a two-part blunder check before every move: what is attacked right now, on both sides? And after my move, what is their best answer? It costs a few seconds. This week, with 91 blunders, I clearly wasn't spending them.
All games this week
| Game # | Date | Colour | Opponent (rating) | Result | Opening | Accuracy | Link |
|---|---|---|---|---|---|---|---|
| 494 | Jun 3 | White | Mick210869 (959) | Win | Queens Pawn Opening Horwitz Defense | 96.9 | game |
| 495 | Jun 3 | Black | Rugbyyyy92 (963) | Draw | Center Game Accepted Paulsen Attack | 69.0 | game |
| 496 | Jun 3 | White | StrawHat_Mike47 (1004) | Loss | Indian Game | 65.4 | game |
| 497 | Jun 3 | Black | marivet88 (997) | Loss | Queens Pawn Opening Chigorin Variation | 80.1 | game |
| 498 | Jun 3 | White | Victorlutte (1002) | Loss | Queens Pawn Opening Horwitz Defense | 67.2 | game |
| 499 | Jun 3 | Black | Bobthebuilder222333 (993) | Loss | Giuoco Piano Game | 59.2 | game |
| 500 | Jun 3 | White | 0yyyzu (1015) | Win | Indian Game | 93.4 | game |
| 501 | Jun 3 | White | alonsomorris (987) | Win | Queens Pawn Opening Zukertort Chigorin Variation | 80.0 | game |
| 502 | Jun 3 | Black | Paul197719 (1063) | Loss | Scotch Game Goring Double Pawn Sacrifice Variation | 63.2 | game |
| 503 | Jun 3 | Black | Nicololucidi97 (926) | Win | Vienna Game Max Lange Defense | 69.3 | game |
| 504 | Jun 4 | White | anishpilot (998) | Loss | Indian Game | 76.8 | game |
| 505 | Jun 4 | Black | the_mask0007 (955) | Loss | Scotch Game Scotch Gambit Haxo Sarratt Variation | 60.5 | game |
| 506 | Jun 4 | Black | ConfusedMatrix (1022) | Loss | Giuoco Piano Game | 51.0 | game |
| 507 | Jun 4 | White | thiago07dantas (1010) | Loss | Queens Pawn Opening Zukertort Chigorin Variation | 77.9 | game |
| 508 | Jun 5 | White | ACOL42 (1008) | Loss | Queens Pawn Opening | 92.8 | game |
| 509 | Jun 5 | Black | kathmandu15 (938) | Loss | Van t Kruijs Opening | 67.0 | game |
| 510 | Jun 5 | White | majid443 (1041) | Loss | Queens Pawn Opening | 96.4 | game |
| 511 | Jun 5 | Black | Ras-stafari (925) | Win | Queens Pawn Opening Pseudo Catalan Variation | 92.9 | game |
| 512 | Jun 5 | White | Ras-stafari (917) | Win | London System | 73.8 | game |
| 513 | Jun 6 | White | 90stx (1035) | Loss | Queens Pawn Opening Horwitz Defense | 71.9 | game |
| 514 | Jun 6 | White | Avazbek1999 (1093) | Win | London System | 72.7 | game |
| 515 | Jun 6 | Black | oceano929 (1087) | Loss | Grob Opening | 70.4 | game |
| 516 | Jun 7 | Black | Edo198426 (1011) | Win | Kings Fianchetto Opening | 80.5 | game |
| 517 | Jun 7 | Black | SulemanShani02 (972) | Win | Unclassified | — | game |
| 518 | Jun 7 | White | mehdismh (987) | Loss | Queens Pawn Opening Accelerated London System | 81.2 | game |
| 519 | Jun 7 | Black | manumanu321 (1082) | Loss | Queens Pawn Opening Accelerated London System | 79.8 | game |
| 520 | Jun 8 | White | Sagar_Makhija_too (934) | Win | Queens Pawn Opening Accelerated London System | 95.7 | game |
| 521 | Jun 8 | White | mohamedAbadr (1030) | Loss | Queens Pawn Opening Zukertort Chigorin Variation | 91.3 | game |
| 522 | Jun 8 | Black | kalleb307 (944) | Win | Queens Pawn Opening | 63.2 | game |
Previous: week 21 · Next: week 23 · Hub: my chess journey · Have a week like this of your own? Review your own games like this.
Frequently asked questions
How do you avoid blundering in chess?
My week points to one check before each move: which of my pieces are attacked right now, and what does my move leave undefended? Many of my 91 blunders were pieces left where they could be taken, like the knight on h5 in game #497.
How many blunders does a 1000-rated player make per game?
For me, about three. Stockfish counted 91 blunders in my 29 games this week, and 81 in my opponents' moves.
What is stalemate in chess?
A position where the side to move has no legal move and is not in check. It is a draw. In game #495 I had a queen against a lone king, Stockfish saw mate in 3, and my 68...Kc6 left White's king with no legal move.