Week 93: a forced mate lost on time in rapid chess

Week 93 (November 3–9, 2025): 18 rapid games, 4 wins, 13 losses and 1 draw, with my rating moving from 1301 to 1229. In nine of the fourteen games I did not win, the engine's winning-chance estimate exceeded 80% at some point. A forced mate lost on time in rapid chess is the most striking case, with two other decisions below.

A forced mate lost on time in rapid chess: the week in numbers

The week had a six-game losing streak, games 1799 to 1804. Of my 41 blunders, 30 were in the middlegame, 8 in the opening and 3 in the endgame, and 10 were played with a low clock, all of them in a single game I'll get to below. As White I went 2–7; my two Englund Gambit games (1.d4 e5) were both losses, one of them a checkmate on move 13.

Three ways my advantages disappeared that week

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

One: answering a threat passively. Against mantizreligiosa (1291) my queen had gone on a pawn hunt, 9.Qxb7 and 10.Qxc6, and Black's queen had taken my b2 pawn in return. I was winning (+4.77). Black's queen on b2 now attacked my rook on a1, and I protected the rook with 11.Nb3. Stockfish's 11.Rb1 moves the rook and attacks the queen at the same time; the queen had no defender, so Black has to spend a move running. After my move, the engine's line for Black starts with a bishop attacking my queen on c6 and later ends with ...Qxf2 check. The evaluation fell to -0.61. My next move, 12.Bd3, was a blunder too, and the game ended in a loss on move 19.

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

Two: taking material with the king in the centre. Against lordfury1995 (1282), in an Englund Gambit, I was clearly winning (+5.78). 15.Qxd5 looks like the dream move: it takes a pawn and attacks three things at once, the b5 pawn, the bishop on f5 and the knight on c6. But my king was still on e1, and the answer was 15...Qa5+, a check on the diagonal. My king had to walk to e2, and the evaluation went down to +2.73. Still good for me, and I still lost the game later. Stockfish's move was 15.O-O, castling first and keeping the pawn grabs for later. Stockfish says my winning chance in this game reached 96.8%.

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

Three: a quiet move when the position needed a sharp one. As Black against Ishaan_Datta (1285), in the Pirc's Geller System, White's 11.O-O was a blunder and I was clearly better (-2.85). I played 11...a6, a useful-looking pawn move. Stockfish's 11...Nd4 attacks the knight on f3, a move after White castled. In its line, my other knight then takes the pawn on h3 with check, and my bishop takes on h3 after the recapture, attacking the rook on f1. My quiet move let White reorganise: the evaluation went to -0.09, and I lost the game on move 13.

Game 1799: a forced mate, lost on time

The worst one didn't make the key positions because it's a whole game rather than a single move. Against mbqwq (1288), in a London System, I made 14 blunders, 33 of my 66 moves were played with a low clock, and the evaluation swung back and forth: from +3.77 to -4.03 with 21 seconds left on move 51, then to a forced mate against me on move 54, then back again. From move 60 to move 66, after every single move I played, Stockfish shows a forced mate for me: mate in 5, mate in 4, mate in 5, mate in 7, mate in 4, mate in 5, mate in 6. My clock after those moves read 4, 3, 3, 2, 1, 1 and 0 seconds. I lost on time.

There's also the draw, game 1798 against maherchugh (1314), where I was at -10.16 on move 29 (completely winning as Black) and played 29...c5. The evaluation went to 0.00 and the game ended in a draw by repetition.

The lesson of the week

Losing winning positions isn't one problem; it's three. When I'm winning, the job changes. I stop needing good moves and start needing safe ones: answer threats with tempo (Rb1, not Nb3), get the king out of the centre before collecting pawns (O-O, not Qxd5), and keep enough time to finish. Game 1799 shows the last one brutally: a won game is worth nothing with zero seconds on the clock. My rule from this week: once I'm clearly winning, I look at my opponent's checks and threats before my own captures, and I play faster, not slower.

All games this week

Previous week: week 92. Next week: week 94. The complete diary is on the journey page. Winning chances, clock times and evaluations all come from the same Stockfish review you get when you review your own games like this.

For a general method beyond this week's games, see the guide to reviewing lost winning positions.