Week 85: how to get better at chess endgames, after a stalemate I didn't earn

By · Updated

Week 85 (September 9 to 14, 2025) took me from 1229 to 1252: 10 rapid games, 6 wins, 3 losses and 1 draw. As White I won all four games. As Black I lost three and drew one, and that draw is the reason this week is about how to get better at chess endgames. I had a winning position, lost the thread in the ending with seconds on my clock, reached a position that was completely lost, and then my opponent stalemated my king. Half a point I didn't deserve.

The week in numbers

What This week
Rated rapid games 10
Score (W–L–D) 6–3–1
Rating change +23 (1229 to 1252)
Average accuracy (Stockfish) 77.4%
My blunders 30
Openings I played most Nimzowitsch Defense Declined Williams Variation (3), Queen's Pawn Opening Zukertort Chigorin Variation (2), Van 't Kruijs Opening (1)

Most of my 30 blunders were in the middlegame (23), as usual. Two came in the opening and five in the endgame, and all five endgame blunders came in the same game. That game, number 1669 against KingCarlos99 (1200), deserves its own section before the key positions.

The stalemate: from winning to lost to drawn

In game 1669 Stockfish says my best winning chance at some point was 94.2%. On move 41 I was still clearly better (-3.95; minus means good for Black, me), with 75 seconds left. From there my endgame moves went like this, according to the engine: 41...g5 cut the advantage to -1.77; 47...Kd8, with 42 seconds left, turned it into +1.91 for White; 48...Kxc7 into +4.97; 50...Rxc6, with 37 seconds, threw away a position the engine had rated about equal; and 55...h6, with 34 seconds, made it +6.64. My accuracy in the endgame phase of that game was 70.2%, and 18 of my moves were played with a low clock.

By move 60 White was completely winning (+17.53 after my 60th move). Then came 61.f7+ and 62.Kf6. My king was on f8. It couldn't go to e8 or g8, because the pawn on f7 covered both. It couldn't go to e7 or g7, because White's king covered both. It couldn't take the pawn on f7, because the king on f6 protected it. My only other piece, the pawn on h5, was blocked by White's pawn on h4. I wasn't in check and had no legal move: stalemate, a draw. My opponent had a winning king-and-pawn ending and forgot that the losing side needs at least one legal move.

I'm not proud of that draw. But it's the most useful game of the week, because it shows both halves of endgame technique failing: I couldn't convert, and my opponent couldn't either.

Three positions where the trouble started earlier

r . . q . r k .
p p p . p p b p
. . . p . . p .
. . . P n . B n
. . . . P . . .
. B N . . . . .
P P P . Q P P P
R . . . . R K .
Move 11 against KingCarlos99: kick the bishop, or reroute the knight?
Pawn from f7 to f6, attacking the undefended bishop on g5 Knight from e5 to d7, opening the line of the bishop on g7 towards the knight on c3

Same game, much earlier. In this Modern Defense, KingCarlos99 (1200) had a bishop on g5 and I kicked it with 11...f6. Stockfish called it a blunder (+0.78 to +3.32). The pawn on f6 also stands right in front of my own bishop on g7. In the engine's line after 11...f6 my bishop even goes back to h8, and White's f-pawn comes to f4 to attack my knight on e5. The engine's 11...Nd7 does the opposite: it moves the knight out of the way and lets the bishop on g7 look all the way down to the knight on c3.

r . . . r . k .
p p p . . p . p
. . . p . . p .
. . . . . . q .
. . . Q . . . .
P b . . . . P P
. . . . N P B .
R . . . . K . R
Move 18 against tarikadkhil: two ways to attack the same knight
Queen from g5 to e7, attacking the knight on e2 Queen from g5 to b5, attacking the undefended knight on e2 and pinning it to the king on f1

Against tarikadkhil (1328), the highest-rated opponent of the week, the position was roughly balanced (-0.52). Both queen moves attack the knight on e2. Only one of them also lines up with White's king: from b5, the queen hits the knight and, behind it, the king on f1, so the knight is pinned and has no defender. From e7 it's just an attack, and White answered 19.Nc3. My next move was a mistake too, and I was checkmated later.

r . . q . r k .
p . n . p p b p
. . B p . n p .
. . p P . . B .
. . . . P . . .
. . . . . P . .
P P P Q N . P P
R . . . K . . R
Move 12 against ArturSvoboddin: my rook on a8 was on the bishop's diagonal
Knight from c7 to a6 Rook from a8 to b8, attacking the undefended pawn on b2

In the Pirc, Kholmov System, against ArturSvoboddin (1246), White's bishop on c6 was looking straight at my rook on a8. I moved my knight instead, 12...Na6, and the evaluation went from +2.17 to +6.47. Stockfish's 12...Rb8 takes the rook off that diagonal and attacks the b2 pawn. Two moves later the bishop took my rook, and the game was over in 14 moves.

The lesson of the week

For me, getting better at endgames starts with the clock, not with books. My five endgame blunders all came with 75 seconds or less. In a 10-minute game I keep spending my time in the middlegame and arriving at the ending with nothing left. The second lesson comes from my opponent: when you're winning an ending, before every move ask "does the other side still have a legal move?" Stalemate is a draw, and a king boxed in by a pawn and a king is exactly how it happens.

All games this week

# Date Colour Opponent Result Opening Accuracy
1668 2025-09-09 Black tarikadkhil (1328) Loss Van 't Kruijs Opening 79.5% game
1669 2025-09-09 Black KingCarlos99 (1200) Draw Modern Defense with 1 e4 77.5% game
1670 2025-09-10 White salohcc (1201) Win Queen's Pawn Opening Zukertort Chigorin Variation 89.4% game
1671 2025-09-10 Black ArturSvoboddin (1246) Loss Pirc Defense Main Line Kholmov System 80.5% game
1672 2025-09-10 White Erik_kuuns (1209) Win Queen's Pawn Opening 86.5% game
1673 2025-09-11 White Milka10lipiec (1228) Win Queen's Pawn Opening Zukertort Variation 88.1% game
1674 2025-09-11 Black Laureano040779 (1208) Win Nimzowitsch Defense Declined Williams Variation 63.8% game
1675 2025-09-13 Black GladeEye (1263) Loss Nimzowitsch Defense Declined Williams Variation 74.4% game
1676 2025-09-14 White La3ert (1230) Win Queen's Pawn Opening Zukertort Chigorin Variation 55.8% game
1677 2025-09-14 Black Clodomir62 (1214) Win Nimzowitsch Defense Declined Williams Variation 78.8% game

Previous week: week 84. Next week: week 86. The rest of the series is on the journey page. The clock data and phase accuracy here come from the same review Chess Coach runs when you review your own games like this.

Frequently asked questions

How do I get better at chess endgames?

Start by checking where your endgame mistakes come from. All five of my endgame blunders this week came in one game, game 1669, and all of them with 75 seconds or less on my clock. For me the first fix is keeping time for the ending, not learning more theory.

What is stalemate in chess?

Stalemate is when the side to move is not in check but has no legal move. The game is a draw. In game 1669 my king on f8 had no legal square and my only pawn, on h5, was blocked, so a position the engine rated +17 for White became a draw.

Did I win more as White or Black this week?

As White I won all four games. As Black I won 2, lost 3 and drew 1.