0
← Back to news

tournaments

Abdusattorov Punishes a Loose Bishop at the Samarkand Olympiad

6 min read• Source: ChessBotBuddies EditorialChess Olympiad 2026, Samarkand Olympiad, FIDE Olympiad, Nodirbek Abdusattorov, Uzbekistan, Tactics, Loose Pieces

Here is a puzzle. Look at the pawn on e4 in the board below. Two black pieces attack it. Two white pieces defend it.

Is it safe?

8
br
br
bk
7
bp
bq
bb
bp
bp
bp6
bb
bp
bp
bn
bp
5
4
wp
wp
wb
wp3
wn
wq
2
wp
wp
wn
wp
wp
1a
b
wkc
wrd
e
f
wrg
h

White to play on move 17. The pawn on e4 is attacked by Black's bishop on c6 and knight on f6, and defended by White's knight on c3 and queen on d3.

Many players would count two attackers against two defenders, decide the pawn is fine, and move on. Richard Rapport of Hungary pushed a different pawn instead. It cost him the e4 pawn, and the game went Black's way from there.

Richard Rapport 0-1 Nodirbek Abdusattorov 46th FIDE Chess Olympiad, Samarkand 2026 · Open Section, Round 5 · 20 September 2026 White: Rapport, Hungary (2726) · Black: Abdusattorov, Uzbekistan (2762) · Opening: Sicilian Defence, Chekhover Variation

Abdusattorov was playing for Uzbekistan, the country hosting the Olympiad, and his rating was 36 points higher. That is a small gap between two elite players.

Two kings on opposite sides

The game began with a Sicilian Defence, which starts with 1...c5.

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Qxd4

This particular version is called the Chekhover Variation. White recaptures on d4 with the queen, bringing the most powerful piece into the game very early.

4...Nf6 5.Bg5 Nc6 6.Bb5 Bd7 7.Qd3 e6 8.Nc3 h6 9.Bh4 a6 10.Bxc6 Bxc6 11.O-O-O Be7 12.Nd4 Bd7 13.f4 O-O

Now look at where the kings went. White castled on move 11 with O-O-O, which puts the king on the queenside, on c1. Black castled on move 13 with O-O, which puts the king on the kingside, on g8.

br8
bq
bk
br
7
bp
bb
bb
bp
bp
bp6
bp
bp
bn
bp
5
4
wn
wp
wp
wb
3
wn
wq
wp2
wp
wp
wp
wp
1a
b
wkc
wrd
e
f
g
wrh

Position after 13.f4, just before Black castles. White's king is already on c1. After 13...O-O, Black's king will be on g8, on the opposite side of the board.

When the two kings live on opposite sides, the game turns into a race. Each player pushes pawns toward the enemy king to open lines for rooks and queens. Whoever gets there first usually wins, and every slow move is a gift to the other side.

How the engine sees it

To see who was winning, we ran Stockfish 17.1, a strong chess engine, on the key positions of this game. It looked about eleven moves ahead, which is a search depth of 22. The engine gives a score in pawns. A score of +1.0 means White is ahead by about one pawn, and -1.0 means Black is ahead by about one pawn.

  • After 13...O-O: White ahead by about 0.7
  • After 15.Nde2: White ahead by about 0.1
  • After 16...Qc7: Black ahead by about 0.2
  • After 17.g4: Black ahead by about 0.9
  • After 18.Qh3: Black ahead by about 1.5
  • At the end of the game: Black ahead by more than 3

Nothing dramatic happened on moves 14, 15 and 16. No piece was lost. But White's small advantage disappeared anyway.

14.a3 Rc8 15.Nde2 Bc6 16.Rhg1 Qc7

White spent moves on a3 and on pulling the knight back to e2, and the engine preferred 15.Rhg1, bringing a rook toward the pawn push that was coming. Meanwhile Black used the same moves to play ...Rc8, ...Bc6 and ...Qc7. Look at the c-file, the column of squares that runs up the board from c1. Black now has a rook, a bishop and a queen lined up on it, all pointing down at White's king.

That is how a slow slide works. Each slow move is small. Together they hand the other side the lead in the race.

Move 17: the pawn that looked safe

Back to our puzzle. It is White's turn, and Rapport pushed the g-pawn.

17.g4 Bxe4

Abdusattorov found the engine's top reply. The pawn on e4 was attacked twice and defended twice, so it seemed to be fine. But counting attackers and defenders is only the first step. The second step is to ask what happens after the trades.

Suppose White takes the bishop and Black recaptures. This is the line that White avoided:

18.Nxe4 Nxe4

8
br
br
bk
7
bp
bq
bb
bp
bp
bp6
bp
bp
bp
5
4
bn
wp
wp
wb
wp3
wq
2
wp
wp
wn
wp
1a
b
wkc
wrd
e
f
wrg
h

The line White avoided: 17...Bxe4 18.Nxe4 Nxe4, with White to move. White's bishop on h4 has no defender, and Black's bishop on e7 now attacks it.

Look at the bishop on h4. Nothing protects it. A piece with no defender is called a loose piece, and loose pieces are the first thing a tactician looks for.

Why is Black's bishop on e7 attacking it now? Because the black knight used to stand on f6, right in the way. When that knight jumped to e4 to recapture, it cleared the diagonal from e7 to h4.

There is more. From e4, the black knight also threatens to jump to f2, where it would attack the queen on d3 and the rook on d1 at the same time. That is called a fork.

The engine's best line for White from here is 19.Bxe7 Nf2 20.Qb3 Qxe7, and Black comes out a pawn ahead. The other tries score even worse for White.

So the pawn on e4 was not really safe. Two defenders could not fix a problem that the count of defenders never showed.

What Rapport played instead

Rapport did not go into that line. He chose:

18.Qh3 d5

8
br
br
bk
7
bp
bq
bb
bp
bp
bp6
bp
bn
bp
5
bp
4
bb
wp
wp
wb
wp3
wn
wq
2
wp
wp
wn
wp
1a
b
wkc
wrd
e
f
wrg
h

Position after 18...d5. Black has won the pawn from e4, and the new pawn on d5 protects the bishop on e4.

Black had won the pawn. The move ...d5 then protected the bishop on e4 and kept the extra pawn safe. After 18.Qh3 the engine had Black ahead by about 1.5 pawns.

What White could have done

Before playing 17.g4, White could have asked a question that costs nothing: which of my pieces has no defender?

The engine's favorite move was 17.Nd4, and it also liked 17.Bxf6. Both keep the score close to -0.2. The second one is easy to understand.

17.Bxf6 Bxf6

8
br
br
bk
7
bp
bq
bp
bp
bp6
bb
bp
bp
bb
bp
5
4
wp
wp
wp3
wn
wq
2
wp
wp
wn
wp
wp
1a
b
wkc
wrd
e
f
wrg
h

Position after 17.Bxf6 Bxf6. White's bishop has traded itself for Black's knight, so there is no loose piece on h4 for Black to attack.

White's bishop takes the knight before the knight gets a chance to jump away. The loose piece is gone, and so is the fork trick.

The finish

Black kept the extra pawn, and in our engine check White never got ahead again. The game record ends after 31...Qc5 with the result 0-1, a win for Abdusattorov, and the engine had Black ahead by more than three pawns.

8
br
7
bp
br
bp
bk
bp6
bp
5
bq
4
bb
wn
bn
wp3
wq
2
wp
wp
wn
1a
b
wkc
d
e
wrf
wrg
h

Position after 31...Qc5, where the game record ends. Black is more than three pawns ahead by the engine's count.

What this game teaches

1. Counting attackers and defenders is step one, not the whole job. The pawn on e4 had two of each. What decided the fight was what each trade left behind.

2. Loose pieces are targets. A piece with no defender is one accident away from being lost. Before every pawn push, scan your own position and ask which pieces are undefended.

3. When the kings are on opposite sides, slow moves cost a lot. White's a3 and Nde2 looked harmless. While Black's rook, bishop and queen arrived on the c-file, the engine's score slid from a small White edge to a small Black one.

4. Three pieces on one file point at a king even when nothing seems to be happening. Look for these lines of pieces. They are quiet, and they are dangerous.

The next time you are sure a pawn is safe, count the attackers and defenders, and then count one more thing: everything that would change once the trades start.

Related Stories

© 2026 ChessBotBuddies.orgPlayAbout UsPrivacy PolicyTerms of ServiceContact