Chess Olympiad 2026, round 7: what the opponent dossiers predicted and what the games decided

Short answer

On the eve of round 7 at the 46th Chess Olympiad, we published comprehensive dossiers on the opponents facing Brazil, the USA, and Canada, breaking down opening repertoires, variations where each player historically scores poorly, and recurring tactical error profiles. Now comes the moment of truth: testing that opening preparation against the 24 games played on 23 September 2026. The accuracy in the opening was remarkable, with the first move anticipated in 17 encounters and the target variations reaching the board in four games. Yet when time trouble strikes and concrete calculation takes over, prep stays on the sidelines. All 14 decisive games of the round were settled in the middlegame and in technical rook endgames, where clock management and over-the-board technique decide the day.

Event: 46th FIDE Chess Olympiad

How this was compiled: Researched on 24 September 2026, with round 7 finished. The 24 games played by Brazil, the USA and Canada were read from the official PGN broadcast on Lichess, and the results checked on chess-results. The dossiers compared are the three published on 21 September, before the round, and they were not changed afterwards. Every evaluation quoted was measured with Stockfish 19 at depth 24, with a dedicated search and a clean hash, rather than copied from the broadcast's automatic annotation; the endgames of Luong Phuong Hanh v Julia Alboredo and Wesley So v Robert Hovhannisyan were checked in the Lichess tablebase. The clock time quoted is what the player who moved had left right after the move, according to the PGN.

The essentials in a few lines

• An eventful round 7 on 23 September 2026: in the open section, Brazil and Canada ground out 2.5 to 1.5 match victories against Tajikistan and New Zealand, while the USA tied 2-2 with a tough Armenian squad. In the women's event, the American team defeated Switzerland 3-1, Canada drew 2-2 with Hungary, and Brazil fell 1-3 to Vietnam.
• First moves in the opening are highly predictable: in 17 of 24 games, opponents stuck to their primary opening weapon. Three chose secondary options already mapped in our database, and only four threw complete opening surprises: Sravan Renjith, Tom Middelburg, Alexandra Kosteniuk, and Pham Le Thao Nguyen.
• Reaching an opponent's statistical weak spot guarantees no easy point: variations where opponents historically struggled appeared on four boards (against Muhammad Khusenkhojaev, Nguyen Thi Mai Hung, Felix Xie, and Kata Karacsonyi), resulting in three draws and one loss for the side targeting them.
• Middlegames and endgames decide the match points: in the decisive games, the turning point occurred between moves 13 and 54. Three games collapsed with under three minutes left on the clock, and GM Andre Diamant watched a winning edge of +2.77 slip away four times against Mekhriddin Sharifov in severe time trouble.
• Opening prep gets you safely to the middlegame, but conversion is entirely on you: thorough theoretical scouting delivers a sound position and saves clock time, but winning demands sharp calculation, endgame technique, and cool nerves under pressure.

The round 7 scoreboard

On the eve of round 7, we put together a thorough scouting report: three comprehensive dossiers profiling all ten players on each opposing squad. We mapped out opening repertoires by colour from classical games on Lichess, flagged variations where opponents score poorly, and compiled technical accuracy profiles with the engine. Since official team line-ups are announced only minutes before the clocks start, we had to prepare for full squads, and four out of every five players we mapped actually sat down to play.

Every Grandmaster knows that checking opening prep against actual over-the-board results is the most valuable habit in chess. To be clear, we make no claim that any team used our reports. Our exercise is practical: we wanted to see how far opening statistics can take a player at this level, and where pre-game preparation hands over to pure calculation and clock management.

The biggest shock for the US team came on Wesley So's board. Playing White, the American super GM lost to Robert Hovhannisyan in an instructive rook endgame where the tablebase confirms a forced win for Black right after 54.Rxd3. Awonder Liang salvaged the match tie for the Americans with a smooth win over Gabriel Sargissian. In the women's tournament, Brazil's standout performance came from IM Juliana Sayumi Terao, who pressed her White pieces to defeat Bach Ngoc Thuy Duong.

The first move: 17 out of 24

There is a common misconception among chess players that opening preparation aims to predict an entire game until checkmate. In master chess, the real purpose of prep is to anticipate the first move and secure a comfortable pawn structure while conserving valuable clock time. Mustafokhuja Khusenkhojaev met 1.e4 with 1...e5, matching 21 of 22 games in his sample. Robert Hovhannisyan did the same, mirroring 26 of 27 games. And Muhammad Khusenkhojaev opened with 1.d4, staying true to 39 of 43 White games. Players preparing against them walked straight into the exact structures studied the night before.

The three secondary choices were also mapped with clear game counts. Haik Martirosyan opened 1.e4, played in 23 of 54 White games. Gabriel Sargissian met 1.e4 with the solid Caro-Kann Defence, used in 4 of 22 games. And Zsoka Gaal chose the English Opening with 1.c4, seen in 6 of 46 games. Facing a secondary opening with few games on record tells a grandmaster that the move is in the opponent's locker, but without enough volume to hide deep theoretical novelties.

The four complete departures illustrate the natural limits of statistical scouting. Sravan Renjith opened 1.c4 despite zero recent classical games in that line out of 20. Tom Middelburg, who played the Caro-Kann in all 11 games against 1.e4 in his sample, unexpectedly chose 1...e6. Former world champion Alexandra Kosteniuk met 1.e4 with 1...e5, a move absent from her 5 recent games against King's Pawn openings. And Pham Le Thao Nguyen, who favored 1.d4 in 19 of 20 games, switched to 1.Nf3. Interestingly, the seven players who deviated from their main weapon scored 2.5 out of 7, while the 17 who stayed true to their habits scored 9 out of 17. The sample is too small to draw sweeping conclusions, but it reminds us that uncharacteristic opening gambits carry their own risks.

The weak lines that reached the board

Librelato's game provides a textbook lesson on the practical limits of opening prep. The Brazilian dossier noted that Nguyen Thi Mai Hung scored a modest 38.9% in 9 games against 1.e4, playing the French Defence almost exclusively, and highlighted the classical line with 3.Nc3 as a low-scoring option for Black: 25.0% across 4 games. The game followed the theoretical sample for 12 half-moves: 1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.Bg5 dxe4 5.Nxe4 Nbd7 6.Nf3 h6, reaching a position where Nguyen was 0 for 2 in the database. Brazil achieved the exact opening position they aimed for, but the struggle continued and Black ultimately won on move 58. A dossier delivers a sound position to your board, but converting middlegames and endgames is entirely down to the player holding the pieces.

Against Muhammad Khusenkhojaev, who held a 38.9% score in 9 games after 1.d4 Nf6 2.c4 e6, Diego Di Berardino went into the line, but Black deviated on move four with 4...Nbd7. The game became an endurance test lasting 131 moves. The diagram captures the peak tension: four moves in a row where the engine evaluation swung between +0.05 and +4.09 without either player finding the finishing blow.

In the other two games, the target lines appeared without deciding the outcome. Golubeva opened 1.Nf3, a move against which Karacsonyi scores 25.0% in 10 games, but departed from theory on move two with 2.b3 and drew comfortably. Felix Xie, who scores 45.0% in 10 games after 1.d4 Nf6 2.c4 e6, accepted the theoretical challenge, varied with 3...d5, and also held the draw. As every GM emphasizes: a low-scoring line merely shows where an opponent had trouble in the past, never an automatic point.

Who ran from the dossier

The Canadian dossier clearly highlighted that Middelburg played the Caro-Kann in all 11 of his games against 1.e4, scoring a modest 36.4%. GM Razvan Preotu played 1.e4 and was met by the surprising 1...e6. The game entered an Advance French with 5...Qb6 6.Be2 Nh6, and Black's position collapsed rapidly despite plenty of time on the clock: 13...Be7 pushed White's edge from +1.37 to +2.64, and 14...Bg5 lifted it to +3.80, both moves played with nearly an hour left on Middelburg's clock. Dodging an opponent's preparation by playing unfamiliar opening structures often leads to far greater trouble than the line you feared.

Hovhannisyan's decision was a masterclass in practical match strategy. The US dossier flagged the Ruy Lopez as his most vulnerable line with Black, scoring 38.9% in 9 games. Wesley So opened 1.e4 and 2.Nf3, but the Armenian grandmaster wisely deployed the rock-solid Petroff Defence with 2...Nf6, his second most trusted setup. So tried to veer away with 4.Nc4, and the game simplified into an equal rook endgame. In mutual time trouble, 53.Kd8 ruined White's position from 0.00 to −4.78. Hovhannisyan offered a momentary lifeline with 53...Re2, but 54.Rxd3 sealed the victory in an endgame that the tablebase declares completely winning for Black.

Vietnamese master Pham Le Thao Nguyen, who favored 1.d4 in 19 of 20 games, switched to 1.Nf3 against Clara Dias, and the game turned on move 17: 17...Ne7 saw White's advantage rocket from −0.07 to +6.61. In Canada's match, Renjith tried 1.c4 against Anthony Atanasov, found himself lost after 30.h4 (dropping from 0.00 to −3.39), but regained the upper hand when 31...Qf7 swung the evaluation back to +2.07. In both games, opening choices were just the prelude to wild middlegame tactics.

Where the games were decided

The clash between Luong Phuong Hanh and Julia Alboredo is the ultimate proof that chess is won through concrete calculation. The Lichess tablebase confirms that the endgame in the diagram is an ironclad draw, but only one move maintains the balance: the defensive check 49...Rc2+. With nearly 14 minutes left on her clock, Alboredo played 49...Kh3, instantly handing White a winning position. Brazil's dossier had correctly identified Luong's weakness with Black in the Sicilian, but in this game she held the White pieces.

GM Andre Diamant endured Brazil's most agonizing game of the round. Facing Mekhriddin Sharifov, an opponent with no flagged opening liabilities, Diamant established a winning advantage of +2.77 or higher at four separate moments. But under severe time pressure, 31.Bg4 let the advantage slip from +3.49 to +0.36 with 2 minutes 52 seconds on the clock. Soon after, 33.Rf4 dropped from +3.28 to +1.76 with 36 seconds, 37.Ng5 fell from +3.41 to 0.00 with 34 seconds, and 53.Ra8 returned +2.77 to equality with 33 seconds. A painful draw. In the women's event, Kathie Librelato also lost control in two moves under heavy time pressure, as shown in the second diagram.

The clock punished both sides equally. Bach Ngoc Thuy Duong played the losing blunder 40...Bg6 with just 1 minute 29 seconds left, allowing the evaluation to soar from +0.28 to +6.72 in Juliana Sayumi Terao's favor. In the open event, Gabriel Sargissian stumbled with 34...Rb2 with 2 minutes 9 seconds, sending Awonder Liang's edge from +1.18 to +4.19. The table below collects these decisive moments where the evaluation shifted permanently.

What Personal Chess Coach would do, and what it would not

Round 7 offers a clear lesson about the division of labor in competitive chess. A dossier is ideal for mapping predictable patterns: the first move in the opening, favourite setups against 1.e4, and variations where an opponent historically drops rating points. But it cannot predict when a player will suddenly abandon their trusted opening, and it has nothing to offer when facing a sole defensive move on move 49 of a rook endgame. Even when opening preparation hits the target, as it did in Librelato's French, prep hands you a fine position on move 6, but there are a hundred moves left before the handshake.

What truly settled the round was the practical mastery every tournament player must develop: converting advantages under time pressure, defending complex endgames, and staying alert when an opponent makes an error. That is precisely what Personal Chess Coach trains in your own games. The engine reviews every move, categorizes tactical mistakes and blunders, groups repeating patterns by opening, and generates exercises tailored to your own shortcomings. The clock times cited in this analysis come directly from the official live broadcast records.

Grandmaster perspective: scouting an opponent's opening repertoire and training your own technical weaknesses answer two distinct needs, and round 7 tested both on the board. The pre-game opening homework proved accurate on 17 of the 24 boards. But it was the tension and calculation at the board that ultimately decided the 14 games that found a winner.

Frequently asked questions

Update log

• 2026-09-24 — First published with round 7 finished: the 24 games played by Brazil, the USA and Canada compared with the dossiers published on 21 September, and every evaluation quoted measured with Stockfish 19 at depth 24.
• 2026-09-29 — The Olympiad ended on 27 September: Uzbekistan won the open and China the women's event. This coverage is no longer updated; the final result is in the standings story.

Sources

• 46th Chess Olympiad Samarkand 2026, Open — round 7 results — Chess-Results
• 46th Olympiad, open, matches 1–12 — round 7 broadcast (USA v Armenia) — Lichess
• 46th Olympiad, open, matches 13–37 — round 7 broadcast (Brazil and Canada) — Lichess
• 46th Olympiad, women, matches 1–25 — round 7 broadcast — Lichess

Related guides

• How to build an opening repertoire: 1.c4 as White and the Sicilian as Black
• How to avoid time trouble in chess: five rules for the last three minutes
• Recurring chess mistakes: how to find the pattern that repeats across your games