Chess Olympiad 2026, round 8: how Personal Chess Coach could have helped Brazil, the USA and Canada

Short answer

On 23 September 2026, the eve of round 8 of the 46th Chess Olympiad, we published three dossiers breaking down the opponents of Brazil, the USA and Canada — mapping each player's repertoire, their low-scoring lines and their error profile down to the wire. Here we set those findings against the 24 real games played on 24 September to show, board by board, where Personal Chess Coach would have made a practical difference. The opening moves were the predictable part: the rival's first choice was catalogued in 23 games, and the chess opening flagged as weak yielded 4.5 points out of 6 for the favoured side. Once the pieces clashed, however, time trouble took its toll — and the most valuable lesson of the day came from a chronic blunder pattern repeated on our own side of the board.

Event: 46th FIDE Chess Olympiad

How this was compiled: Checked on 25 September 2026, with round 8 finished. The 24 games of Brazil, the USA and Canada were read from the official PGN broadcast on Lichess, and the match scores checked on chess-results. The dossiers compared are the three published on 23 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, not copied from the broadcast's automatic annotation. 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

• Scores on 24 September 2026: in the open, Brazil 4–0 Japan, USA 3.5–0.5 Serbia and Canada 1.5–2.5 Turkey; in the women's section, Brazil 2–2 Chile, USA 1.5–2.5 Armenia and Canada 3–1 Spain.
• The opponent's first reply was nailed down in the dossier in 23 of the 24 games: 18 times as their primary choice in opening moves and 5 times as a charted secondary weapon. The lone exception was Kohei Yamada, whose 9-game sample the system had already flagged as inconclusive.
• The chess opening line marked as vulnerable reached the board in six games, and the favoured side racked up 4.5 points out of 6: clinical wins for Caruana, Liang, Fier and Ouellet, a tough draw for Abrahamyan and a single slip by Rodrigue-Lemieux.
• The error profile struck gold on Carvajal, who let another winning margin slip away, and on Otsuka, who went astray in the middlegame with almost 28 minutes to spare. It missed on Serbia, whose grandmasters cracked in level positions rather than throwing won ones.
• In 8 of the 12 games with major swings, the decisive move was played with under 5 minutes on the clock. Kathie Librelato saw her position collapse with 35 seconds left in round 8, mirroring her 36-second stumble in round 7: the classic time-trouble bug that only post-game training exposes and cures.

The round 8 scoreboard

On the eve of the round, we published three comprehensive dossiers covering all five players of each opposing team: their White and Black repertoires, the opening moves where each scores below average, and an error profile evaluated with the engine. Because official team sheets only appear on the bulletin minutes before the clocks start ticking, our sweep covered the entire squad. Of the 30 names analysed beforehand, 24 sat down across the board for battle.

Anyone who has played in an open tournament or a club championship knows the feeling: staying up all night aimlessly browsing mega databases is the textbook recipe for turning up exhausted and tumbling into time trouble. This article takes the opposite approach, with total transparency: putting the round under the microscope to show exactly where Personal Chess Coach would have given a tangible practical edge to whoever sat on our side of the board, and where no tool can work miracles. No guesswork: we compared what our pre-round scouting revealed with what the official PGN recorded on the scoresheets the next day.

And round 8 delivered a feast for chess fans. Brazil's open team steamrolled Japan without conceding even half a point, featuring commanding Black wins by Supi and Diamant and masterclass positional grinds by Fier and Di Berardino with White. The USA dispatched Serbia with quiet authority, as Awonder Liang dismantled his rival with Black in the exact line highlighted by the dossier. Canada had the most intriguing round: in the open, they tripped up on the single board where a player entered the theoretically favoured chess opening; in the women's section, they scored an emphatic win on the board where the recommendation was followed to the letter.

The first reply: 23 out of 24 in the dossier

There is an enduring myth among amateur chess players — and even in the minds of some experienced masters — that chess opening preparation is about memorising twenty moves by heart and delivering checkmate on cruise control. In real high-level chess, the goal is far more practical: sitting down at the board knowing exactly which pawn structure will arise, neutralising unpleasant surprises, and conserving precious minutes on the clock to spend in the middlegame where games are actually won and lost. On that front, the round 8 scouting was remarkably precise. Shinya Kojima opened with 1.d4, matching 100% of his 12 recorded White games. Mariam Mkrtchyan answered 1.d4 with 1...d5, echoing every single one of her 20 recorded games. Sabrina Vega Gutierrez rolled out her trademark 1.e4, confirming all 66 games in her profile. Anyone who prepped a sound structure against these opening moves was already halfway to a great result.

The secondary choices did not catch anyone off guard either, because they came backed by transparent numbers. Shou Otsuka met 1.e4 with the Sicilian Defence, a weapon he had already deployed in 7 of his 27 games against the king's pawn. Alexandr Predke did the same, repeating a choice seen in 12 of 35 games. Emre Can shelved his usual Sicilian in favour of the solid Caro-Kann, present in 3 of 33 games against 1.e4. Monserrat Morales Flores opted for 1...e5, a line appearing in 7 of 20 outings. And Lilit Mkrtchian went with 1.d4, her third-string choice, used in 11 of 47 games. Uncovering a secondary weapon during prep is pure gold: it tells you the opponent understands the broad plans, but lacks the same deep muscle memory and will almost certainly burn extra clock time trying to figure out the subtleties.

The only player who went completely off book was Kohei Yamada, and for an obvious technical reason: in the scarce Black games available in his database, he had only ever faced 1.e4. Against Brazilian GM Diego Di Berardino's 1.d4, Yamada answered 1...Nf6 2.c4 e6 3.Nf3 d5, entering a Queen's Gambit Declined without a single precedent in the sample. And here is the honest sportsmanship every serious player deserves: Personal Chess Coach's dossier refused to recommend a speculative plan against him, because our algorithm flags samples with fewer than 10 games as low-confidence rather than inventing baseless advice. Di Berardino converted cleanly in 47 moves anyway. Knowing with mathematical certainty what the data proves — and where the sample is too thin — is the cornerstone of our platform.

When the flagged line reached the board

The cleanest demonstration of the round came on board 1 of the United States match. The dossier flagged that Alexey Sarana answers 1.e4 with the Sicilian in 17 of 22 games, but that his scoring rate in the Open Sicilian with 2...d6 plummets to 44.4% across 9 outings. Super-GM Fabiano Caruana struck right at the weak link: 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6, shadowing Sarana's database for ten half-moves until the prophylactic 6.g3 nudged the Serbian GM out of his comfort zone. As the diagram reveals, Sarana lost the thread under pressure: with just 8 minutes and 56 seconds remaining, he rushed with 21...Qg6? and watched the engine evaluation surge from +0.81 to a crushing +4.28. Caruana opted for the solid 22.Be3 rather than the engine's lethal 22.Bh3, temporarily easing the bar to +0.87, but maintained complete positional dominance to wrap up the full point on move 43.

On the other side of the clash, the dossier highlighted that Velimir Ivic was scoring only 57.7% in 26 games after 1.e4 e5 2.Nf3 Nc6, well below his overall 68.1% White average. Awonder Liang took the challenge head-on: he followed 12 half-moves of Ivic's historical sample via 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Bxc6 dxc6. After 5...Be7, the Serbian's scoring rate sank to a meagre 28.6% across 7 games. The final bill came due in a tense rook endgame: with his clock ticking down to just 34 seconds, Ivic blundered with 62.Rc6+?, sending the evaluation from 0.00 to −4.57, and Liang sealed the win on move 67.

In Brazil's match, the most eye-popping stat in the scouting report was Shou Otsuka's vulnerability with Black: 25.0% in 52 games, dropping to 23.2% against 1.e4 and a disastrous 7.1% in 7 games when playing the Sicilian. And guess what he rolled out against GM Alexandr Fier? Sicilian, full throttle. Fier navigated patiently with 2.Nc3 before transposing with 3.Nf3 and 4.d4, and on move 22 Otsuka committed a typical middlegame blunder with 22...Ne3?, spiking the evaluation from +1.85 to +5.05 despite having nearly 28 minutes on his clock. Fier promptly delivered the engine's 23.Rd3 and wrapped up the game in 30 moves. In the women's section, Maili-Jade Ouellet opened 1.d4 against Marta Garcia Martin — whose score plunges from 75.0% against 1.e4 to 46.9% against 1.d4 — and cruised to victory in 47 moves. Meanwhile, Tatev Abrahamyan pushed 1.e4 against Susanna Gaboyan, who had a 36.1% record in 18 games against the king's pawn, holding a draw in a sharp tactical melee.

The counter-example came from Canada, providing a healthy reality check against blind trust in statistics. The dossier noted that Turkish teenage sensation Yagiz Kaan Erdogmus was scoring 41.7% in 6 games against the Sicilian with 2.Nf3, without a single victory. Rodrigue-Lemieux looked to test that record with a Sicilian chess opening, but after 2.Nf3 he went with 2...e6, a move not present in any of those 6 games in the database. When the clock tightened, with 4 minutes 1 second left, the Canadian blundered with 29...gxh6?, flipping a balanced +0.14 into an immediate +5.32 disaster, resigning on move 36. As any GM will remind you: six games without a win do not constitute a structural refutation; they are simply six games. In Personal Chess Coach, every number is accompanied by sample size so you never mistake a statistical quirk for universal truth.

Those who took another road

At the board, theoretical preparation is a compass, not a pair of handcuffs. A smart dossier highlights odds and historical trends, but the player pushing the pieces always owns the final move: your personal playstyle, your confidence on the day, and match strategy take precedence. WGM Oksana Golubeva, for instance, had the recommendation to meet 1.d4 with 1...Nf6 and 2...e6 against Ines Prado Acebo, a line where the Spaniard scored a modest 31.2% in 16 games. Golubeva chose instead to fight in the aggressive Dutch Defence with 1...e6 and 2...f5, scoring a 31-move win. The decisive blow came from a tactical oversight no opening book can forecast, shown in the diagram: with nearly 10 minutes on the clock, Prado Acebo blundered with 27.Be4?, letting Black's rook invade and crashing the position from +0.10 to −6.99.

In Brazil's women's match against Chile, the general blueprint called for 1...e5 against the Chilean squad and 1.d4 with White. None of the four Brazilians followed the recipe to the letter, producing mixed fortunes. The seasoned Juliana Sayumi Terao took up the fight in the Sicilian Defence against Francisca Sanchez Tobar — who had scored 62.5% against the Sicilian compared to 41.7% against 1...e5 — and showcased her class to score the win. Agatha Nunes chose the French Defence against Maria Belen Carvajal Mendoza and was ground down in a cramped setup. Clara Dias and Kathie Librelato preferred opening moves with 1.e4. Clara squared off against 1...e5, Morales's pet defence (78.6% score), and ground out a gritty win; Librelato, meanwhile, was outplayed by Rocio Contreras Suazo in a treacherous Kan Sicilian.

For the USA, legendary grandmaster Irina Krush was advised to take the solid 2...e6 route against Elina Danielian in the Queen's Gambit, where the Armenian had won 54.2% across 12 games. Krush chose to accept the gambit pawn with 2...dxc4 and succumbed on move 43, after blundering with 40...Qxa6? under extreme time pressure with just 44 seconds remaining, watching the position plummet from 0.00 to +9.14. For Canada, Yunshan Li was steered towards the Sicilian with 2...Nc6 against Sabrina Vega Gutierrez (36.7% score), but decided to test the Spaniard in her stronghold with 1...e5 (86.4% in 11 games). It led to a Scotch Opening where Li salvaged a gritty 67-move draw. The takeaway for your own journey is clear: choosing whether or not to follow the dossier is entirely in your hands, but knowing in advance the cost and risk of every option is the greatest competitive edge you can take to the board.

The weak spots the error profile pointed to

Stepping into a game knowing your opponent's Achilles' heel is worth half a point in any tournament. Carvajal entered round 8 losing an average of 50 centipawns per move in the middlegame, backed by a chronic record of evaporating leads. Against Agatha Nunes, that exact script played out on move 39: with a frantic 39 seconds on the clock, 39.Kf4 let an evaluation of +1.59 slip to +0.12. What no dossier can foresee is how the other side of the board will react under the same pressure, as the first diagram highlights. Scouting your opponent's flaws gives you sharp tools while there is time to calculate; when the clock drops to 38 seconds, your own ingrained habits and tactical composure take over.

Otsuka's profile highlighted that his blunders tend to erupt in complex middlegames, just like the 19.Rxd5 blunder that cost him against Monaco. His miscalculated 22...Ne3 against Fier was of the exact same breed, with the added sting of being played with nearly 28 minutes left on the clock. Marta Garcia Martin, who had squandered a massive +9.96 in round 6, blundered with 23...Qg4 with 27 minutes 10 seconds remaining, catapulting the engine evaluation from +1.57 to +5.11 in favour of Ouellet — who missed 24.Qe8+ and played 24.Nf3 (+0.87), winning later on. Monserrat Morales Flores, with a 49-centipawn loss rate in middlegames, slipped with 28...Be8 with 38 seconds left, sending the bar from 0.00 to +6.57 for Clara Dias, who found the clinical 29.f5.

Serbia, on the other hand, defied its historical blueprint. The USA dossier outlined a team prone to drawing winning games; yet in round 8, the Serbian setbacks erupted straight from level balances, starting with Sarana's rushed 21...Qg6 and ending with Ivic's fatal 62.Rc6+. An error profile tracks recent form, but each game is an unwritten contest. On the Armenian side, the report showed Susanna Gaboyan losing 39 centipawns per move in middlegames, having already seen a +7.40 lead vanish into a draw in round 1. Against Abrahamyan, the second diagram captures that same rollercoaster, featuring back-to-back slips as both sides missed their chance to put the game to bed.

The clock, and the mistake that repeats

Every competitive chess player — from club players in late-night blitz to grandmasters in classical tournaments — knows the terror of time trouble. In round 7, with 36 seconds remaining, Librelato played 40.Ba4 and saw her position against Nguyen Thi Mai Hung crater from −0.18 to −2.87. In round 8, against Rocio Contreras Suazo, the nightmare played out at the identical pressure point: with 2 minutes 38 seconds left, 23.Bg5 surrendered her +0.72 edge to −0.99; and with 35 seconds, the fateful 27.Bc2? proved fatal, crashing from −0.56 to −6.05 in the diagram. The scouting report warned that Contreras drops 60 centipawns per move in endgames and regularly wobbles in winning positions. But the game never made it that far: the clock claimed the victim first.

This is precisely where chess opening theory hits a human wall, and where Personal Chess Coach transforms your training. Losing two consecutive games in the exact same fashion — with under a minute on the clock and tossing away fully defensible positions — is not bad luck; it is a subconscious pattern. Amateurs and masters alike fall into the trap of pouring money into opening courses while repeatedly hemorrhaging rating points to the same chronic time-trouble blunders. Personal Chess Coach runs every single move of your games through the engine, flags repeat errors grouped by game phase and pawn structure, and powers the 'Fix my mistakes' training mode under Exercises, giving you clock-free practice on your own critical moments until the correct plan becomes second nature.

The clock extracted a heavy toll on both sides of the room. With 53 seconds left, Francisca Sanchez Tobar pushed the reckless 55.Ne7 in a dead-drawn position against Terao, sinking to −5.30. With 44 seconds, Irina Krush grabbed the poison pawn 40...Qxa6?, handing Danielian victory on a platter (+9.14). None of those blunders had anything to do with opening preparation. And even the three huge mistakes made with plenty of time — Otsuka's 22...Ne3, Garcia Martin's 23...Qg4, and Prado Acebo's 27.Be4 — were pure conceptual breakdowns in the middlegame. A sharp dossier arms you with an advantage up to move 10 or 15; what happens on move 40 is entirely down to how you train your own flaws.

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

The Olympiad dossier we published was generated using official broadcast classical games. Inside the Personal Chess Coach platform, you hold that same level of grandmaster scouting power for any opponent: simply type in their username from Chess.com or Lichess. The system gathers their games, constructs their White and Black opening trees, highlights where they underperform their average, and delivers a concrete game plan, always displaying the sample size beside every statistic. If the database is sparse, the tool transparently flags the data as inconclusive, exactly as we did with Kohei Yamada. You can test the public dossier preview for free without even creating an account — perfect for checking before your weekend tournament whether your round 1 opponent has a leaky Sicilian waiting to be punished.

We believe in complete transparency with our players: two elements featured in this Olympiad coverage were custom-built and are not part of the standard product workflow. The game-by-game centipawn loss profiles were manually calculated by our editorial team, whereas the automated dossier concentrates on practical repertoire and scoring rates. Furthermore, our platform does not track the remaining seconds recorded in the PGN clock tag; it pinpoints positional inaccuracies, leaving time management reflection to the player's study routine. Yet, as round 8 demonstrated, neither gap diminished the insight: the scouting accurately called the stumbles of Carvajal and Otsuka, and the clock was plain for anyone to see.

Round 8 of the Olympiad leaves a lesson every experienced tournament player will vouch for: Personal Chess Coach solved the entire predictable phase before the clocks started, nailing 23 opening moves on the board, navigating six vulnerable positions, and yielding a resounding 4.5 out of 6 for those in the driver's seat. The rest was settled in the unseen details — 34 seconds here, 38 seconds there, and the mental grit not to crack in time trouble. When all is said and done, the most dangerous opponent you will ever face does not possess a FIDE rating or a Lichess handle: it is your own clock and your own recurring blunders. And against those two, the most effective training begins the moment you analyse your games.

Frequently asked questions

Update log

• 2026-09-25 — First published with round 8 finished: the 24 games of Brazil, the USA and Canada compared with the dossiers published on 23 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 Olympiad, open, matches 1–12: round 8 broadcast (USA v Serbia) — Lichess
• 46th Olympiad, open, matches 13–37: round 8 broadcast (Brazil and Canada) — Lichess
• 46th Olympiad, women, matches 1–25: round 8 broadcast (USA and Canada) — Lichess
• 46th Olympiad, women, matches 26–50: round 8 broadcast (Brazil v Chile) — 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
• How to analyze a chess game with Stockfish without just copying the engine