Brazil's round 5 opponents at the Olympiad: how Ireland and Hungary play

Short answer

In round 5 of the 46th Chess Olympiad, on 20 September 2026, Brazil play Ireland on board 31 of the open section and Hungary on board 12 of the women's. Drawing on 699 classical games broadcast by Lichess over the last three years, this piece lays out each player's repertoire in both squads, the lines where they score below their own average, where the engine finds the mistakes and — for the first time in this series — what the verified online accounts show of their blitz and rapid play, always with the sample size attached.

Event: 46th FIDE Chess Olympiad

How this was compiled: Compiled on 19 September 2026, with round 4 finished and round 5 not yet played: it starts on 20 September at 3pm Samarkand time (10:00 UTC). The pairings were checked on chess-results; the official PGN that names each team's line-up is only published when the round starts, so this dossier covers both full squads rather than four boards each. The repertoire uses over-the-board classical games from the last three years broadcast by Lichess, linked to each player by FIDE ID and excluding round 5 itself: 300 games across the five Irish players (13 to 135 each) and 399 across the five Hungarian players (40 to 126). Chess960 games were excluded from the classical sample starting this round. The error profile was measured with Stockfish 19 at depth 16 over each player's 12 most recent classical games, and every move quoted in the text was re-measured at depth 24 with a dedicated search. The blitz and rapid repertoire comes from Chess.com accounts verified one by one, over the last 90 days, and sits in a section of its own.

The essentials in a few lines

• In round 5, on 20 September 2026, Brazil play Ireland in the open section (board 31) and Hungary in the women's (board 12). Neither line-up appears until the official PGN at the start of the round, so this dossier covers all ten players in the two opposing squads.
• Kata Karácsonyi of Hungary opened 1.e4 in all 56 of her White games in the sample — 100% of them. Against 1...e6 she drops from 57.1% to 28.6% in 7 games.
• Adam Collins carries the most reliable edge in the dossier: with Black against 1.d4 he scores 46.2% in 13 games, against his own 55.7% baseline, and it is the only line in either match rated high confidence.
• Tom O'Gorman plays an online repertoire he does not have over the board: in person he splits his White games between 1.e4 and 1.d4, and across 77 White games in the last 90 days on Chess.com he opens 1.Nf3 in 30 and 1.b3 in 14 — scoring 28.6% with the latter.
• The most expensive endgame of the match belongs to Nomin-Erdene Davaademberel, at 41 centipawns per move, and it comes from a single game: over 158 moves against Hatice Asli Mustu in the Turkish league she let advantages of 7.4, 8.8 and 6.3 pawns slip — and won. Petra Papp has the highest middlegame figure across both squads, 46 against 5 in the opening.
• The figures are from the opponent's point of view and hold as tendencies: samples from 13 to 135 games, an error profile over 12 games per player, and every percentage carries the number of games behind it.

The round 5 picture

Brazil reaches round 5 after beating Iraq 2.5-1.5 and Albania 4-0. The two matches ahead have the same shape with the signs reversed. In the open section Brazil and Ireland are level on 5 match points, but Ireland leads the first tiebreak, game points, 10 to 8.5. In the women's it is Brazil that leads that tiebreak, 12 to 10, with both teams on 6 match points.

The ratings invert too. Brazil is the 31st seed in the open section and faces the 66th; in the women's it is the 44th and faces the 18th. Put another way: in the open section Brazil is the favourite on paper and behind on the table, and in the women's it is the other way round. The Olympiad runs eleven rounds, and this is what the Swiss system does — it pairs the tied, not the alike.

The sample column in the table below says how much of each dossier holds up. Alexander Baburin, the squad's only GM, has 13 broadcast classical games in three years — the smallest sample of the match, and only just above the 10-game line at which the dossier starts flagging low confidence. At the other end, Tarun Kanyamarala has 135 and Kata Karácsonyi 126. That is 699 games in total, and they are not spread evenly: a dossier describes well whoever plays often and says little about whoever plays rarely.

This round brings something new, and it comes from away from the board. Eight of the ten opponents have a verified Chess.com account — title verified by the platform, real name and country matching the FIDE record — and six of those have games in the last 90 days. It is the first time in this series that a blitz and rapid repertoire enters, and it sits in a section of its own, because it answers a different question.

Ireland: the most reliable edge is on the lower boards

Tarun Kanyamarala is the strongest player in the squad and the most consistent with himself: 68.8% with White across 69 games and 67.4% with Black across 66, a rare symmetry. There is no first move that opens an edge against him — with Black he answers everything in the sample well. What exists is on the other side of the board: when he has White and meets the French, 1.e4 e6 2.d4 d5, he scores 16.7% in 3 games against a 68.8% baseline. That is three games, and the dossier marks the confidence as low. The King's Indian, 1.d4 Nf6 2.c4 g6, gives him 41.7% in 6 — same direction, slightly better sample.

Adam Collins is the opposite case, and the most useful one. His gap between colours is wide, 64.3% with White against 55.7% with Black, and the hole appears exactly where the sample allows a statement: against 1.d4 he scores 46.2% in 13 games. It is the only line in either match that the dossier rates high confidence, and it narrows further in the Queen's Gambit Declined — 1.d4 d5 2.c4 e6 gives him 37.5% in 8 games. With White his system is the King's Indian Attack, which appears in three of his four most played openings.

Tom O'Gorman splits his White games into two almost exact halves: 1.d4 in 13 and 1.e4 in 13, out of 29. The difference is what follows. With 1.e4 he scores 84.6% — 11 wins, no draws, 2 losses — and with 1.d4, 53.8%. Against him the dossier points to 1.c4, which gave him 25% in 4 games against his own 55.8% baseline with Black; four games make that an indication, not a plan. The King's Indian also troubles him with White: 25% in 4.

Alexander Baburin and Kavin Venkatesan are the two the dossier says least about, for different reasons. Baburin has 13 games in three years, eight of them with White, and in all eight he opened 1.d4 — a hundred per cent, though eight games do not license the word 'always'. Venkatesan has 27 and the squad's lowest averages in both colours, 53.3% and 45.8%; against him the dossier points to 1.e4, which gave him 33.3% in 6 games, at medium confidence. With White, Venkatesan opened 1.d4 in 13 of 15.

Hungary: 1.e4 in fifty-six consecutive games

Kata Karácsonyi has the lowest rating in the Hungarian squad and the second largest sample of the match, 126 games, which is why the dossier describes her more sharply than anyone else here. With White she opened 1.e4 in all 56 games in the sample — not 54, not 'nearly all': 56 out of 56. A repertoire like that keeps no secret at move one, and the price of that is that everyone knows where it hurts. Against 1...e6 she scores 28.6% in 7 games, and the French as a whole, 1.e4 e6 2.d4 d5, gives her 33.3% in 6, against a 57.1% baseline with White. With Black she is more fragile, 44.3% across 70 games, and against 1.Nf3 she falls to 25% in 10.

Nomin-Erdene Davaademberel has the squad's most balanced sample, 99 games, and a wide asymmetry between colours: 66.7% with White and 53.7% with Black. Both of the lines where she scores below her own White average grow from the same move — the symmetrical knight development after 1.e4. Against the Sicilian with 2...Nc6 she scores 41.7% in 6 games, and against 1.e4 e5 2.Nf3 Nc6, 50% in 8. With Black, the Caro-Kann gave her 10% in 5 games, the lowest figure across both squads; that is five games, and the dossier marks it low confidence.

Zsóka Gaál is the squad's top rating and the player the dossier has least to say about, which in preparation is information rather than a gap. Across 94 games she has no recorded weak line in either colour, and no first move opens an edge against her: with Black she scores 51.0% in 49 games, and the best the dossier finds is 1.e4, at 52.0% in 25 — above her own average, which is to say no edge at all. With White she opens 1.d4 in 29 of 45.

Szidónia Lázárné Vajda and Petra Papp have 40 games each, the squad's median sample. Vajda is the best scorer of the team with Black, 62.5% in 16 games, and with White opens 1.d4 in 15 of 24, scoring 76.7%. Papp is the reverse: 50.0% with Black in 23 games, and against 1.c4 she scores 16.7% in 3 — the lowest number of the match, on the smallest sample of the match, and the two facts should be read together. With Black against 1.d4, the King's Indian gives her 30% in 5.

Where the engine finds the mistakes

The table gives average centipawn loss per move by phase. In the opening all ten sit between 5 and 9 centipawns, which says more about memorised theory than about play — it is the phase in which everyone is still inside what they studied. The difference shows up when the preparation runs out.

With 12 games per player, this figure compares the phases of one player, not players against each other. A five-centipawn gap between two names in the table supports nothing; the distance between Petra Papp's opening and her middlegame, 5 to 46, supports a great deal. That is 306 middlegame moves measured, and it is the largest jump across both squads.

The most expensive endgame in the table belongs to Nomin-Erdene Davaademberel, 41 centipawns over 246 moves, and the lazy reading would be that she plays endgames badly. The three moves the shallow pass flagged as her worst were re-measured at depth 24, as the rules of this house require, and all three come from the same game: round 10 of the Turkish league, on 10 August 2026, against Hatice Asli Mustu. On move 78 she was 7.4 pawns up and played Be4, which left her at 1.3. On move 117, 8.8 pawns became 0.4 with Qf3. On move 140, 6.3 became 0.7 with Bd3. The game lasted 158 moves — and she won it.

This is why a high endgame figure is not a synonym for bad endgame play. Hers is the number of someone who takes the game to the end, presses a winning position, lets the win slip three times and picks it up on the fourth. The 99 games in her sample confirm the habit: a median length of 89 moves and 64.6% of them passing move 40, the highest proportion in the Hungarian squad. Play twice as many endgames and the engine has twice as many chances to find an expensive move.

Kata Karácsonyi has a different shape: 42 in the middlegame and 40 in the endgame, two high and level numbers, describing trouble that starts early and does not improve. On the Irish side the figures are tighter, and the interesting one is Adam Collins, at 17 overall with all three phases between 6 and 20 — the steadiest player in the table, and precisely the one whose repertoire weakness is the most reliable in the dossier. The two facts point opposite ways, and neither cancels the other. Tom O'Gorman is the only Irish player whose middlegame, 29, is worse than his endgame, 21.

Two things are left out of this table on purpose. Blunder counts, because the classification uses uncapped loss: a winning player who picks a slower mate collects a blunder for a move that changes nothing, and average centipawn loss, which caps each evaluation at 1000, does not suffer from that. And the worst moments themselves, which only enter the text after being re-measured at depth 24 — of the moves checked this round, most sat in positions that were already decided, won or lost, where a shallow ruler charges fortunes for moves that change no result.

What the online accounts show, and what they do not

Six of the ten verified accounts have games in the last 90 days: Kanyamarala (55), O'Gorman (153), Collins (300), Venkatesan (279), Lázárné Vajda (11) and Papp (4). Zsóka Gaál's and Nomin-Erdene Davaademberel's accounts exist and are verified but record no games in that window — which does not mean the account is not theirs, only that there is no sample. Alexander Baburin and Kata Karácsonyi have no account meeting the proof standard, so they do not appear here.

O'Gorman's online repertoire is the reason this section exists. Anyone preparing from the over-the-board record alone would see a player who opens 1.e4 or 1.d4 and nothing else; anyone looking only at the online account would see a 1.Nf3 and 1.b3 player, the Nimzo-Larsen Attack, which shows up in 11 of his most played openings and gives him 4 wins against 7 losses. Both descriptions are correct, and neither is evidence for the other. This is why the dossier never adds the two sources together: in a classical Olympiad game, with 90 minutes and an increment, the repertoire that counts is the one from the board.

What the online record adds is a more recent and wider portrait. Adam Collins, over the board a King's Indian Attack player, has 300 online games in the last 90 days, and with White his most played opening is against the Caro-Kann, with 10 games and 70%. Kavin Venkatesan has 279, and with Black the Sicilian appears in 15 of them, at 60%. These are blitz and rapid figures, with everything that implies: they are evidence of taste, not of preparation.

Petra Papp has 4 online games in the window, and the dossier itself marks the sample as low confidence. Four games describe no one's repertoire, and they are quoted here for the same reason Baburin's 13-game sample sits in the first table: stating the sample size is part of the figure, and leaving it out would be the one way to lie with true numbers.

What to watch in round 5

It is worth separating what the dossier does from what it does not. It shows where an opponent has scored below their own recent average, with the sample size attached. It does not say what the Brazilian players are good at, does not measure anyone's form this week, and does not predict a result. The two only meet when the line where the opponent scores less is also a line our side plays — and in round 4 that happened on one of the eight boards.

In the open section, the most direct reading is Adam Collins, if he is fielded: 1.d4 against him is the only recommendation of the match that the dossier rates high confidence, and inside it the Queen's Gambit Declined narrows further, to 37.5% in 8 games. Against Tarun Kanyamarala the dossier finds no edge with White, and the only recorded weakness requires him to open 1.e4 and meet the French — a situation that depends on him, not on whoever faces him.

In the women's section the figure that stands out belongs to Kata Karácsonyi, and it comes with a condition: all 56 of her White games began with 1.e4, so the French only comes into play if she is fielded and has White. The converse is on record too: if the Brazilian has White against her, 1.Nf3 gave her 25% in 10 games, at medium confidence — a better sample than the French, and a less spectacular recommendation.

Finally, the standing reminder, which round 4 illustrated well. André Diamant opened 1.c4 against an opponent who, across 29 games with Black in the sample, had never faced that move: the dossier had nothing to say, and the game was drawn. Playing outside the opponent's dossier is not a mistake — it is a legitimate decision, and sometimes the best one. What the dossier guarantees is that the decision is taken knowing what is being left unused.

Frequently asked questions

Update log

• 2026-09-19 — Initial publication, with the round 5 pairings checked on chess-results and the round not yet played: repertoire from classical games broadcast by Lichess, error profile with Stockfish 19 at depth 16, quoted moves re-measured at depth 24 and, for the first time in this series, a blitz and rapid repertoire from verified online accounts. The dossiers cover both full squads; once the line-ups are out, the same file will show the four boards and the round's game.
• 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 5 pairings — Chess-Results
• 46th Chess Olympiad Samarkand 2026, Women — round 5 pairings — Chess-Results
• Kata Karácsonyi — broadcasts under FIDE ID 769584 — Lichess
• Tom O'Gorman (toog) — titled profile that verifies the account used for the online repertoire — Chess.com

Related guides

• How to build an opening repertoire: 1.c4 as White and the Sicilian as Black
• Corporate chess leagues: what changes when the point belongs to the team
• A 7-day chess tournament preparation plan: from repertoire to energy management