Arjun Erigaisi vs Prin Laohawirapap: how a +5 advantage became the Olympiad's biggest upset

Short answer

In round 1 of the Samarkand Olympiad, Thai IM Prin Laohawirapap (2363) beat GM Arjun Erigaisi (2759), India's top board, with Black in a 54-move Colle. This analysis uses engine measurements to show that White's attack was correct, that the advantage reached +5.58 while Black was short of time, where it drained away between moves 43 and 47, and which move, 50.Qg7, turned a level position into a loss.

Event: 46th FIDE Chess Olympiad

How this was compiled: Round 1 game of the open section, played on 16 September 2026. Moves, ratings and clock times come from the official PGN broadcast on Lichess. Evaluations measured with Stockfish 19 at depth 24, one dedicated search per position (Threads=1, 256 MB hash, cleared between positions); Lichess's automatic annotations were not used. Clock times are those shown after the move, including the 30-second increment.

The essentials in a few lines

• IM Prin Laohawirapap (Thailand, 2363, born 2006) beat GM Arjun Erigaisi (India, 2759) with Black in 54 moves in round 1 of the Samarkand Olympiad on 16 September 2026; India won the match 3-1.
• Erigaisi's 18.f5 measures 0.00 on the engine: it was not an unsound attack. White's advantage came from Black's mistake 18...Re8 (+1.58) instead of 18...exf5 (+0.05).
• The advantage peaked at +5.58 after 38...Rd8, with Laohawirapap down to 46 seconds; 40.Rg3 kept +5.10, while 40.Qh7 dropped to +3.49.
• After the move-40 time control, Erigaisi's clock fell from 31:22 to 9:25 between moves 42 and 49, and the evaluation went from +4.79 to 0.00; the move that levelled it was 47.Nf3, when 47.Nxe4 kept +2.21.
• The decisive move was 50.Qg7??, played with 3:38 on the clock in a level position: 50.Qf4 held 0.00, while 50.Qg7 measures −7.14, with forced mate from move 53.

A quiet Colle, and an attack that was sound

The usual explanation when a 2759 loses to a 2363 is the favourite who forces the issue in a position that won't bear it. It is a comfortable explanation because it fits every upset. Here it struggles with the first twenty moves: 1.d4 Nf6 2.Nf3 d5 3.e3 c5 4.c3 e6 5.Nbd2 Nbd7 6.Bd3 Be7 7.O-O O-O 8.Ne5 is one of the most sober set-ups available after 1.d4, chosen to play without theory, not to finish the game by move twenty.

White followed the textbook plan for the structure: exchange on e5, push f4, open the centre with c4 and e4, and only then advance f5 and f6 against Black's castled king. The engine objects to none of those steps. After 18.f5 the evaluation is exactly level, and the move has a concrete purpose: opening the b1 bishop's diagonal and gaining space next to the enemy king.

It was Laohawirapap who tipped the balance, with 18...Re8, and the clock helps explain why. By move 17 he had 21 minutes left against Erigaisi's 55, having spent more than an hour on his first seventeen moves. From then on Black was playing in time trouble, and much of White's advantage over the next twenty moves came from the opponent's haste rather than from any reckless plan.

Up to +5.58, against an opponent out of time

Black's second serious mistake came on move 31: 31...h5 took the evaluation from −0.18 to +2.55, played with 2:51 left. For the next nine moves Laohawirapap never had more than 2:03 after moving, living on the 30-second increment, and White's advantage grew to its peak on move 38.

Chess.com's coverage pointed to 40.Rg3 as the simpler route, and the measurement agrees: +5.10 against +3.49 for 40.Qh7. The gap is large, but it does not decide the game. At +3.49 with move 40 made, White was still winning, and the Olympiad regulations add 30 minutes to each clock after move 40. Laohawirapap's time shortage vanished on the same move that White's advantage stopped being overwhelming.

This is where the game changes character. Up to move 40, White was playing an opponent with no time, and the pressure did half the work. After the control both players had more than 25 minutes, and the position asked for something else: a run of accurate moves to convert, with no help from the other player's clock.

Moves 43 to 47: the advantage drains away

The collapse was not one move, which is what makes it instructive. 43.Rg3 gave up almost two points, 44.Qg5 another 0.75, and each move looked reasonable on its own. Over that stretch Erigaisi's clock went from 31:22 after 42.Qxh5 to 9:25 after 49.Rxf3: 22 minutes for seven moves, in a position the engine still rated as winning.

On move 47 the choice was between winning the material back at once with 47.Nxe4 or keeping the tension and improving the knight. The game move looks more ambitious because it avoids the exchange. In practice it allows 47...Re2+ 48.Kh3 Bxf3 49.Rxf3 Qe1, after which Black's queen and rook are active against the white king in a position the engine calls exactly level. Chess.com also flagged 47.Nf3 as the move that threw the advantage away.

With the evaluation at zero and nine minutes left, Erigaisi was in a position every club player recognises: the won game has become a level one, and the scoreboard in your head still says it ought to be won.

Move 50: attacking a level position

This is the one moment where the overpressing story finds support in the numbers, and it matters precisely because it is a single move. The queen on g7 hits f7 and, with the pawn on f6, sets up mating ideas. Against a passive opponent it would be a fair try. With Black's queen and rook one move from the white king, it hands over the tempo Black needed to start checking first.

The finish is short: 50...Qh1+ 51.Kg4 Re4+ 52.Rf4 Qd1+ 53.Kg3 Qg1+ 54.Kh3 Re3+, and Erigaisi resigned. The engine already shows −7.14 straight after 50.Qg7 and finds forced mate from 53.Kg3. No later error changes the result: move 50 decided the game on its own.

It is worth noting what the game does not show. It does not show a favourite who looked down on his opponent in the opening, nor a premature attack. It shows a large advantage built against an opponent in time trouble, slowly wasted once the time trouble ended, and a decision to play for the win taken exactly when the position had stopped being winning.

What a tournament player can take from this game

What follows is our reading, based on the numbers above. Nearly every tournament player has lost a game like this: winning, letting the advantage slip without noticing the exact moment, and then, with the position level, carrying on as if still winning. The pattern appeared twice on the same day at the Olympiad: in Brazil's match against Botswana, IM Roberto Molina had the better position and lost to an active move played with three minutes left.

What the game offers to train is not 'don't push'. Erigaisi pushed at the right moments, and that is how he reached +5. What failed was the reassessment: after 47.Nf3 the position needed a new goal, and move 50 was still chasing the old one. A simple clock rule decided before the game protects against that better than any resolution to be humble made halfway through it.

• When an advantage slips, tell yourself what the evaluation is now, not what it was.
• With under five minutes left, prefer the move that keeps your king defended to the one that creates a threat.
• After the game, note the move where the advantage started to fall, not just the final mistake.
• Review losses from winning positions separately: conversion errors need a different kind of training.

Frequently asked questions

Update log

• 2026-09-17 — First published, with moves and clock times from the official PGN and evaluations 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 FIDE Chess Olympiad Samarkand 2026 | Open — Erigaisi vs Laohawirapap, round 1 (PGN with clock times) — Lichess
• 2026 Chess Olympiad Round 1: Thai IM Upsets Indian Number-2 As Favorites Prevail — Chess.com
• Samarkand Olympiad Day 1: Favourites, fireworks and a shock — FIDE
• Laohawirapap, Prin — rating profile — FIDE

Related guides

• How to bounce back after a loss in a chess tournament: a three-step mental reset
• How to avoid time trouble in chess: five rules for the last three minutes
• The Colle-Zukertort System: the set-up, the promised attack, and what the engine says about it
• Recurring chess mistakes: how to find the pattern that repeats across your games