The Blank Cell on the Analysis Sheet: Vietnamese Chess and the Limits of Data

Da Nang, a night in August, half an hour before airtime. The twenty-seven-year-old analyst laid an A4 sheet in front of me, printed with four familiar columns: average centipawn loss, engine match rate, live rating, head-to-head record. I scanned down each row. In every cell, instead of a figure, three identical letters sat there with a cold finality: N/A. Insufficient information to assess. He shrugged: "The system finished running, but it couldn't pull anything out, boss."

On the monitor, two young national-team players were shaking hands in a qualifier. The game ran five and a half hours and closed in a drawn rook-and-knight versus rook-and-bishop endgame. For those five and a half hours I talked without stopping, telling viewers about the breathing rhythm of the player with the white pieces, about how he shifted his knight exactly one square sideways, about the scrape of a wooden chair on the floor when the coach stood up. None of it appeared on that sheet.

Then I thought: perhaps that blank sheet was telling a different story, and a more honest one than any dashboard I had ever held.

Ten years of data covering the board

Vietnamese chess has entered a decade in which data is a compulsory travelling companion. The International Chess Federation (FIDE) publishes its rating lists on a regular monthly cycle, and every professional player lives with three separate columns: classical, rapid and blitz. Training centres at home now boot the engine before they open the board, and any commentator walking into a studio is handed a dossier of metrics.

Le Quang Liem, Vietnam's number one, was the first to push the country's classical rating past 2700 — a mark he reached in the early 2010s according to published FIDE rating lists — and at one stage climbed close to 2740. Nguyen Ngoc Truong Son held a place among the region's elite for nearly two decades. The generation that followed, alongside women players such as Pham Le Thao Nguyen, turned Vietnamese chess into a familiar force at continental and regional events.

Alongside the results, the measurement toolkit thickened. Average centipawn loss (ACPL) tells you how far a player's moves deviate, on average, from the engine's best choice. Engine match rate measures how closely a player mirrors the machine. Live rating updates after every game, while performance rating reflects form within a single event. At the format level, major events settle matters through playoffs: rapid first, then blitz, and when everything is still level, an Armageddon game — White receives more time but must win, while Black advances with a draw.

Those tools are genuinely useful. They also have genuine limits, in ways only someone sitting in the playing hall can see.

What the engine cannot measure

Based on my experience following games across many national and international events, the prettiest numbers usually belong to the safest player. Someone matching the engine ninety-two percent of the time will post a low ACPL, an analysis sheet glowing green, and will often close the game without a win. By contrast, the player who dares to push a move the engine ranks "second best" — a move that forces the opponent to solve a difficult problem with ten minutes left — is often scored lower by the dashboard, even while holding the better position on the board.

The machine can count error, but the machine cannot count fear. A move that is not technically optimal can be the human-optimal move, because the opponent across the board also has only one clock, one tired memory, and one evening that has already run long. The engine analyses a position under unlimited time. A human analyses a position with three minutes left. Those are not the same problem.

I once sat through an endgame in which both sides played perfectly according to the machine for the final forty moves. The engine match rate touched near-total, ACPL hovered near zero, the analysis sheet was as clean as a textbook page. Online viewers had left the room by move twenty. The metrics said this was a high-quality game. My memory says it was an evening nobody wants to recall.

That is why I increasingly distrust dashboards that are full to the brim. When every cell holds a figure, people feel they have understood the whole game. That feeling is a soft trap, and it costs far more than a blank sheet.

Another paradox sits in the draw rate. In elite round-robins, draws routinely exceed sixty percent, and every draw is a commercial loss: spectators do not buy tickets to watch two men nod at each other after thirty moves. Organisers respond by cutting time, adding playoffs, bending the rules toward drama. But manufactured drama and chess quality are not twins. Cutting time to thirty minutes per side increases the number of errors, and errors sell better than a perfect game. The ACPL of the whole event will drop, meaning technical quality falls, while revenue rises. A data system that can measure both directions at once would be a complete one.

Opening preparation is the category that stays nearly invisible on camera. A second can spend three weeks reconstructing a line that his player will use exactly once, across exactly seven moves, and never reopen. If that line has never appeared in a database, it has a name: a novelty. The analysis sheet will mark it with a question mark, because the engine has nothing to compare against. Three weeks of human labour appear on screen as one blank cell — one of the most beautiful blank cells I have ever seen.

Formats and invisible gaps

Most of the public only meets chess through playoffs. There, a player can win six gruelling classical games, then lose one fifteen-minute rapid game and be eliminated. A high rapid rating does not guarantee a player is good at rapid under knockout pressure; shaking hands and a quickening breath are not in any engine's database. But the final scoreboard still records only one line: who advanced.

The Armageddon game is the clearest illustration of rules creating an asymmetry of risk. White gets added time but must win. Black loses time but is allowed a draw. Two players walk into the same game with two completely different problems — and afterwards their ratings are compared as though they had just played the same rule set.

At the qualification level, the complexity is thicker still. The World Cup is a knockout event, where one bad afternoon can erase a year of preparation. The Grand Swiss is a large Swiss-system event, where every draw is a survival piece. And a rating path always exists, a route only a very small group in the world can reach. Three roads, three kinds of pressure, three personality types suited to them. A single rating column cannot express all three.

Young players and the adult rhythm

There is one place where the data usually goes silent, and goes silent alarmingly: junior events and wildcard entries.

A fifteen- or sixteen-year-old talent pushed into adult competition at five or six hours per game is usually greeted by the media as a symbol. But their body and their research base are not yet complete. They are written into the rating list before their capacity to withstand long pressure has been built. The rating will fall, and that fall easily becomes a label stuck to a biography.

Wildcards into invitational events are another form of privilege. They place a player in a prestigious field without passing through qualification, and they are rarely scrutinised the way a chain of real results is. Oversight systems built to track the path of official qualifying places turn out to be soft at this kind of entrance. Most of the public looks at an event list and sees a lovely name. Few look at that name and ask: why.

At home, the national championship and regional games long served as the most important training ground for a generation of players, before online open events ballooned and pulled young players into a very different competitive rhythm. A qualifying game in a national event may carry a rating several hundred points lower than an online open, but it teaches what no online event can: sitting still in one place for five hours, with no exit button and no other window to open.

At the regional level, chess has featured in the SEA Games programme, and medals there created a generation of fans who knew players' names before they knew the names of international events. The Asian Games has also seen Vietnamese players push into the leading group. Those milestones sit inside no single metric; they sit inside the memory of television viewers, and memory does not update monthly.

The blind spot of collective memory

We remember the sacrifices of material. We forget the forty moves of quiet accumulation that nobody photographed. Data behaves in the opposite direction: it records the average, the repeated, the things nobody narrates. The two memory systems drift apart, and that drift is exactly where misreadings of a player are born.

In years of doing this work, I have watched the public canonise a player on the strength of three beautiful games, then turn away from the same player over one bad game in a playoff. The engine has never lied — it simply answers a far narrower question than the one we imagine we are asking.

A blank sheet is more honest than a full one. An analysis sheet marked N/A has confessed that it saw nothing. A sheet full of numbers quietly claims it saw everything. Between those two attitudes, I choose the first, because it leaves room for the person sitting at the board.

What remains after the bell

I thought I had been exiled from the board; it turned out the board was walking me deeper into myself. My commentary career grew out of the very night I understood that I would never fully tell a game, and that the attempt to tell it fully is the most beautiful part of the job.

A time will come when tournaments in Vietnam have enough machinery to produce analysis sheets with no blank cells left. When it does, the discussion worth having will not be about which metric is right, but about what we want to keep of the long nights nobody filmed. Because after the bell ends the game, the only thing left on the board is not a string of numbers, but two people who have just walked through five and a half hours of their lives.