Table Tennis: When the Data Table Looks Better Than the Truth

In a small analysis room in Shenzhen, I once received a forty-page document about a table tennis tournament. The cover was beautifully printed, the table of contents immaculate, each match on its own page with full charts, curves and colour blocks. The person who handed it to me said: "This is all our data." I opened the first page. Then the second. Then the fifth. Every box had a heading, every column had a label — but inside, everything was empty. Not a single score, not a single name, not a single point of contact.

That perfect document contained exactly as much information as a blank sheet of paper. I sat still for a long while. Then I did something I would not have dared to do ten years earlier: I called the sender back and said three words — "I don't know".

It sounds simple. In this profession, it is the hardest sentence to say.

In Croatia, in the summer of 2026, I learned that a midfield does not run after the ball, it runs after space. That lesson followed me into table tennis and still holds. People count balls, count points, count trophies. Very few count the gaps.

Modern table tennis is a sport of tables and charts, and every time the rules change, the meaning of all previous data changes with them. In 2026, the ball went from 38mm to 40mm, reducing flight speed and spin, and instantly ageing every statistic about points won directly on serve. In 2026, the format was cut from 21 points to 11, rendering meaningless a whole set of numbers about the ability to come back in a long game. In 2026, the hidden-serve rule arrived, removing most of the server's advantage. In 2026, VOC-based speed glue was banned. In 2026, celluloid balls were replaced by plastic ones.

Each of those milestones is a reset for every table of statistics that precedes it. Yet on forums, people still set points from 2026 beside points from 2026 as if table tennis had stood still for thirteen years.

I am not against data. I live on data. A tactical wizard is not someone who sees more, but someone who looks where others have forgotten to look. Yet the most forgotten place, in every chart I have ever read, is the ground beneath the coloured cells.

The WTT ranking system is the clearest example. Points are calculated over a rolling 52-week window: old points expire automatically, new points are added. That means a player holding a high ranking is not necessarily playing well — they may be living on points earned twelve months ago. Look at the points column and you see stability. Look at the calendar and you see the pressure of defending points. Two different pictures, one table.

That is why I am wary of heat maps. On the court, a heat map highlights where a player stood most, where the ball landed most. It is beautiful. It can be printed. It can be sold to sponsors. But it cannot distinguish a player standing still in the middle of the table because they are deliberately blocking and squeezing, from a player standing still in the middle because they could not move in time. The same red square, two entirely opposite stories.

The problem is that the heat map has become the new fortune-telling of the table tennis world. It hides a player's real role in the tactical system, replacing it with a patch of colour that is easy to look at, easy to quote, easy to argue about on social media. The completeness of form has never been proof of the validity of analysis — a fully populated table can be as empty as that forty-page document.

Based on my experience watching matches, the gap between the chart and the truth usually shows up in the smallest details. I once sat beside a coaching bench at a tournament with no spectators. The arena was empty, but I could still hear the coach shouting instructions down to the metre. He was not looking at the statistics. He was looking at his player's feet, at the corner of the table the opponent had just left, at the gap opening up after each rally. That bench had no algorithm. It had human eyes.

The biggest blind spot in modern table tennis analysis is not missing data. It is data that looks complete. A table with enough columns, enough colours, enough charts creates the feeling that the question has been answered. The reader nods. The writer relaxes. Nobody goes back to check whether the ground behind the coloured cells is real.

The consequence is that analytical platforms come under growing pressure to publish a table every day, a chart every week. That pressure does not come from the need to understand the match, but from the need to fill the page. When the speed of production outruns the speed of verification, the table begins to detach itself from reality. The reader gets the feeling of having understood, not understanding itself.

I have seen analyses thousands of words long built from a single column of ranking points. The writer was not wrong on the numbers. But they overlooked one thing: a ranking is the outcome of a process, not a description of ability. Reading a points column without reading the calendar, without reading the opponents, without reading physical condition, is like reading a book's table of contents and thinking you have finished the book.

Saying "I don't know" is not surrender. It is the first step towards knowing what must be found. In my line of work, the only honest response to an empty dataset is to send it back where it came from, ask for it to be filled, rather than invent a story to make it look weighty. In Croatia, they did not need a Zidane to control the ball; they needed an invisible web of passes running through the right places. Analysis works the same way: value does not come from a flashy fact, but from a web of evidence thick enough to hold a conclusion.

So in this major-tournament season, when every statistical table is beautiful and every chart is ready for the page, perhaps the most worthwhile thing to do is not to add another layer of colour. It is to go back and ask: is the ground behind that colour real? If the answer is no, then the kindest thing an analyst can do is to say those three words plainly — and wait for real data.