The Empty Report and the Honesty Line of a Table Tennis Data Reader

Over 14 months at the analytics desk of a sports data company based in Munich, I signed off on 212 files sent to clients, most of them German table tennis clubs and a handful of football sides. Seven of those files ended with two words: NULL RETURN.

The bodies of those seven files were empty. No point-win rate on serve, no long-rally control index, no head-to-head table, not a single player name. Only the domain label, table tennis, was populated; everything else was blank.

A head coach called back to challenge me: “We pay for a blank sheet of paper?” I told him that sheet was worth more than any spreadsheet I had ever sent, because it stated exactly one fact — the input source held nothing to read, and any index I filled in at that moment would have been a product of imagination, not of data.

The lesson I have drawn from 18 years watching this industry is this: an empty report, honestly declared, is still a finding; a report stuffed with numbers but untraceable in origin is the real disaster.

My job, in the end, is to find out whether the data actually exists.

Outsiders picture sports data analysis as sitting in front of a mountain of gold: switch on the machine and the numbers flow in, leaving only the arranging to be done. The reality is the opposite. Most of an analyst’s time goes into verifying that the data exists at all.

How we work runs through two layers. The first layer reads the source — a news item, a match sheet, a video segment, a scoreboard — and extracts verifiable events: who, beat whom, what score, at what time, from what origin. The second layer takes those fragments and builds deep analysis: trends, comparisons, forecasts, confidence ratings.

When the first layer returns an empty list, the second layer has exactly one valid choice: to declare emptiness. Every other choice is fabrication.

It sounds simple. The pressure to fill a template is what is hard to resist. A report has nine sections, and every section has an empty field waiting for a number. The client pays for a product that looks complete. And there is a very human professional temptation: an empty field wants filling, an empty frame wants painting. I have seen enough reports “painted” that way to understand that formal completeness has never equalled analytical value.

This is where I want to speak to anyone reading table tennis transfer news this season: the credibility of a report lies not in its word count but in whether it states the origin and the certainty level of each claim. A two-thousand-word item that never says where its information comes from is more suspect than a single line reading “unverified”.

The seven empty files shared one telling feature. They were not short of data in the sense of “sparse”; they were labelled but hollow — the system had recognised the table tennis topic but extracted no content. That is the signature of a failure at the collection layer, or of an inaccessible source: a deleted article, one behind a paywall, a truncated file, a document with revoked access.

The danger of such a file is not that it is empty. The danger is that it still looks valid. It has a domain label, a nine-part frame, all its subheadings. A hurried reader can skim it, take it for a finished analysis, and carry it off to cite. And so an empty result, through a few retellings, becomes an assertion about sport.

I call that downstream propagation. An empty field at the first layer, if mistaken for a finding, breeds a wrong conclusion at the second layer, then a wrong decision at the third — choosing a player, setting a schedule, spending transfer money. No one in that chain means to lie. They simply fail to read the line saying the data does not exist.

Table tennis is a sport where data, read correctly, says a great deal. A serve is not merely the action that starts a point; it is a tactical instruction. Serve-point win rate, receive-attack success rate, average rally length, points won inside the first three strokes — these indices sketch a player’s style more sharply than any qualitative description.

I often use the table tennis model to talk about football, and the reverse. The structure of a table tennis exchange — serve, receive, rally, point-ending stroke — projects onto a football set piece with almost identical rhythm. A player who lets an opponent seize the initiative from the second ball is like a midfield that loses possession right after the first pass. Both are failures in the opening beat, and both are measurable.

In Germany, where Timo Boll and Dimitrij Ovtcharov once pushed table tennis into the group of seriously followed sports, people are used to reading a player through an index table before reading them through a news item. But to measure, you first need data. That is why I show no leniency toward reports that invent numbers.

In January 2026, when I was 25 and still an analyst in Munich, I published a fourteen-page report on TSV 1860 Munich. At the time the club had twelve matches left in the German second tier. My figures showed their average xG per match stood at 0.78 — the lowest in the division in five years. The local press mocked me, because 1860 Munich were a beloved club and I was just a foreigner with a laptop.

On 28 May 2026 they lost their relegation play-off, dropped to the fourth tier and lost their licence. The editor-in-chief of the paper that had mocked me later called back and commissioned a series on decoding the data of relegation-threatened teams.

I tell this story not to praise myself. I tell it because it shaped a professional habit I kept all the way into table tennis: never open with sentiment or a club’s brand name, always lead with one index, and always declare where the warning threshold lies. An xG below 0.8 per match, to me, is a red alert. In table tennis I set a comparable threshold for the rate at which an attacking player loses points on the serve beat.

In the summer of 2026, at 26, thanks to that 1860 Munich report going around, a national broadcaster hired me as a data expert for the World Cup in Russia. Before the round-of-16 tie between Japan and Belgium on 2 July 2026, I issued a warning. Japan were pressing with a PPDA — the number of opponent passes allowed before a challenge — so low that it worried me. They swarmed the front line, but the space behind the midfielders was a killing zone against a side that passes long so well.

Japan led 2-0 in the second half and lost 3-2 to lightning counter-attacks. Japan’s PPDA of 6.2 in 2026 was no accident; it was a manifesto written in numbers. The Japanese proved that pressing is not instinct, it is an exercise in arithmetic. The lesson I carried into table tennis is the same: a player who attacks in two quick beats does not do so out of “fire”, but because their placement structure and footwork rhythm allow a two-beat attack.

In May 2026, at 28 and already a mid-level editor in charge of data, I launched a project tracking every remaining Bundesliga match after the league restarted with empty stadiums. The home-win rate fell from roughly 42 percent to roughly 25 percent.

I immediately sent a recommendation to a client club fighting relegation: push your line higher away from home, because home advantage had vanished with the stands empty. They won four of six away matches and stayed up.

The summer of 2026 emptied the stands but filled the data tables — it turned out football had been missing that. When the ground no longer roars, you hear the keystrokes of the calculations more clearly. Since then, every piece I write treats the crowd — full or empty — as an independent variable, an experimental condition.

In table tennis that variable is even sharper. An arena with eight thousand spectators leaning toward one end of the table can change how a young player handles a serve at match point. But we only see it if we measure it, and we only measure it if we have data.

The transfer window is when data drowns in noise. Rumours appear faster than signed documents, and they appear precisely in the gap that data leaves behind. When a deal has no confirming figures, a story fills the void.

I rank every transfer item by four levels of evidence. The first is an official document from a club or federation, with dates and a seal. The second is a statement from a named agent or official. The third is information from an unnamed source, the kind journalism calls “a person close to the situation”. The fourth is a rumour circulating on social media, with no source and no date.

The first three can still be worked with, provided I mark the certainty level of each line. The fourth should be treated as noise. The summer transfer market is merely a slower version of the stock market: the numbers decide, not the rumours.

The subtle part is this: German table tennis clubs do not publish transfer fees. They publish contract lengths. For a player, the contract term and the release clause are the real story, not the figure a news site attaches to the deal. A contract with one year left and one with three are entirely different goods at the negotiating table, even if both are rumoured to be “valuable”.

I have often sent clubs a table with just two columns: years left on contract, and the point-win rate against direct rivals. That table is far shorter than a transfer news item, but it is usable. Based on my experience tracking matches, a player who shines only against weaker opponents yet wilts against stronger ones is the sign of a contract worth less than its price.

Here is the most counter-intuitive part.

This industry rewards confidence, not accuracy. A confident report always gets a call back; an honest report saying “we do not have enough data to conclude” gets a complaint. The incentive is inverted, and that inversion breeds countless analyses that sound very solid but have nothing beneath them.

The consequence is a paradox: the more reports, the less knowledge. People pump numbers into gaps not out of malice, but out of fear of leaving them empty. The fear of being thought incompetent pushes people to prefer a wrong but tidy conclusion over a right but open answer.

I also want to lay out my own blind spots. Table tennis data analysis has three limits I always disclose when I file a report.

The first is that correlation is not causation. A player who wins more when serving with sidespin may not be winning because of that serve. It may simply be that they met only lower-ranked opponents that week. Context must be controlled as a strict variable, or we are merely describing coincidence.

The second is sample size. At club level, table tennis has very few matches per season. Twelve matches is far too small a sample to assert any trend. Many reports I receive build an entire tactical system out of three matches.

The third is data drift. Since the plastic ball replaced celluloid, the measurable spin per exchange has changed fundamentally. Any cross-era comparison becomes skewed unless we state exactly what is being compared with what.

Together, those three limits mean every conclusion of mine must carry a confidence interval. A claim without a confidence interval is an unfinished claim.

And here is where the seven empty files became a professional ethics test. When the data comes back short, there are two roads. The first is to declare emptiness, take the complaint, and preserve the honesty of the report. The second is to fill it, take the praise, and put a decision built on fiction into a coach’s hands. I chose the first road, seven times in fourteen months.

I do not think that is a virtue. It is an operating rule. Fate was written in advance — we simply need enough data to read it. When there is not enough data, the most honest thing we can do is say we cannot read anything yet.

So how should the signal be tracked in the next round of the table tennis transfer window? I would suggest three things.

First, track contract terms, not rumours. A contract with one year left is a real, measurable signal that depends on no unnamed source.

Second, track a player’s point-win rate inside the first three strokes against direct rivals, not against the field at large. A figure measured against a specific opponent group is always more honest than a whole-season average.

Third, state the certainty level of every line. A transfer item annotating the confidence of each sentence is a usable item. One without such annotations, however long, is only noise.

When a source is inaccessible, when a document returns empty, the correct reflex is not to paint over the gap. The correct reflex is to close the file, mark it unverified, and go back to the collection layer to find the origin. An empty file correctly declared leads us to where the data is. An empty file painted over leads us to a wrong decision — and we will never know where we went wrong, because everything on paper looks reasonable.

In table tennis people still speak of unreturnable serves, of ripped loops, of ten-stroke rallies that hush an arena. I have come to believe that every magical night of sport has a hidden equation behind it. But the equation only solves when the variables are filled correctly, and a variable that does not exist must be recorded as non-existent, not assigned a value to please the reader.

I will keep reading table tennis through numbers, and through these transfer-window days I will keep returning empty reports when the input source is genuinely empty. Attentive readers will understand why that blank sheet is the most honest part of the whole file.