When the Spreadsheet Returns N/A: Data Discipline in the Middle of a Major Season

The match ended at 10:40 p.m. Haiphong time, and over the next twelve minutes my phone received nineteen messages. All of them said the same thing: a Vietnamese player had just won a Challenger match on hard court, and he was ready for a Grand Slam main draw. I opened my tracking file — the file I keep on every player who has ever made me sit in front of a screen past midnight. Eleven official matches in twelve months. Four on hard court. Two against opponents inside the ATP top 200. Cells with enough sample for me to allow myself a conclusion: three.

Nineteen messages. Three cells of data. That gap is where my work begins, and also where it is forced to stop.

Data is never in a hurry. The people in a hurry are the ones who get it wrong.

I write about tennis for Vietnamese readers from an apartment in Haiphong, but my working habits were forged in newsrooms where I learned that a grown-up sports piece should resemble a court file more than a bulletin. Precedent first, evidence second, and a verdict only once there is enough material on the table. Vietnamese football taught me that with one specific match, not with a theory.

In the summer of 2026, at Lach Tray, Hai Phong FC generated 1.92 xG against SLNA and lost 0-1 to an individual error. The opposing goalkeeper made 11 saves in that match, 3.8 times his own season average. I wrote that the result was unjust in a purely random sense. I was mocked for two weeks. In the third week, the Hai Phong head coach mentioned my numbers at a press conference, and the mockery stopped. From that day I set myself an invariable rule: no verifiable data, no conclusion — and the raw table ships with the article instead of staying on my hard drive.

Major-tournament season is temptation season. The calendar is dense, every week brings a new story, and every story demands a verdict that same night. I understand the pressure, because I have written on deadline too. But between a player with a full spreadsheet and a player with an empty one, the gap is not talent. It is the number of times someone has actually been measured.

In June 2026, before Germany faced South Korea in the World Cup group stage, I published an analysis showing that Germany's pressing coefficient had fallen from 8.1 PPDA in 2026 to 12.6 in 2026, with average distance covered down 6.2 kilometres per match. Germany collapsed in my spreadsheet before it collapsed on the pitch. That day they held 74% of the ball and lost 0-2. The lesson I kept was not the result: it was that a dense enough spreadsheet can see a collapse coming, while an empty cell says nothing at all. Those two situations require two different behaviours, and most failures in sports writing come from treating them as the same.

My standard analytical table has five columns. The first measures sample size. The second measures opponent quality. The third measures surface distribution. The fourth measures performance at decisive points. The fifth measures workload and physical condition. A player counts as readable only when at least three of the five columns carry enough sample, and the threshold for each column is set in advance, never adjusted by the result.

Sample size is the cruellest column and the easiest to skip. For first-serve points won, the standard error is the square root of p times (1-p) divided by n. A player serves roughly 57 times a match; with 11 matches I have about 630 service points to work with. Setting p at 0.6, the standard error lands near 1.95 percentage points and the 95% confidence interval spans about 3.8. With two matches, roughly 110 points, the standard error jumps to 4.67 percentage points and the interval swells to nearly 9.2. Same metric, two entirely different levels of reliability. What deserves writing is the width of the confidence interval. A 61% rate with a margin of plus or minus 9 points does not say the player serves well. It says he has not been measured enough.

Opponent quality is the column I check before reading any percentage. Eleven wins is a fact, but it only means something once you know where those eleven opponents sit in the rankings. In my file, two matches were against top-200 opponents; the other nine were against players outside the top 400. A 72% win rate does not describe the player's ability; it describes his schedule. This is the error I meet most often in domestic coverage, and it does not come from ignorance. It comes from the habit of reading a single column.

Surface distribution explains the rest. Four of the eleven matches were on hard court; the remainder were spread across clay and indoor carpet under entirely different conditions. No player converts form across surfaces at the same speed, and major-tournament season is where that difference gets magnified. Based on my own experience watching matches across a number of Asian Challengers, a player can win five straight in Nonthaburi and then lose in the first round in Europe with nothing that qualifies as decline. The difference is conditions, not mentality.

Performance at decisive points is the column I read last, because it is the noisiest of all. For break-point conversion I set myself a floor of 40 opportunities before I permit myself to speak. Below that floor, a player can look like a specialist in nerve simply because two balls landed on the line. I track matches with a paper notebook, and in it every break point is logged with its situation: second serve, after a long rally, or after a medical timeout. Circumstance weighs as much as outcome.

Workload is the final column and the one I regard as life-or-death. Spectators can leave the stands, but physical data never takes a break. Matches inside twelve months, consecutive weeks of competition, the number of three-set matches inside four days — these facts forecast the future better than any story about character. At 22, a player can carry that density. At 33, the same density is a loan with interest.

And here is where I have to write about the second line of every return timetable. In my files, a comeback is always recorded in two lines: the announced line and the actual line. The announced line is controlled by the communications department, and it is always optimistic on purpose. The actual line only appears when the player walks out and moves laterally. When a statement says wait until the weekend, I enter a question mark in the fifth column, and in most cases that question mark outlives the statement. On 4 June 2026, Novak Djokovic withdrew from Roland Garros before the quarter-final with a knee injury, having won four matches. No performance metric in my file predicted it, because the workload of a 37-year-old at a major cannot be read off a first-serve points won figure. I write that limit down instead of hiding it.

In Vietnam I use Ly Hoang Nam as the public benchmark, because his record is something anyone can verify. His career-high ranking is 231 in the ATP, reached in November 2026. In 2026 he won the Wimbledon boys' doubles title alongside Sumit Nagal. Those two facts are enough to build an axis, so that every comparison with younger Vietnamese players has a root. And that axis shows the real distance: a men's Grand Slam singles draw has 128 places, of which 104 are direct acceptances by ranking, 16 come from qualifying and 8 are wild cards. Climbing from 231 to 104 is roughly 127 places, equivalent to about three hundred ATP points, depending on the date.

Those three hundred points do not come from one historic week. They come from tournament structure: a Challenger 100 title is worth 100 points, a Challenger 125 is worth 125, an ATP 250 title is worth 250. The shortest and least glamorous road is three consecutive deep weeks at small events, a complete support team, and one unchanged surface. At this level, consistency is worth more than a single serve or a single inspired evening. Structure also explains why every player has to choose a schedule: the Roland Garros 2026 prize pool stood at 53.5 million euros, with the men's singles champion taking 2.4 million according to the organisers, while a week on standby at an ATP 500 can pay a few thousand euros and no points at all.

People remember results. I remember the conditions that produced them.

The counter-intuitive angle I want to raise is that our problem is not a shortage of data but the wrong kind of data. Those eleven matches will not become more reliable when I add a twelfth if the sample stays skewed: the same surface, the same region, the same group of opponents outside the top 200. Adding numbers to a distorted sample only makes the error look more precise. In statistics, that is how you build a house on sand and render the facade in cement.

A correlation is also not a cause. Winning several matches in a row does not create a Grand Slam main draw place. What creates it is the ranking; what creates the ranking is the points structure; and what creates the points is the schedule chosen by the player and the coaching team. Reversing that order is how form becomes a mystical word, and also how a 22-year-old gets placed on a scale of expectations he does not yet have the data to carry.

What bothers me most about this profession is the incentive structure. A wrong verdict is remembered for years; a refusal to conclude is forgotten over lunch. Market rewards therefore run against data honesty. I resist that with one simple habit: I put the error margin in the body text, right beside the figure, rather than burying it in a note at the end. When the margin sits inside the sentence, the reader has the right to hold me accountable, and I lose the free gift of vagueness.

There is one category of fact that is usually treated as no data at all. When a team does not publish workload, when a statement does not specify the location of an injury, when a player withdraws without an accompanying explanation — that absence is itself a fact. I enter it in the fifth column with a symbol, and that symbol carries the same weight as a metric, because it describes the player's own willingness to share information.

When the next major-tournament season opens, I will follow three signals. The number of consecutive weeks in which the player contests three matches or more. The share of top-200 opponents in his total matches. And whether that serve metric holds when opponent quality doubles. Those three signals arrive before any ranking, and they cost less than any prophecy.

That young player has the right to be measured enough before he is judged. My job is to keep the empty cell in the right place until that right is honoured.