Nine Blank Tabs and a Single N/A: The Discipline of a Data Writer

Nine Blank Tabs and a Single N/A: The Discipline of a Data Writer

6:42 a.m., Nha Trang. The spreadsheet that had been scheduled to run overnight had long since finished. Nine tabs, each one a layer of data: competitive results, the athlete's personal progression curve, the context and tier of the event, the qualification mechanism, the compliance and anti-doping file, the coaching and support system, the risk map, the media narrative and expectations, and the transmission chain out into the wider industry. All nine tabs returned the same three characters: N/A.

No wind reading. No altitude above sea level. Not a single split. No athlete name. No competition date. A dossier nine layers deep, and all nine layers empty.

In my trade, that is a result, not an incident. That afternoon I filed the longest report I had ever written for a case with no data: nearly 1,400 words, and the only conclusion cleared for publication was that no conclusion could be drawn. The desk asked me three times. I answered three times with the same sentence: no sample, no verdict.

I wrote in my notebook the line that has followed me for twenty-five years: "Numbers never lie. They simply wait for someone sober enough to listen." But an N/A is also a number. It says that somewhere in the collection chain, a mesh has torn. The job of a data writer is not to invent the fish. It is to point at exactly where the net broke.

Why a dossier can come back entirely blank

Sports operates on a fixed habit: it asks for the conclusion before the evidence. A coach wants to know whether his athlete can meet the continental standard. A press office wants to know whether the new signing is a good contract. An organiser wants to know whether a ticket is worth buying. Every one of those questions is legitimate, and every one needs the same thing: a data series long enough to compare. When that series does not exist, my job is to say the hardest sentence available.

I learned that discipline at thirty-two, in a small newsroom in Nha Trang where I was the only data reporter. After matchday 20 of the 2026 V.League season, I published a series using expected goals against to show that the Thanh Hoa defence, which the media had crowned the best in the league, was in fact conceding more than the quality of chances it allowed: xGA of 1.9 per match, with the goalkeeper saving only 64 percent of shots on target. The coaching staff called me the man who sits in the cold room. On 7 February 2026, Thanh Hoa lost 0-3 to Ulsan Hyundai in the AFC Champions League play-off round, exactly the script the numbers had drawn three weeks earlier.

I do not retell that story to praise myself. I retell it because it taught me something that runs against its own surface. What made my spreadsheet right was not that it was clever. It was that it had real data to calculate with: every shot, every position, every season of comparison. Had I held ten matches instead of twenty rounds, I would not have been allowed to write. Had I lacked the save percentage, I would not have been allowed to write either. The line between analysis and guesswork sits precisely there.

A year later I travelled to Russia for the 2026 World Cup at thirty-three, on the credibility of that same series. While the domestic press praised Germany's defence, I pointed out that their PPDA had risen from 7.3 in the 2026 cycle to 12.8 in 2026 qualifying, meaning the number of passes they allowed opponents before each defensive action had nearly doubled, the signature of a lost high press. I wrote that Germany would go out in the group stage and was laughed at by colleagues. On 27 June 2026, Germany lost 0-2 to South Korea and finished bottom of Group F. I did not win because I was smarter. I won because I chose a measurable indicator over an unmeasurable reputation.

Since then my rule fits in one sentence: what cannot be measured should not be written. But that rule has a second half few people notice. When there is nothing to measure, the data writer must write about the emptiness itself.

The four kinds of blank in an athletics dossier

Back to those nine blank tabs. They were not blank in the same way. After years of building tables and cross-checking them, I sort blanks into four types, and each demands a different response.

Blank type one: the data does not exist. This is the cleanest type and the easiest to accept. If nobody records it, nobody can read it. The clearest example is wind measurement at junior athletics meets. An athlete who runs 10.20 seconds with a 1.9 metres per second tailwind and another who runs 10.35 seconds into a 0.5 metres per second headwind are not in the same class, yet the results sheet prints two lines that look nearly identical. World Athletics sets the wind limit at +2.0 metres per second for records at 100m, 200m, long jump and triple jump for exactly this reason. Usain Bolt's world record of 9.58 seconds in Berlin on 16 August 2026 was ratified with a wind reading of +0.9 metres per second, which means it stands even after the supporting condition is stripped out of the equation. But at a junior meet with no wind gauge, an unratified mark is not a poor mark. It is an unreadable one. And a ranking system cannot digest what it cannot read.

Altitude belongs to the same family. The track in Mexico City sits around 2,240 metres above sea level, where thinner air lets a body run faster and hold speed over short distances, which is why a wave of sprint records fell there from 2026 onward. A mark set at that altitude, with no altitude declared, becomes a floating number.

Blank type two: the data cannot be verified. This type is far more dangerous because it travels with prestige. Training marks, leaked internal figures, numbers whispered in a gym and reprinted in a newspaper as though they were competition results. Eliud Kipchoge ran 1:59:40 in Vienna on 12 October 2026. It was one of the greatest athletic performances I have ever watched, and it is not a world record, because the run was purpose-built, paced by a rotating team and staged outside the official competition system. The difference between 1:59:40 and a ratified record is not speed. It is the verification file.

In football this type wears transfer-market clothing. An agent says his client has passed a medical. A social media account declares the deal done. A fee is floated to test the reaction. Three sources, none verifiable, all quoted equally. I call that noise, and noise has no specific weight.

Blank type three: the sample is too small. Here the problem is not missing numbers but numbers that are not yet enough to say anything. One race is not a level. In my consulting work I only commit to a judgement on an athlete when I have at least three consecutive seasons. Nguyen Thi Oanh won three gold medals at the 31st SEA Games in Hanoi in May 2026, in the 1500m, 5000m and 3000m steeplechase, and the value of that performance lies in the fact that it does not stand alone: it is a link in a long chain recorded over years. Quach Thi Lan's gold in the 400m hurdles at the 30th SEA Games in 2026 follows the same logic, a milestone within a series rather than a solitary flash.

Based on my experience watching matches, the small sample is the most common trap in both football and athletics. When my model showed that Morocco's defence at the 2026 World Cup broke the offside trap successfully 71 percent of the time, and that goalkeeper Bounou exceeded post-shot expected goals by +3.2, I drew no conclusion from a single match. I drew it from seven matches, and from the fact that the pattern repeated against opponents with completely different styles. Seven matches is still a small sample, but it is a structured small sample, and structure is not coincidence.

Blank type four: the data is contaminated. The hardest type, because it is not empty but full, only full of the wrong thing. Part of any performance comes not from the human being but from the equipment, the conditions, the pacemaker. From January 2026, World Athletics capped road racing shoe sole thickness at 40mm and permitted only one rigid plate in competition shoes, after a period in which carbon-plated models with pressurised foam kept breaking records. Before that, a 2026 study published by the manufacturer itself found that the famous carbon-plated shoe improved running economy by roughly 4 percent over the previous model. Four percent is not small: over a marathon it is worth about five minutes. When an athlete smashes a personal best by three minutes, my first question is not what he trained. It is what he wore.

An indicator is a messenger, not a judge

Every metric I use daily is a proxy. xG is not goals. PPDA is not defensive quality. A wind reading is not form. They are narrow windows onto a wide room, and their value lies in our remembering that the room is wider than the window.

Germany's PPDA rising from 7.3 to 12.8 did not cause their elimination. It was a symptom of a structure that had aged, and the structure is what eliminated them. Likewise, Morocco's 71 percent offside-trap success rate does not explain their run to the semi-finals. It is one piece of the puzzle, and I would have been wrong to sell it as the only one.

Missing splits are the most concrete form of distortion. A 5000m runner who clocks 13:40 can have split it two very different ways: an even effort from the gun, or sitting in the pack until the last 400 metres and then unleashing a final lap. Same final result, two opposite forecasts for the next race. If my table has no splits, I am forced to treat those two athletes as one. That is not analysis. That is grouping.

When every layer is blank, what remains

Nine blank tabs gave me three actionable diagnoses, and I always try them cheapest first.

Diagnosis one: the data source does not exist. If the meet has no wind gauge, no electronic timing, no split cameras, the answer lies in organisation, not in analysis. The correct action is to upgrade the meet, not to write a long piece filling the gap with inference.

Diagnosis two: the source exists but is not published. Many clubs and federations keep GPS data, load data and medical data locked away. I have been in that position and had to choose. In 2026, when Khanh Hoa were fighting near the bottom of the table, I faced the choice of disclosing internal data to keep my role as a journalist, or keeping it sealed to protect the team. I chose the team, and my old newsroom cut ties with me. Since then I have kept a two-role principle: never mix a club's proprietary data into a public article, use only data from official platforms. A blank caused by confidentiality is not the same as a blank caused by laziness, and readers deserve to know which one they face.

Diagnosis three: the data exists, is published, but is not standardised. This is the most expensive type because it demands fixing a process rather than a single article. As a reporter I once had to persuade the desk to standardise a data box at the foot of every match report: xG, xGA, save percentage, passes allowed per defensive action. At first it was treated as a nuisance. Within months it became mandatory for every reporter. A fixed template is cheaper than an argument about who is right.

The transfer window is where all three diagnoses mix most violently. The transfer window is not a market fair. It is a cost-optimisation problem across individual metrics. A contract carries three layers of numbers that few people read together: the structure of the release clause, the wage bill after the new deal is added, and the player's actual minutes at elite level. When a player who has not played 50 top-flight matches is valued at 100 million euros, the third layer is empty and the first layer is carrying the entire belief. That is the signature of a bubble, and a bubble always bursts at the thinnest point.

The reverse trap: filling the gap with story

What worries me is not a blank dossier. What worries me is the industry's reflex when it meets one: to fill it with narrative.

That reflex runs on a recognisable mechanism. When data is missing, adjectives replace it. When the sample is small, a single moment does. When verification is absent, the speaker's reputation does. And when all three tiers are missing, a headline strong enough that nobody has time to ask for the source. I have watched this mechanism operate often enough to know it is not fraud. It is laziness, well organised.

In athletics it appears as unratified training marks printed in the present tense. In football it appears as small models presented as large laws. In both places it shares the same blind spot about execution: a model describes, a human decides. A spreadsheet does not run the last 400 metres, does not choose the pass, does not carry the weight of twenty thousand people in the stands.

I once managed to measure the strength of what a spreadsheet cannot measure. In March 2026, when the pandemic emptied every stadium, I compared 14 home matches played by Binh Duong with crowds, generating xG of 1.85 per match, against 10 matches without crowds, generating xG of only 1.31. Home advantage was inflated by 29 percent. But I could only draw that conclusion because I had two groups large enough to compare. With three matches, I would have stayed silent. And I remember this: the empty stadium taught me to hear what twenty thousand people used to drown out.

The same disease appears somewhere else, where I am less comfortable saying it. When video referees draw an offside line down to the millimetre, a measuring tool becomes a verdict machine. A goal is erased because a toe pokes past the line, while the attacking instinct of players is eroded by the very fear of it. The referee becomes the editor of the match. A tool meant to reduce error is redefining the game, and that is far from its original purpose.

There is one variable I always refuse to set to zero. Luck is the residual that the model cannot explain, and I never reduce it to zero. A striker who hits the post three times in one match is not a bad striker. A team that wins four in a row with a negative xG differential is not a strong team. That residual exists, and the job of the analyst is to name it correctly, not to delete it from the equation so the equation looks prettier.

I worship data, but I pray through real-world verification. And part of that verification is admitting that some things must be recorded in words rather than numbers. An athlete's emotion before a final is qualitative data. I record their exact words, the moment they said them, and the physical state that came with them. Three years later, looking back, that qualitative chain usually explains the collapses the model left blank.

Signals for the next data cycle

Back to those nine blank tabs that morning. I did not delete the file. I scheduled it to run again and attached three checkpoints.

Checkpoint one: whether a wind reading line appears in the official results. Without that line, every performance comparison drops one tier of confidence.

Checkpoint two: whether split times are published alongside the result. With splits, I can separate the even runner from the late kicker, and the forecast for the next race changes completely.

Checkpoint three: whether ratification status and equipment declaration come with it. A mark without those two pieces of information is just a nice line of text.

When those three checkpoints light up, my verdict changes within a morning, with no further argument required. Before believing in a reputation, I need to see the data behind it. And if that data does not yet exist, the most useful thing a sports writer can do is publish the emptiness, with the exact address of the break.

Tomorrow the spreadsheet runs again at 6:42 a.m. I am leaving all nine tabs in place.