The Blank Column: Women's Football and the Cost of What Nobody Records

At nearly two in the morning on October 14, 2026, in Incheon, I reopened my WK League tracking sheet. The expected-goals column for Incheon Hyundai Steel Red Angels had already been coloured in with four different pens. But in the seventh column — the one linking minutes played to a player's name — three cells were empty. Those three empty cells belonged to a 22-year-old substitute forward named Choi Yu-ri. She played 214 minutes across the entire season, less than two full matches. And in those 214 minutes, her expected-goals figure was 3.2, the highest in the squad.

I sat in front of the screen for a long time. My spreadsheet was not wrong. It was simply empty. And in women's football, that emptiness is routinely misread as a verdict: there is nothing worth noticing here. From those numbers, I saw a person waiting to be called by name. That night I wrote the first piece in a series that later became the professional fingerprint of my career.

A habit of record-keeping abandoned long ago

Women's football has been systematically under-recorded compared with men's football, and that shortfall did not appear by accident. In 2026, the English Football Association banned women's teams from playing at member clubs' grounds. The ban lasted exactly fifty years, until it was lifted in 2026. Across those five decades, hundreds of thousands of women's matches were played on local pitches, in industrial estates, on bare ground — and almost none of them were captured in a decent written record. No record, no data. No data, no memory. No memory, no legacy.

The first FIFA-organised Women's World Cup took place in 2026 in China with twelve teams. By 2026, in Australia and New Zealand, the number had reached thirty-two. Vietnam's women's national team reached the finals for the first time, drawn with the United States, the Netherlands and Portugal. They lost 3-0 to the United States, 2-0 to Portugal and 7-0 to the Netherlands. Three matches, no goals scored, twelve conceded.

Those numbers were recorded fully and accurately, and I have no intention of denying them. But they are the visible tip of an iceberg whose submerged mass is barely measured at all. Across those three matches, how many kilometres did Vietnam's women cover? What was the average pressing distance of each line? Which player retained the ball best under pressure? Nobody can answer, because no positional-tracking camera system was installed to serve that purpose.

In Vietnam's national women's championship, eight teams compete inside a short calendar, and most matches are filmed from a single angle. If a coaching staff wants to review a phase of play, they have to film it themselves on a phone from the stands. I know this because I have stood in those stands, holding a paper notebook, counting passes by hand.

In South Korea, where I work, the WK League has eight teams and Incheon Hyundai Steel Red Angels have dominated for years. Even in a relatively professional league like that, the number of matches with complete positional-tracking data is far lower than in the men's K League. Major data providers such as Opta, StatsBomb and Wyscout all cover women's football, but the coverage is thinner, the depth shallower, and the commercial value of those data packages is priced far lower.

There is one fact I repeat at every seminar: FIFA's prize fund for the 2026 Women's World Cup was 110 million US dollars. The prize fund for the 2026 men's World Cup in Qatar was 440 million US dollars. That one-to-four ratio is not an accounting accident. It is the outcome of a chain of decisions, and the first link in that chain sits at the point of record-keeping.

When the model has nothing to read

I came to women's football through sports broadcasting, then moved gradually into data analysis from 2026. My main tool at the time was expected goals, known as xG. The underlying idea is not complicated: every shot is assigned a probability of becoming a goal, based on distance, angle, shot type, defender pressure and goalkeeper position. Add all those probabilities together and you get an estimate of how many goals a player should have scored from the quality of chances she created.

The problem is this: to calculate xG, you need event data for every shot. In the top European men's leagues, each match generates thousands of automatic data points. In the WK League, in Vietnam's women's league, in most Asian women's leagues, that number is many times smaller — and in some matches it is zero.

When I ran my model on 2026 WK League data, the system returned exactly one line for Choi Yu-ri: insufficient sample. Two hundred and fourteen minutes is far too little for a statistical model to assert anything. Technically, the model was right. Humanly, it missed a signal.

The lesson I took from that case became a principle I still hold: when the data says "not enough", the analyst's job is not to stop, but to change the unit of measurement.

I abandoned the automated model and switched to hand notation. I rewatched every match Choi Yu-ri appeared in, counted every touch, marked where she received the ball, logged the moments she entered the penalty area. Three hundred and seventy-two touches. Eleven shots. The xG figure I recalculated by hand was 3.2 — and I must be explicit that this is my own calculation, not the league's official number.

The point is not the figure itself. The point is that eleven shots in 214 minutes represents an unusually high density of chances, far higher than the starting forwards in that squad. In other words, every time she came on, she generated chances faster than the players who were trusted. But because nobody recorded it, and because her cumulative totals were so small, that signal sank into the white space of the spreadsheet.

A month after my analysis was published, the Incheon Hyundai Steel Red Angels coaching staff handed Choi Yu-ri a starting place in the final match of the season. She scored twice. The team won the title.

I tell this story not to praise myself. I tell it because it illustrates a mechanism of systemic distortion. Statistical models learn from past data. If the past has overlooked a group of people, the model will keep overlooking them — and will do so with a perfectly objective expression on its face. That manufactured objectivity is more dangerous than an open prejudice, because it wears the coat of mathematics.

When data is missing, the person suffers, not the machine

In 2026, thanks to that series of data-driven analyses, I was invited to commentate at the U-20 Women's World Cup in France. During the group-stage match between South Korea and England, I mispronounced the name of forward Ellie Brazil three times in the first half alone. Social media reacted before half-time. I remember the feeling clearly: my face burning, my ears ringing, and an entire studio watching my back.

Two weeks after the tournament, I sat down and rewatched every recording. I built a new spreadsheet listing the name, shirt number, date of birth, club and pronunciation of all 352 players at the tournament. I asked native colleagues in sixteen countries to record each name onto their phones, then I recorded those and listened for twenty minutes a day for a month.

Mispronounce a name once, remember it for life; correct it, and you value it more. I am not retelling this to apologise again. I am retelling it because I realised something: my record-keeping system, until that point, had not even included the name. I had been calculating xG for women players without knowing how to say their names correctly. That was a far more serious gap than any statistical error in a model.

Since then, whenever I open a women's football dataset, I place another file beside it: an identity file. Who is the person behind this metric, where is she from, how old is she this year, which clubs has she passed through, is she still playing while holding down a second job.

One example has haunted me for years. At a women's tournament in the region, I watched a match in which the away side had only thirteen registered players. Two of them had taken overnight flights from elsewhere, slept four hours, then walked onto the pitch. On the statistics sheet, they are two lines reading "substituted on, minute 70" and "substituted on, minute 84". Nothing else. Read only the sheet and you conclude those two players were insignificant. You would never know they played that match on the legs of people who had not slept.

When data is missing, the person suffers, not the machine. The machine is not undervalued. The machine only records what someone chooses to feed it.

A defensive metric and an asymmetry nobody mentions

In modern tactical analysis, one of the metrics I use most is PPDA — the number of passes an opponent is allowed per defensive action by your own team. The lower the PPDA, the more aggressive the pressing. It is an elegant concept, because it measures intent rather than only outcome.

But to calculate PPDA you must be able to count every defensive action in every zone, in every phase. That is only feasible with complete event data. In most women's leagues, that data does not exist. So we take a metric system designed for wealthy men's football, apply it to underfunded women's football, and conclude that women's football is hard to analyse.

I have hand-counted PPDA for several Vietnam women's national team matches in qualifying and at the 2026 Women's World Cup. The method is crude: watch the recording, press a key each time the opposition passes the ball in Vietnam's half, press another key each time a Vietnam player performs a defensive action — a tackle, an interception or a pressure. Then divide one by the other. I did this for three matches, which took nearly forty hours, and I am fully aware that the error margin of a single human counter is larger than that of a ten-camera system.

But even a number with an error margin is better than a blank cell, provided you state the error margin. What I learned from hand-counting was not really about the metric itself, but about a much larger asymmetry: in the same competition, under the same federation, in the same season, video assistant referee technology was deployed in some men's matches while most women's matches had none.

I hold a professional view on that technology that I have maintained for years: VAR reviews take far too long and shred the rhythm of a match. A two-minute wait is enough to cool a goal that has just been scored, to quiet a crowd, to leave a player standing in the centre circle with her fingers laced together. I say this as someone who has sat in commentary boxes and had to keep a story alive through that dead time.

But I must also say the rest. A women's match without VAR is not a fortunate match that has preserved its rhythm. It is a match in which a refereeing error can destroy an entire year of work for twenty players, with no mechanism to correct it. The absence of technology from women's football is not purity. It is legalised asymmetry.

A blank column is not a neutral column

There is a misreading I encounter constantly in conversations with colleagues: people treat an empty cell in a dataset as a neutral state, a place where nothing has happened yet. Not true. An empty cell is the product of a decision. It is the decision not to install cameras, not to hire a coders, not to buy a data package, not to send a reporter to the ground. Every empty cell has a decision-maker behind it and a budget that was cut.

On the market side, this causal chain is very clear. No data means no analysis. No analysis means no stories. No stories means no audience. No audience means no sponsors. No sponsors means the data budget is cut again. That loop feeds itself, and it is not fate. It is a structure that can be dismantled.

People often say women's football is not yet attractive, so it has not yet earned data investment. I would argue the causality runs the other way: because it has not been invested in with data, people have not yet seen how attractive it is. A three-player combination in Vietnam's women's league can be every bit as good as a similar combination in the men's first division. The difference is that the men's move gets cut into a thirty-second clip with motion graphics, while the women's move sits in a video file nobody opens.

But I also want to speak plainly about the trap on the opposite side, because I see many colleagues in the industry fall into it. That is the temptation to turn a lack of information into something poetic. People write that women's football is "a land of many mysteries", that there is "not enough data to conclude", and then they leave the story hanging there. To me, stopping at "insufficient information" is an abdication of work disguised as caution.

I receive many analysis reports in which every single cell reads "insufficient data". After reading, I know nothing more about the player. Such a report is methodologically correct but professionally useless. Had I done that with Choi Yu-ri in 2026, she would have stayed on the bench, and I would never have known what I missed.

And there is another trap, the one I find myself most prone to. For years I had a tendency to defend women's football unconditionally and to hesitate before criticising the tactics of women players, while being perfectly willing to dissect every misplaced pass by a male player. I realised I was treating them as sacred ground that must not be touched. That kind of coddling is a form of contempt. The deepest respect, for me, is fair criticism. People only criticise severely what they consider important.

A phone call during the pandemic

In 2026, when competitions shut down, I lost my working rhythm entirely. For three straight weeks I lay in my apartment in Incheon and could not write a single line. I reread my old pieces and found they read like technical reports written by a machine.

One evening I made a video call to Ji So-yun, the South Korean midfielder who spent many years at Chelsea. She told me that at fourteen she had left her family to pursue football, living alone, and that there were nights she cried in her room and nobody knew. My eyes filled as I listened. One woman telling the story of her childhood, and another woman on the other end of the line hearing it with the very childhood she had lived herself.

A phone call during the pandemic taught me that sport heals without an audience. After that night, I changed how I wrote. Every piece of mine begins with an ordinary moment in the life of a woman player, and only then leads the reader into the numbers. That sequence is not a writing trick. It is a moral choice: the person first, the metric after.

What is changing

There are signals worth noting. FIFA has expanded the Women's World Cup and increased the prize fund with each cycle. On May 17, 2026, at the FIFA Congress in Bangkok, Brazil was chosen to host the 2027 Women's World Cup — the first time the tournament will go to South America, a football market full of passion but also full of inequality. Women's leagues in Europe are selling broadcast rights at higher prices than before. Attendance at some women's leagues has exceeded the capacity of older stadiums, forcing organisers to move to larger venues.

In Vietnam, the women's national team has received more attention since the 2026 World Cup. Huỳnh Như, the captain and the first Vietnamese player to play professionally abroad when she joined Lank FC in Portugal, has become a name that young girls cite as a standard. Goalkeeper Trần Thị Kim Thanh has been discussed more often since the matches in which she faced attacking lines that were physically far superior.

But that attention is still largely confined to the few weeks around a major tournament. Afterwards, the spreadsheet closes again. Matches in the national women's championship are still filmed from a single angle. Nobody is still paid to sit and count passes.

At fifty, I understand that the pitch has no borders, but the heart has coordinates. And my coordinates lie in the columns that were left blank.

Every contract is a life being moved. Every time a woman player signs professionally, behind that signature is a family that rearranged the seating at the dinner table, a mother who learned to talk through a screen, a father who sold something. If the spreadsheet does not record those things, then the writer must.

I do not believe we will soon have complete positional-tracking data for every women's match. I believe in something smaller and more achievable: each of us in this profession can impose a discipline of record-keeping. Record the name correctly. Record the shirt number correctly. Record the minutes correctly. Record also what never appears in a table — the overnight flight, the tired legs, the three weeks when you could not write a line.

A women's football culture that is properly documented will not automatically become fairer tomorrow. But it will no longer allow anyone to say there is nothing worth noticing here. And as I learned on an October night in Incheon, sometimes all a player needs is for someone to open that blank cell and read her name aloud.