Null Result: When a Football Analyst Has to Say 'Insufficient Data'

On my desk in Guangzhou sits a nine-layer document. The first layer covers tactics and technique. The second covers club finance and the transfer market. The third covers results and the public-opinion cycle. The fourth covers the league landscape and where a club sits in the food chain of resources. The fifth covers rules and compliance. The sixth covers the coaching staff and dressing-room health. The seventh covers the risk profile. The eighth covers media narrative and the gap between expectation and reality. The ninth covers the flow of an entire football industry, from academies to agents to broadcast rights.

In all nine layers, the data field is filled with the same sentence, repeated verbatim: insufficient information.

That document is in front of me for a very specific reason. The source material has no title. It has no club name, no player name, no timestamp, no provider, and no assessment of source reliability. Someone handed me an empty box and asked me to weigh it, measure it, and write three thousand words about it.

My job is to read football through data. But data does not grow out of thin air. A match can be told in hundreds of ways, yet only a small number of those tellings survive contact with video, with the match report, with transfer history and head-to-head records. It took me eleven years to learn the difference between those two groups of tellings, and most of that time was spent learning to recognise when I did not yet have enough to speak.

The pressure of the empty cell

Outsiders assume writing about a specific match is harder than writing about an abstract theme. For me it is the reverse. With a match in hand, I have a skeleton: goals, cards, minutes, starting line-ups, substitutions, the shape of the scoreline. I can be wrong in my interpretation, but I cannot invent events.

With nothing in hand, anything can become true. That is the most dangerous moment in this trade.

I have a nine-layer template, and a template always demands to be filled. There is a very concrete pressure in sports content work: an empty cell counts as a mistake. The site needs length. The algorithm needs something new every day. The editor needs copy before broadcast. And the writer, facing an empty box, begins to do the worst thing possible with it: filling it with adjectives.

'The team is in good form.' 'The defence is switching off.' 'The coaching staff has lost the dressing room.' Those three sentences sound a great deal like analysis. They are not grammatically wrong. They are wrong in one place only: they could have been written in advance, about any club, in any matchweek, and still be true. A sentence that is true of every match says nothing about any match.

I once wrote a piece nobody read. Three years later, it became my teaching file.

Nine layers, and what they are actually for

The nine-layer template is not a ritual for appearance. Each layer is a question that can be answered wrongly, and therefore a trap that can be spotted before it detonates.

The tactical layer forces me to name the structure rather than talk about spirit. The financial layer forces me to separate cash from a conditional receivable. The results layer forces me to separate process from the table. The league-landscape layer forces me to ask where a club sits in the hierarchy of resources, not merely where it sits in the standings. The rules layer forces me to remember that some things cannot be bought with money, even when you have money. The dressing-room layer forces me to remember that some things cannot be bought with tactics. The risk layer forces me to ask what happens if my first assumption is wrong. The media layer forces me to measure the gap between expectation and reality. The final layer forces me to look outside the club, at academies, agents, broadcast rights, at the entire industry standing behind a single match.

When all nine layers return a null result, that does not mean the analyst failed. It means the input was empty. A table with nine blank cells is an indictment of the source, not a confession by the writer.

Here is the point I want to hold throughout this piece: the value of a null result lies in stopping a wrong conclusion before that conclusion is written. In an industry that produces thousands of assertions every week, the most expensive thing is a refusal to assert, delivered on time.

Three matches do not create an identity

Every Monday I read at least a few pieces declaring that a club has 'found its identity'. Behind that judgement there are usually three matches.

Three matches is an almost meaningless sample in football. Across three matches, a team can win all three without playing better than its opponents on any process metric. Football has a low scoring density, and at low density luck carries weight. A ball off the post, a refereeing decision in the 88th minute, a save made with the fingertips: those three events are enough to flip a table but not enough to flip a system.

Based on my experience watching matches in the V.League and in the AFC Champions League, I have noticed a fairly stable pattern: teams that change structure after three matches are usually the teams that change structure for the fourth time after seven. Instability dressed up as flexibility.

In the V.League the effect is stronger because the quality band between clubs is narrow. A mid-table side that wins three rounds in a row can climb into the leading group and is immediately described as a tactical phenomenon. But place those three matches next to the fixture list and you will often find all three opponents were playing midweek in another competition, or were missing key players. Sample size is not only the number of matches. Sample size is also the context of each match.

This leads to a practical consequence for readers. When you read a post-match piece, look for the sample size before you look for the opinion. If the piece talks about form, ask over how many matches form was measured. If it talks about a new system, ask how many minutes that system has played with the strongest XI. If it talks about a turning point, ask which minute the turning point began and who decided it.

What xG explains, and what it misses

Expected goals, known as xG, has become the common language of the analytics world. I use it. I do not trust it to be enough.

xG answers a narrow question: given the location and circumstance of a shot, what was the historical average scoring probability. That is a metric of chance quality, not a metric of team quality. The two are routinely mixed up, and most online arguments about xG are arguments between two different definitions.

xG does not know who is shooting. A shot from a position with a 0.15 probability, taken by a first-choice striker and taken by a left-sided centre-back, describes two different events in execution even though it describes one event in data. xG does not know the state of the game: a team leading 2-0 shoots differently from a team trailing 0-2, even from the same position. xG does not know the referee's standard that night, and therefore does not know which team was permitted to play its own way. Nor does xG know the positions of the players who were not involved in the shot.

That last point is the largest blind spot. As a former player, I do not need to watch tape to know who is running in the wrong place. I look at the gap between two centre-backs and I know who arrived two steps late. No column in any data table records 'the right-sided centre-back is standing three metres higher than his partner in the 63rd minute'. But the second conceded goal usually starts from exactly that three-metre gap.

So when I read an analysis built only on xG, I know the author is answering half the question. The other half lies in things that cannot be measured, or can be measured but cost far more time than opening a data page. An analyst's job is not to pick the easier half.

Read a transfer by position, not by price

In recent seasons, loan deals with an obligation to buy have become widespread in Southeast Asian leagues, including the V.League. On the surface it is a sensible instrument: the receiving club pays nothing up front, the sending club secures future income, and a young player gets minutes.

Inside that structure there is a very clear transfer of risk. A big club sends a young player to a small club, letting the small club pay wages and training costs during the raw phase. When the player's value rises, the big club recalls him or sells him on under a clause signed long before. The small club gets one good season and loses the player at the exact moment he starts lifting them up the table.

The harder part to see sits in the accounts. An obligation-to-buy figure is a conditional receivable. A small club cannot use it to pay this month's wages, yet it must still count it in its financial planning and its transfer strategy. The result is a club locked in an intermediate position: rich enough not to be treated as weak, poor enough to be unable to keep anyone.

Based on my experience watching matches, the tactical consequence of this contract type is very concrete. The small club is forced to build a system around a player it knows it will lose, usually at the end of the season. The coach must choose between optimising the current campaign and preparing a squad without that player. Neither option is free.

I read a transfer not through the price, but through where the player will stand in the system.

A winger bought for a system that does not use wingers is a loss, whatever the fee. Conversely, a cheap player in exactly the right structural position can be one of the signings of the season, and nobody will call it a major transfer. That is why I always read the positional map before the final line on the invoice.

119 matches with no crowd

In 2026 everything collapsed. I stood up and rebuilt from the rubble.

In March 2026 the major leagues stopped almost simultaneously. I was twenty-one, with a run of cancelled pieces and a question I could not answer: what do you write about when there are no matches to write about. The Bundesliga was among the earliest leagues to return, on 16 May 2026, with matches played in empty stadiums.

Together with a group of students, I decided to turn that period into a large-scale social experiment. We took the matches played after the lockdown and compared them with the pre-pandemic period, controlling where possible for fixtures and opponent quality.

The result: home teams took only about 38 percent of available points, against roughly 47 percent beforehand. Home advantage did not vanish, but it narrowed sharply. That suggests a substantial share of what we call home advantage sits in the stands, not in the grass or the travel distance.

The video series reached roughly eight hundred thousand views on Bilibili within two months. But its real value lay elsewhere: it taught me that when ordinary data stops flowing, an analyst must find a new variable instead of waiting. The absence of a crowd is a variable. It exists, it is measurable, and it forces people to redefine a worn-out concept.

Croatia and the price of decisiveness

In June 2026 I wrote a piece arguing that Croatia were not dark horses. I gave two data points: passing accuracy of roughly 86 percent in qualifying, and squad depth superior to most teams in their bracket. The piece was mocked, largely because it ran against the story being told.

On the evening of 11 July 2026, in the semi-final, England led 1-0. In the live commentary I wrote one short line: England will fall. Hundreds of comments mocked me immediately afterwards. Croatia won 2-1 after extra time, Mario Mandžukić scoring the decisive goal in the 109th minute.

I was right. And I learned a lesson that was not in the data.

World Cup 2026 taught me one thing: hesitation is what wrecks every plan. But in the same week it taught me the opposite too: being decisive does not license me to insult the reader. I was right about the result and wrong about how I treated the people following me.

Since then I have changed how I write a strong argument. I keep the decisiveness but add a door: if the data holds. That phrase does not weaken the argument; it states precisely what the argument depends on and what would break it. It is a form of transparency, not a form of timidity.

38 turnovers in midfield

In August 2026 I wrote the first analysis of mine that got noticed, on the AFC Champions League quarter-final between Guangzhou Evergrande and Shanghai SIPG.

In the first leg in Shanghai on 22 August 2026, Guangzhou Evergrande lost 0-4. The familiar explanation was that the defence played badly. That explanation is not wrong, but it is useless, because it does not say what to fix tomorrow.

I counted 38 turnovers by Guangzhou Evergrande in the midfield zone. That is a metric of structure, not of attitude. When both full-backs push high in a 4-3-3, the midfield zone holds only three players against counter-attacks. Losing the ball there means the opponent has open ground in front of the back line within two passes.

I proposed a switch to a 3-5-2 with inverted wing-backs: one full-back pushing high, one tucking inside to keep the midfield thick when the ball is lost.

The piece had seven views after three days.

A month later Guangzhou Evergrande won 2-0 in a domestic match using a broadly similar structure. Forums began resharing the old piece, and it drew about twelve thousand reads. In the second leg on 12 September 2026, Guangzhou Evergrande won 5-1 and lost on penalties, in a match where their back line held a markedly higher structural line than in the first leg.

I tell this story not to praise myself. I tell it to show what those seven views taught me: a correct conclusion does not need to be read immediately in order to be correct. It needs to be recorded, with a timestamp and with numbers, so it can be checked later. Seven views were seven self-audits. Twelve thousand views were a social event, not scientific evidence.

Two independent sources, and the silence in the tunnel

In this trade I have a network of internal sources. It is my greatest advantage and my greatest temptation, because it lets me stand in front of a story before that story becomes a headline.

My rule is very simple and very hard to keep: two independent sources before publication. If an agent tells me the coach has lost the dressing room, I have a person with a motive, not yet an event. The agent benefits when the story is printed. A source's motive is part of the data, and ignoring it is a form of systematic error.

The lesson from the tunnel: the silence before a match says more than any press conference.

I have seen dressing rooms that were loud before a defeat and silent before a win, in both directions. But I have never seen a dressing room stay silent for three weeks and still produce good results. That is a signal, not a conclusion. A signal needs two more independent observations before it becomes a printable sentence.

In the chaos of a mid-season, what an analyst needs most is the clear-headedness of an outsider.

Vietnam, January 2026, and the flood of conclusions

On the evening of 5 January 2026 in Bangkok, Vietnam beat Thailand 3-2 in the second leg of the ASEAN Championship final, winning 5-3 on aggregate. Nguyễn Xuân Son scored twice before suffering a serious injury.

In the seventy-two hours that followed, I counted hundreds of pieces drawing conclusions about the national team's future, about one player's transfer value, about the correctness of a football philosophy, about a second golden generation. All of it from one match.

One match can be data. It cannot be evidence for a decade.

If I were allowed to keep only one sentence from that night, I would keep this: Vietnam's defence held its structure through the final thirty minutes of the second half, when Thailand pushed its entire midfield line high. That is a verifiable observation, with a timestamp, and it can be used for the next match. The rest is emotion, and emotion belongs in its own place, not in the place of data.

In the ninth layer of the template, industry flows behave the same way. A final raises broadcast viewership for three weeks, raises the valuation of a few players for three months, and raises the expectations of an entire football nation for three years. Those three curves do not share a gradient, and confusing them is the origin of most bad planning.

The other side of the null result

Here I have to argue against myself, because that is the job of the seventh layer.

A table full of insufficient information can be a sign of honesty. It can also be a sign of laziness dressed in technical language. The two look identical on paper, and in many newsrooms they are treated identically.

The difference sits underneath: did the analyst actually go and look, or simply wait for the data to arrive? Refusing to conclude is only valuable when there is a concrete list of what is missing, where to find it, and when it is needed. Without that list, a null result is an evasion in careful packaging.

I have fallen into exactly that trap. After 2026 I spent too long waiting for one more dataset, one more control match, one more season. I had a draft on tournament-football tactics that took me nearly two years to publish. By the time it appeared, three of its four arguments were out of date. Hesitation did not protect me. It only made my work older.

So the right null result and the wrong null result sit on either side of a very thin marker: have you done the homework. If you have, and the data is still empty, say so, with dates and with a list of what is missing. If you have not, do not call waiting a method.

The second uncomfortable side is this. When all nine layers are empty, the table is in fact an indictment of the input. At some layer, someone passed an unprocessed thing into the system and called it an article. In most cases I have encountered, the fault was not in the data. The fault was that someone wrote the conclusion before the data existed.

What to watch in the next matchweek

Next time you read a post-match piece and meet the words identity, system, or turning point, ask two questions: what is the sample size, and what would change the author's mind. If the piece cannot answer the second, it is not an analysis. It is a statement.

As for me, what needs watching in the next matchweek lies elsewhere. I will not be watching who wins. I will be watching which team keeps its structure when it falls behind in the sixtieth minute. That is the data the league table never records, and in a long and noisy annual season, it is the only thing that tells the truth about a club before the headline tells it for them.