From Anfield 2026 to an Empty Dataset: Football Still Needs Human Eyes

On 7 May 2026, at Anfield, Liverpool beat Barcelona 4-0 after having lost the first leg of the Champions League semi-final 0-3. Divock Origi scored the fourth goal in the 79th minute, from a quickly taken corner while the Barcelona defence was still adjusting its positions. Before kick-off, probability models gave Barcelona more than a 90 percent chance of progressing. No model wrote into its spreadsheet a variable called the collective loss of concentration of a team that believed the job was done.

The next morning, I opened my own analysis. Nine data groups. Seven comparison tables. Three scenario models. Every cell was filled. I sat looking at it for nearly two hours and could not write a single line, because no cell in it explained what had just happened. That was the morning I understood that a complete analysis can be hollow.

Football today runs on a nine-dimensional system of the same kind. The technical and tactical dimension, with xG - expected goals, measuring chance quality - and PPDA, the passes allowed per defensive action, a measure of pressing intensity. The club finance and transfer market dimension, using Transfermarkt valuations as a reference point. The results and public-opinion cycle dimension. The league landscape and team positioning dimension. The rules and governance dimension, where UEFA's Financial Fair Play and the Premier League's Profit and Sustainability Rules sit. The dressing room. Risk. Media narrative. And the transmission chain of the entire football industry.

That framework has no shortage of slots. It has a slot for everything, including a slot for emptiness.

On 6 February 2026, the Premier League announced 115 charges of Profit and Sustainability Rules breaches against Manchester City, spanning the 2026-2026 to 2026-2026 seasons. On 17 November 2026, Everton were deducted 10 points - the heaviest sanction ever applied to a Premier League club. On 26 February 2026, after appeal, the penalty was reduced to 6 points. On 18 March 2026, Nottingham Forest were deducted 4 points. Throughout that period, every analysis had enough cells to fill: transfer fees, wage-to-revenue ratios, contract structures, durations, add-on clauses, sell-on clauses. The spreadsheet was packed. The real question - why a system allows the gap between paperwork and grass to grow so wide - had no cell to answer it.

The mechanism producing hollow completeness is simple. An analytical framework is designed to always produce an output, so it can be formally complete while substantively empty. It never says it does not know. It says there is insufficient information. Formally, those two statements are far apart. Substantively, they are identical. And once a framework has committed to always producing an output, filling the gap becomes a professional reflex.

I have seen that reflex across many years of watching matches in both Europe and Asia. Based on my own experience of tracking matches, I have noticed a rule: the more formally complete an analysis is, the more likely it is to miss what is actually happening. When every indicator sits inside a safe range, nobody goes looking for the anomaly.

The transfer market is where this mechanism is most visible. Every window, thousands of valuations are updated, hundreds of contract structures are dissected, dozens of scenarios are drawn. Every number has a source, every source has a date. But almost nobody can answer why a club decided to spend that amount on a 19-year-old on a Friday afternoon. The transfer market is where dreams are priced in black and white, and also where those prices are never verified again.

In the 2026-2026 season, Leicester City won the Premier League. Before the season, many bookmakers listed that scenario at 5000 to 1. Forecasting models placed Leicester near the bottom, based on squad value, previous-season performance and squad depth. Jamie Vardy scored 24 Premier League goals that season, at the age of 29, after having played in semi-professional leagues. The spreadsheet was not wrong arithmetically. It simply had no cell for one thing: a group that believed in itself for ten straight months.

Euro 2026 unfolded in a similar way. Greece won, with pre-tournament odds among the lowest. No xG model predicted that a low-block team, counter-attacking through set pieces, could go the whole way. The ball does not lie, and it knows how to tell a story in a way a spreadsheet cannot.

The paradox also lies in technology. VAR was introduced in the Premier League from the 2026-2026 season with the aim of reducing refereeing errors. Published data after each season show the rate of correct decisions rising. But the volume of public controversy rose with it. More data means more disagreement, because data cannot erase the difference in how humans read the same frame at 50 frames per second.

The problem is that we ask data questions it has no capacity to answer. xG measures the quality of a chance. It cannot measure the hesitation of a defender in the 88th minute, when he knows that a single mis-touch will dissolve an entire season. No metric has a name for that moment. Barcelona's 6-1 win over PSG at Camp Nou on 8 March 2026, with three goals in the final seven minutes - Sergi Roberto's winner in the 90th plus 5th - is another example of the same limit. Before the match, the models gave PSG a very high chance of progressing. After the match, people still used those same models to explain what had happened.

I am not writing this to deny data. I earn my living reading numbers, and I believe in them. But an analysis consisting entirely of the words insufficient information is the most honest analysis a system can produce under conditions where it is forbidden to invent. The danger lies in the other form: the analysis that looks complete, where every cell has content, and the reader can no longer tell what is measured data and what is speculation written in the same confident voice.

My profession has an unspoken principle, written in no code of conduct: when you do not know, leave the gap. But the profession also faces reverse pressure, from publishing rhythms and from readers who want answers before the referee blows the whistle. That pressure turns the gap into something nobody wants to see. So we fill it, with sentences that are plausible, readable, and unverifiable.

When the stadium was empty, I learned to listen to the echo. Chinese football froze in March 2026, and for many weeks the only thing left in my work was sound that no metric could capture: wind across the stands, the footsteps of a groundskeeper, my own breathing. Those weeks taught me that a sporting event always leaves behind a part that cannot be digitised, and that part is usually the part that lasts longest. A match is never just 90 minutes. It is an entire life compressed.

What I want to leave behind is not a warning against technology. Nine-dimensional frameworks help a great deal: they detect patterns the eye misses, they preserve traces for debate, and they teach sports writers to interrogate themselves harder before asserting. What I want to argue for is a professional attitude. An analysis has value when it dares to leave a cell empty, when it dares to say that the most important thing last night lay outside its own framework. That honesty does not weaken commentary. It is the condition for commentary to survive the seasons ahead, when every metric has been standardised and every match has been cut into thousands of frames.

At 61, I still believe the most beautiful match is the one that has never been played - no scoreboard, no spreadsheet, just two teams walking onto grass and something happening that nobody predicted. When the next generation of commentators starts writing, what I want them to carry is not a new analysis tool. It is the instinct to force themselves to ask, after every match, whether what they have just written is a discovery, or merely a gap filled with words.