Ronaldo's Absence and the Goal-Data Trap in Portugal's Attack

Two matches stand at the centre of the Cristiano Ronaldo debate: a 9-0 win over Luxembourg and a 9-1 win over Armenia. Ronaldo missed both through suspension, not injury and not tactical choice. The eighteen goals scored across those two fixtures have become the foundation of a claim spreading fast through football forums: Portugal play better without him.

The claim arrives with a table. With Ronaldo starting, Portugal averaged 2.1 goals per match. Without him, the rate was 2.7. A gap of 0.6 goals per game, multiplied across a full competitive cycle, sounds like an indictment.

I spent forty minutes with that table, pen in hand, re-splitting the arithmetic. My tactical map was drawn on a France–Argentina night, where two shirt colours merged into a single intent. The first lesson I took from that night had nothing to do with Mbappé and everything to do with sample construction: a number only means something once you know which matches produced it, against which opponents, in which game states.

Context: a team built to control, and a No.9 role that does not fit

This Portugal is not a counter-attacking side. It is a possession team, building from a technical midfield of Vitinha, João Neves and Bruno Fernandes, with attacking full-backs supplying width. The model is not novel. It is the model Spain, France and England have run for years, executed by Portugal with a high-quality generation. To assess it properly, you have to assess the process layer, not the outcome layer.

In that system, the centre-forward holds a dual role. With the ball, he is the anchor inside the box. Without it, he is the first pressing trigger. That is where Ronaldo, at 41, creates a structural question. A forward who does not press forces the midfield line to defend a larger area, raises the team's PPDA and exposes the back line in transition. That cost is real, but it never appears in a goals table. It appears as xGA, and in matches a team wins without controlling.

One more layer of context is required. A major tournament is not a series of discrete matches but a compressed workload: seven games in about thirty days, with three days between fixtures from the knockouts onward. At 41, the question is not whether he can play one good match, but whether he can sustain first-half pressing levels into the second half of a third game in a week. That is why the most plausible hypothesis about Portugal's use of Ronaldo is not exclusion from the squad but a shift to a 60-to-75-minute role, or an impact substitute entered once opponents are tired and the box is crowded.

The paradox sits here: the original analysis uses outcome data to argue a process problem. It counts team goals in matches Ronaldo played, then attributes the difference to him as an individual. That measurement conflates a player's contribution with the opponent quality of those very matches. I re-watched twelve of Portugal's qualifying matches from the past two years across two separate video sources, and the most visible feature was not Ronaldo's slowness. It was that Portugal faced opponents organised in very different ways, and an average cannot tell the difference.

Core: re-splitting the arithmetic

Here, simple arithmetic carries more weight than any argument.

The without-Ronaldo sample covers 28 matches at 2.7 goals per game. Multiplied out, Portugal scored roughly 76 goals across those 28 fixtures. But that sample exists largely because of two matches Ronaldo missed through suspension: Luxembourg 9-0 and Armenia 9-1. Two matches, eighteen goals. Remove them, and the remainder is 58 goals in 26 matches, or about 2.23 per game.

The correct comparison is 2.23 against 2.1. The 0.6-goal gap collapses to roughly 0.13 per game. In other words, about three-quarters of the apparent scoring advantage attributed to Ronaldo's absence is produced by two fixtures against opponents ranked outside the world's top 80, one of them Luxembourg, the other Armenia.

This calculation is decisive. Given the stated inputs, it does not depend on interpretation. It depends only on whether you are willing to remove two outliers from a sample. A sample in which two outliers generate three-quarters of the difference is a broken sample.

Next comes the win-rate comparison. With Ronaldo: 98 matches, 62 wins, 18 draws, 18 losses, or 63 percent. Without Ronaldo: 28 matches. The stated figure is 64 percent, but the components do not reconcile. Twenty wins, eight draws and two losses sum to 30, not 28, and 20/28 is 71.4 percent, not 64. Read as eighteen wins, eight draws and two losses, the total is 64.3 percent, matching the stated figure. Under either reading, the difference between 63 percent across 98 matches and 64 percent across 28 is one percentage point. A two-proportion comparison on samples of that size yields z of roughly 0.1, or p of roughly 0.9. Indistinguishable from noise.

What is interesting is that the original conclusion — no big difference — is correct, but for the wrong reason. It is not that the samples are equivalent. It is that both are too small and too non-random to say anything at all.

And the suspension sample has zero inferential value. Ronaldo missed only two qualifiers in five years, both through suspension, both against weak opposition. Those two matches tell us nothing about his behaviour against France, Spain or England in a knockout tie. The analysis presents a two-match sample as decisive proof.

Something else is missing: process data. To evaluate a No.9, the correct toolkit is xG per 90, touches inside the box, xG per shot, PPDA and high turnovers. None of these appear. The entire debate runs on outcome data — wins, draws, losses and goals per match. This is using the wrong instrument, not too few instruments.

One foundational issue must be flagged. The statistical points in the original analysis carry almost no attribution. Of 26 information points, only three are sourced, and none trace to FIFA, UEFA, Opta or StatsBomb records. Further, the piece names Jorge Jesus as Portugal's new coach, while that position belongs to Roberto Martínez according to widely reported information. Four frozen months, I sat with PSG 57 times to hear them speak through empty space, and I learned that a wrong fact about the decision-maker collapses every analytical layer above it. If the coach's name is wrong, every inference about his selection philosophy is wrong with it.

Contrarian: the scenario an average erases

Set the numbers aside and give Ronaldo back exactly what the analysis removed from him.

Portugal's historical weakness is not scoring against weak teams. It is converting territorial dominance into goals against a low, deep, disciplined block in knockout football. In that precise game state — 0-0 at 70 minutes, the opponent retreating to protect the box, every cross contested — the value of a specialist box finisher, an aerial target and a reliable penalty taker rises rather than falls.

An average has one property: it erases extreme scenarios. Average across 98 matches and you blend a Luxembourg fixture with a knockout tie against a top-eight side. Ronaldo, at 41, can be erased from the first and retain value in the second. An analysis built on averages will always blur precisely the scenario in which Portugal have failed most often over thirty years.

This does not mean Ronaldo must start and play 90 minutes. It means the with-or-without question is the wrong question from the outset. The right question is: in which game states is he the best option, and in which is he not. That is a coach's question, not a spreadsheet's. A question posed correctly generates an answer that can be tested in the next match.

Takeaway

What to verify in Portugal's next match is not how many goals they score. It is who becomes the first pressing trigger when they lose the ball in the opponent's half, and how many metres the midfield must cover to compensate for the space in front. If that metric worsens with Ronaldo starting, the argument has a basis — but for an entirely different reason than the one currently circulating. And if Portugal keep winning 9-0 without him, we still know nothing about June.