From Rimario's 0.32 xG to a champion's PPDA: reading the transfer market with data

In 14 appearances for CLB Hải Phòng in the 2026 season, Rimario Gordon posted an expected goals (xG) figure of 0.32 per match, the lowest among the ten attacking foreign players in V.League that year. I built the table, printed it, carried it into the press room, and predicted the striker would score around five goals. A senior editor waved it away: "What does a woman know about strikers?" By the end of the season, Rimario had scored exactly five goals, been released, and the room had gone quiet. That story is not about a spreadsheet winning. It is the reason I start every analysis with a data source instead of an instinct.

V.League is a market where a player's value is set by three things: goals, age, and the memory of whoever is paying. Rimario arrived at Hải Phòng for a fee of roughly 250,000 USD, no small sum for a port-city club. But goals are a lagging indicator. They appear only after a chance has already been created. To read a striker properly, you have to look at what happens before the goal: shot volume, the quality of the shooting position, the conversion rate. xG does exactly that by assigning every shot a probability of becoming a goal based on distance, angle, the type of pass and defensive pressure. A striker with 0.32 xG per match generates, from chances of that quality, less than a third of a goal per game. Multiplied across 30 rounds, the theoretical ceiling sits near ten goals, and only if conversion runs exactly to expectation. Rimario did not reach it.

What I learned was not in the final number. It was in the fact that a transfer market can run on belief and ignore a leading indicator. V.League had no standardised database covering every player at the time; scouting reports were mostly impressionistic, padded with a few raw figures such as goals and appearances. A striker who scored 12 goals in a lower division will always sell more easily than one who scored 8 in a higher division, even when the quality of their chances was completely different. The buyer looks at the price table; the analyst has to look at the movement table.

A night in Hải Phòng taught me this: people look at the price table, I look at the movement table. A price is a single slice. A movement table is the trend of data over time, and the trend says more than the absolute level.

Take a simple defensive metric. PPDA, the number of passes an opponent is allowed before each defensive action, measures how actively a team presses. The lower the PPDA, the more aggressive the press. A team that holds a PPDA of 11.4 and then drops to 9.8 has changed the nature of its football, not merely its results. If a team's PPDA has fallen steadily over its last three matches while its points total stands still, that is an early signal the league table has not yet priced in.

I tested this at a larger scale. In the summer of 2026, when the Bundesliga returned to empty stadiums, I compared 26 rounds with crowds against 9 without. Home advantage fell from 55% of matches won to 43%, a drop of 15.3%. Yellow cards rose 22%. Away teams' PPDA fell from 11.4 to 9.8, meaning visiting sides pressed harder once the crowd no longer weighed on them. With the stands empty, I realised I had been failing to count a variable: emotion does not sit inside a spreadsheet.

At Euro 2026 I was wrong in a different way. I picked Belgium to win because they had the tournament's highest total xG, and I ignored Italy's PPDA of 8.7, the lowest of all 24 teams. Italy won by pressing actively rather than by the power of their attack. I then built a pressing dataset across 14 major competitions and found that every European champion from 2026 onward had kept a PPDA below 10. One dimension of data always misleads; two dimensions begin to talk to each other.

A serious scouting dataset for V.League needs at least five axes: finishing output (xG, xGOT), chance creation (key passes, xA), defensive workload (pressing, ball recoveries), sustained physical capacity (high-intensity running distance in the second half), and season-to-season stability, meaning the variance of the metrics above. The fifth axis is the most neglected. A player with 0.4 xG per match who swings between 0.05 and 0.9 from round to round is a gamble; a player who holds xG between 0.3 and 0.4 is an asset. The market pays for peaks, but a club lives on stability.

Applied to the transfer market, the principle holds. A striker judged only by goals is a striker read with half the data. You need both xG and xGOT, both the number of touches inside the box and the share of aerial duels won, both high-intensity running distance to know whether a player can sustain pressure for 90 minutes or only flares for 60.

A player's true transfer value lies in the gap between leading and lagging indicators; the club that can read that gap buys talent cheap, the club that cannot pays for memory.

Rimario is the case of a gap that went unread. He was not a bad player; he was a player placed against the wrong expectations. A striker with 0.32 xG needs a system that creates better chances, or needs to be used as a second spearhead rather than the main centre-forward. Putting him in the role of carrying the goals and then concluding he failed is misreading the map.

But stopping there would repeat the very mistake I made in 2026. Germany left the 2026 World Cup; every model has its day of bankruptcy, only historical data remains. I once predicted Germany would reach the semi-finals on the strength of 67% possession, 2.1 xG and 91% pass accuracy. Germany lost their opener to Mexico and were eliminated by South Korea. The spreadsheet was not wrong; the context had changed. The temperature of the pitch, Mexico's high press, the psychology of the reigning champion, none of that lives in any column of a spreadsheet.

Correlation is not causation. A striker with low xG may have a system that does not serve him, not a lack of quality. A team with a low PPDA may press well, or may have gone behind and been forced to push up. The same number, two explanations, and the poor analyst is the one who picks the explanation that fits the bias they already hold.

There is one variable I always undercount: emotion. My numbers do not need applause. They need to be right; time is the referee. But a player whose hands shake in the 90th minute, a stadium so quiet a defender can hear his own breathing, a coach who has lost the dressing room after a run of defeats, a spreadsheet cannot record any of it. Data is a map, not the territory. Even a good map-reader has to accept that the road out there may be flooded.

Three in the morning, the market is asleep. That is when the numbers are at their most clear-headed. I still sit with my dataset, not to find a striker who scores 20, but to find the signal for the next round: a player whose xG is rising steadily while his goals stand still, a defence whose PPDA is falling while the points have not yet moved. Those signals are where the market has not yet priced in. Whoever reads them first buys cheap.

And the question I leave for this season is not who will win the title, but this: in the upcoming transfer list, how many names are being valued by the goals they scored before rather than by the chances they will create next?