The Two-Tier Goalkeeper Market: Paying for Feet, Forgetting Hands

On 11 July 2026, at Wembley, Gianluigi Donnarumma saved two penalties in the Euro 2026 final against England and was named Player of the Tournament. On 14 July 2026 he signed a five-year contract with Paris Saint-Germain. The transfer fee was zero, because his AC Milan deal had expired on 30 June. That same summer, Kepa Arrizabalaga sat on the bench at Stamford Bridge. Chelsea had paid 71.6 million euros to sign him from Athletic Bilbao in August 2026, then a world record for a goalkeeper. Three years after that signature, he was the number two behind Édouard Mendy. I wrote both lines on the same notebook page, four boxes apart. During this major tournament season I have reopened that page often, whenever someone asks me why a goalkeeper who shines in a short tournament does not automatically become expensive. I heard that boy's name from the B stand before any big club knocked on the door. For goalkeepers the sentence works differently. The name rises in a single tournament; the price is set somewhere else entirely. Two price lists, side by side Before 2026, European goalkeeper valuation was simple: saves made, save percentage, clean sheets. All three depend on how many shots the defence in front allows, which means they measure the collective more than the individual. The market used them for nearly two decades anyway. In June 2026, Manchester City paid Benfica 40 million euros for Ederson. That fee sat among the highest ever for the position, and the reason was not shot-stopping. Ederson was bought for the 60-metre pass, for standing on the edge of the box, for the first phase of build-up. Pep Guardiola paid 40 million for a fifth defender who plays with his hands. That break point created a new category on the scout's desk. From 2026/18, goalkeeper profiles at big European clubs were split into two columns: distribution and shot-stopping. The two rarely score high in the same person, and when forced to choose, most big clubs pick the first column. That sounds reasonable until you look at the price structure. Distribution can be seen with the naked eye across three matches. Shot-stopping needs at least a season and a half to separate from noise. A market that prices what is easy to see will always pay for what is easy to see. What the numbers say The metric I use most when judging goalkeepers is the gap between goals conceded and post-shot expected goals, the goals prevented figure against xGOT. It answers one question: on shots that were on target, did this keeper stop more or fewer than average. It strips out most of the effect of the defence in front and does not depend on whether the team attacks or defends. When I went back through goalkeepers transferred for 25 million euros or more between 2026 and 2026, a pattern emerged. The group with a positive goals-prevented figure in two consecutive seasons before the move largely kept the number one shirt at the new club. The group bought mainly on distribution had a markedly lower retention rate. Alisson Becker is the cleanest example of the performance-first group. Liverpool paid Roma 62.5 million euros in July 2026, plus around 10 million in add-ons. At Roma he had posted two straight seasons of positive goals prevented against xGOT inside a defence that was not especially solid. Liverpool bought proven performance, and his distribution was merely decent. André Onana sits on the other side. In July 2026, Manchester United paid Inter Milan around 50 million euros. The trait mentioned most often was short passing under pressure and playing out of the press with his feet. His shot-stopping at Inter was good, but his time in a high-pressure environment was short relative to the fee. David Raya went the other way. Arsenal loaned him from Brentford in summer 2026 and made the move permanent for around 27 million pounds in July 2026. That fee sits well below the going rate for the group above, while his goals-prevented figure and clean-sheet rate in the Premier League were among the best. His distribution was good enough, never spectacular. Arsenal paid for performance and received performance. What stands out is that both groups were priced at the table by the same two measures: top-flight appearances and age. Goals prevented appears in the appendix of the file, after the fee has already been agreed internally. Financial rules push the same way. Since July 2026, UEFA caps the amortisation of transfer fees at five years rather than spreading them across the full contract length. For a 22-year-old keeper on a six-year deal, the annual charge is still lower than for a 30-year-old on a three-year deal at the same fee. Big clubs therefore have an extra financial reason to buy young, whatever the shot-stopping numbers say. Seven matches and a shootout A World Cup or a Euro gives a first-choice goalkeeper at most seven matches. Across those seven, the number of on-target shots faced usually lands between 15 and 25. A shootout adds four to eight attempts from around 11 metres, under conditions nothing like open play. At 25 shots, the standard deviation in save percentage is wide enough that an average keeper can finish a tournament at 85 percent and a top keeper can finish at 55 percent. Neither result says much about the following three seasons. This explains why major tournaments keep producing goalkeeper stories and very few big transfers to match. Euro 2026 lifted Donnarumma to the top, yet he left on a free because his contract had expired, not because any club was ready to pay 70 million euros for him at that moment. Four years later, when he moved from Paris Saint-Germain to Manchester City, the reported fee sat below 35 million euros for a keeper who had just won the Champions League. Young goalkeepers on high fees take a different road. Bart Verbruggen left Anderlecht for Brighton in July 2026 for around 20 million euros at the age of 20. Liverpool agreed a deal for Giorgi Mamardashvili from Valencia in August 2026 for around 35 million euros, at 23. Today's young player is tomorrow's expensive player, I wrote that in 2026, and it holds most strongly for goalkeepers because the long development curve keeps supply thin. Both were priced on the distribution column more than on goals prevented. Verbruggen rose at Anderlecht on his ability to play with his feet in a possession system. Mamardashvili rose at Valencia on his reflexes, but a 35 million euro fee at 23 reflects an expectation he will add distribution over the next three seasons. Contract mechanics push prices too. In Spain and Portugal, the release clause is mostly a negotiating tool rather than a wall. Diogo Costa's clause at Porto has been reported in the Portuguese press at 75 million euros, set well above his actual market value at the time of signing. After every major tournament, agents press clubs to re-sign a player with a fresh release clause, turning a seven-match run into a floor for three years of talks. What clubs leave out The biggest blind spot in the goalkeeper market sits in the 28 to 31 age bracket. One bad season can wipe 10 to 15 million euros off a keeper in that group, when the cause is usually outside his control: a rebuilt defence, a changed marking scheme, or a wrist injury that dulls reflexes for a few months. Emiliano Martínez is the case to remember. In September 2026, Aston Villa bought him from Arsenal for around 20 million pounds at 28, after nearly a decade mostly spent on loan. Over the next four seasons he was among the Premier League's best on goals prevented and won the 2026 World Cup with Argentina. In the other direction, summer 2026 saw Paris Saint-Germain spend on a 23-year-old leaving Lille for a reported fee near 40 million euros, right after the club won the Champions League. That is an investment aimed at the next ten years rather than the next season, and this way of pricing is becoming the norm. That mismatch creates exploitable space. Mid-tier clubs in Serie A and the Bundesliga are starting to realise they can buy a 29-year-old keeper who was marked down after one hard season for a third of the fee commanded by a 22-year-old with fewer than 60 professional matches. Over the first two seasons, the performance gap usually favours the 29-year-old. The hardest part of a goalkeeper deal is not the fee. It is whether the buying club accepts adjusting its way of playing to suit him. A contract with a signature is a transfer; everything else is a rumour. The next domino I am on the training ground at six in the morning, because sporting directors never answer emails at night. That habit formed after I asked a scout about a 27-year-old Danish keeper playing in the Dutch league and was told his data department had never put the position on the shortlist. With goalkeepers, verification moves slower than at any other position because the data points per season are fewer. Over the same number of matches, a midfielder generates more than two thousand touches; a goalkeeper generates a few hundred. Anyone wanting to judge the position on data has to accept a three-season horizon, and most clubs do not have that patience when the head coach's chair empties every 14 months. The goalkeeper market will split further into two tiers over the next two to three seasons. The top tier will keep paying for distribution under the age of 24, and the next record fee will most likely go to a 22-year-old with more than 60 senior starts rather than to the hero of a World Cup. The lower tier will hunt the 28 to 31 group coming off a bad season, buy with modest cash and collect the shot-stopping. The gap between the two tiers is where the return sits, and that is where I will spend most of my time watching in the next window.