Worst transfers of the window, ranked: From Fernandez to Tielemans

It's easier to identify a bad transfer than a good one. For starters, most transfers fail. If you pick out 20 random transfers from this summer and say, "I bet that guy doesn't even play half of the available minutes with his new team," you are going to be correct more often than not. Most soccer clubs are dumb organizations. And I'm not saying the people involved are dumb! But rather, the way these clubs are structured makes it so very few decisions get made with any kind of rigorous, objective criteria. Almost no one is talking in terms of the goals or points a new player might add and why. It's mostly still just vibes-based decision-making. That's absurd when you're spending upward of $100 million on fees that just allow you to then sign the player to an expensive long-term contract. And it's absurd when you think about all of the different ways a given transfer could fail: The player isn't as good as you think, the coach doesn't like him, he gets hurt, his family can't adjust to the new country, you didn't properly account for league-translation issues, etc. But even beyond all of that, there's a simple heuristic to follow. A bad transfer requires only one team to make a bad decision. For a good transfer to happen -- a talented player acquired for way less than it seems like it should be -- it requires every other team to pass on the player. That means overpaying for players who don't play is way more likely than underpaying for players who become stars. So, rather than trying to identify the best moves from this summer, I'm going to do what I did last summer and identify the transfers from this transfer window that are most likely to fail. Here are the top 11 from across Europe. How we're identifying bad transfers For the third year in a row, we're using the basic projection system that NFL analyst Kevin Cole helped me create. We took four years' worth of top-end transfers and looked at what the player's estimated value on the site Transfermarkt was a year after the transfer. What we found, as I outlined last year: To varying degrees, a lower age, a lower transfer fee and a higher market value at the time of the transfer made it more likely there was an increase in value after a year. Then, we can take those factors and create a formula to predict an increase or decrease in value for any big transfer. In other words: A year from now, is a player's market value likely to be higher or lower than their transfer fee, and by how much? This is a basic analysis, and it doesn't account for the extra costs of player wages, which can vary significantly. Plus, we're using estimated numbers from Transfermarkt to come up with the market values and the fees, which often contain add-ons. However, studies have found that Transfermarkt values tend to be pretty close to true player value on average, and it also lets us harness the power of the wisdom of the crowds: The market values on the site are a pretty good representation of what the world thinks of a player. At least based on our analysis, when teams have paid significantly more than the Transfermarkt value for a player, those moves have tended to not work out. Last year's top -- or bottom -- 11 included first-season Newcastle flops Nick Woltemade and Yoane Wissa, the player they were replacing in Liverpool's Alexander Isak, and Arsenal's Eberechi Eze, who played just 1,812 minutes. Looking at this summer, though, one thing immediately stands out: The price has gone up. Last year, just four players projected to have a year-out value decrease of 30% or more. This season, there are 11 -- in Europe alone. I've excluded deals from Saudi clubs, so we start in North London. 11. Mateus Fernandes, midfielder, West Ham to Tottenham Age: 21 Fee: €99 million Market value: €50 million Projected negative differential: 30.3% This model is really forgiving when you sign a young player. For example, Manchester City signed 18-year-old Ayyoub Bouaddi from Lille for €95 million -- €15 million more than his Transfermarkt estimate. But it still sees his value a year from now as increasing 25% beyond the fee City paid. So why not the same for Fernandes? Bouaddi is widely considered to be the next great midfielder, while Fernandes was pretty good for a really bad team last season. More than that, though: Bouaddi is three years younger. 10. Ezri Konsa, center back, Aston Villa to Arsenal Age: 28 Fee: €59.6 million Market value: €45 million Projected negative differential: 33.06% It's no surprise that the people who own the Los Angeles Rams also own Arsenal. The NFL team is famous for trading away draft picks for established stars. After they won the Super Bowl, Rams general manager Les Snead wore a shirt that said, "F--- them picks." They do whatever they can to win now. They'll worry about the future whenever it comes. After years of patient planning and investing in young talent, the Gunners entered their version of that era last summer. Their three biggest signings were in the latter years of their primes. And with William Saliba out for an undetermined amount of time with a back injury this season, they made Konsa the most expensive 28-year-old center back in the history of the sport. In case you can't tell: They want to win, right now. 9. Rodri, midfielder, Manchester City to Barcelona Age: 30 Fee: €60 million Market value: €55 million Projected negative differential: 34.1% It might seem like Bouaddi and Yan Diomande were this summer's classic World Cup overpays: players someone got excited about because of a couple of good games against subpar opposition. But both guys were already on superstar paths at the club level. Instead, allow me to make the case that Rodri, perhaps the best midfielder of his generation, was this summer's World Cup overpay. He tore his ACL two years ago, and no one thought he was close to back to his best for Man City last season, as he played only about 1,600 minutes. Then, in the much slower and much easier World Cup environment, he looked great and the soccer world seemed to collectively forget the past two years. He's 30, and he's about to play in one of the most physically demanding systems in Europe. 8. Randal Kolo Muani, forward, Paris Saint-Germain to Juventus Age: 27 Fee: €41.2 million Market value: €20 million Projected negative differential: 34.52% On loan at Tottenham last season, he scored one goal and added one assist across nearly 1,700 minutes. Yes, he is an attacker. 7. Youri Tielemans, midfielder, Aston Villa to Manchester United Age: 29 Fee: €41 million Market value: €30 million Projected negative differential: 34.9% It seems like he's the Casemiro replacement, but he's not -- or at least he shouldn't be. Casemiro provided most of his on-ball value at Manchester United by making runs into the penalty area, while Tielemans is your classic touch-the-ball-as-much-as-you-can ball progressor. They're not remotely similar players. The concern with Tielemans -- beyond his age -- is, well, how much is he going to play? Sandwiched around a mostly healthy 2024-25, he missed 11 games with injury in 2023-24 and 23 games last season. The moves for the much-younger midfielders Andrey Santos and Carlos Baleba seem to suggest that United are at least aware that Tielemans might miss time this season and might soon decline, like most players his age do. 6. Sandro Tonali, midfielder, Newcastle to Tottenham Age: 26 Fee: €108 million Market value: €80 million Projected negative differential: 36.2% Tonali has always been misunderstood. I guess because he has long hair and is Italian and did play for AC Milan, he was compared to Andrea Pirlo at the beginning of his career. Pirlo, though, is one of the greatest long-range creative passers the sport has ever seen. Tonali completed 13 passes last season that broke the opposition defensive line -- tied for 139th most in the Premier League. He is an above-average midfielder who does a little bit of everything. Tottenham paid like he's a ready-made superstar. 5. Bruno Guimarães, midfielder, Newcastle to Arsenal Age: 28 Fee: €87.5 million Market value: €70 million Projected negative differential: 36.25% Much of what I'm about to say can be explained by the new squad-cost-ratio rules in the Premier League that effectively limit clubs from spending a percentage of their revenue on player costs every season, but that's not going to stop me from pointing it out. It's supposed to go like this: If Americans buy your team, they are going to cut costs and do whatever they can to raise revenues because Americans own sports teams because they want to make money. If a sovereign wealth fund buys your team, it's going to spend as much as the rules will allow, if not beyond, because these countries buy teams because they want the reputational benefits that come with winning lots of soccer games. Put another way: Newcastle (owned by Saudi Arabia's Public Investment Fund) are supposed to be the team acquiring Guimarães from Arsenal (owned by Stan Kroenke) -- not the other way around. 4. Sávio, winger, Manchester City to Tottenham Age: 22 Fee: €87.7 million Market value: €35 million Projected negative differential: 40.17% If you're still wondering why this model loves when teams sign young players, then look no further than the artist formerly known as Savinho. Manchester City acquired him from their sister club Troyes for €25 million, a number that was clearly negotiated in good faith after the Brazilian had a breakout season at age 19 while on loan at another City Football Group club, Girona. In his two seasons with City, Sávio scored two Premier League goals and started 28 total Premier League matches. Despite things going close to as poorly as possible, another team was willing to bet nearly €90 million on his potential. I really liked Sávio as a post-hyper-sleeper-type signing, and the model would have, too -- but Tottenham paid like he's an established star. 3. Morgan Rogers, attacking midfielder, Aston Villa to Chelsea Age: 23 Fee: €138 million Market value: €90 million Projected negative differential: 40.8% Let's just accept the premise that Rogers can break math or briefly turn into Lionel Messi when his foot touches the ball. He has scored 18 goals from 13.41 expected goals and generated 16 assists from 9.6 expected over the last two seasons. Even with that overperformance, he has averaged 0.49 non-penalty goals+assists per 90 minutes across his Villa career -- very good but not close to most-expensive-British-player-ever territory. This is one of those deals that, because of the size, has almost no upside. Chelsea paid for the absolute best-case scenario. To turn it around: Rogers is the third Villa player on this list. And Al Hilal's €58.4 million move for 30-year-old Ollie Watkins would have landed at No. 1. Villa have, undeniably, done a fantastic job in making the inefficiencies of the market work in their favor. But the question for Villa is: Did they have to do it all at once? On Monday, Villa became the second team in the past 20 years to fail to have a shot on target in their first two matches of a Premier League season. The other was Swansea, who got relegated that season. 2. Gonçalo Ramos, forward, Paris Saint-Germain to AC Milan Age: 25 Fee: €30 million Market value: €74 million Projected negative differential: 48% If I'm investing that much money into a PSG castoff who played a ton of sub minutes in Ligue 1, I want his numbers to look like Messi's, not João Pedro's. Much like Sávio, Ramos would have been a smart buy-low who instead became a buy-high. 1. Enzo Fernández, midfielder, Chelsea to Manchester City Age: 25 Fee: €145 million Market value: €100 million Projected negative differential: 49.1% The founders of the stats app Futi created a possession-value model that estimates the value a player provides his team based on how much his actions increase his team's chances of scoring and decrease his team's chances of conceding. The number then gets normalized based on opponent, position and overall team performance when the player is on the field. In his three full Chelsea seasons, Fernández ranked 15th in 2023-24, seventh in 2024-25 and ninth in 2025-26 among players in his position who played at least 2,000 minutes in the Premier League. He is a very good Premier League player who just became the most expensive Premier League player of all time.