Daddy, Where Do Transfer Fees Come From?

And shouldn’t Thiago’s be higher—like, a lot higher—than Wijnaldum’s?

Welcome to space space space, a letter about making soccer make sense. Sign up here to get it delivered to your inbox.

It’s been precisely four weeks since Bayern Munich folded Barcelona’s tattered pride into a flag-code triangle for their trophy case and less than three since they plopped the actual trophy upside down on their heads. Which is weird, in my opinion, because this whole Thiago-Wijnaldum transfer drama didn’t really take off till after that and yet we’ve been talking about it for at least 300 years.

In case you’ve just woken up from a very long nap: Bayern, Liverpool, and Barcelona are locked in one of those third-act Sergio Leone standoffs where eyes dart and trigger fingers hover and horns swell and time itself keels over from the stress. When the smoke clears, the idea is that Thiago Alcantâra will move from Munich to Liverpool while Gini Wijnaldum heads south to Barcelona. Bayern will be fiscally responsible while keeping their squad young, Liverpool will snap up a world class player who’s mysteriously undervalued, and Barcelona will overpay for yet another aging midfielder who doesn’t suit their style—ho hum, your typical offseason.

But even if you know these clubs’ recent histories, you may, as a person who has had occasion to watch both Thiago and Wijnaldum play soccer, find it a little disconcerting that press reports have pegged their pricetags in the same €25 millionish range. Shouldn’t players with comparable transfer fees be, you know, comparably good?

In a word: lol.

My buddy Tiotal Football, who’s been writing the Great American Newsletter about soccer front office theory, likes to joke about his “Where do transfer fees come from?” article sitting in drafts that’s just a picture of a stork. More seriously, he likes to remind us that players can’t be “bought” and “sold,” because they’re not bags of potatoes, and transfer fees aren’t actually about pricing soccer skill—they’re deals done to terminate one employment contract and negotiate a new one, with all the non-soccer complications that involves.

Maybe Thiago’s always dreamed of living on the banks of the Mersey and refuses to play anywhere else. Maybe he’s demanding outrageous wages, or has a rider that calls for a bowl of M&Ms before every game with the brown ones picked out. There are lots of unknowns that have nothing to do with sports but everything to do with how much clubs are willing to pay each other to try to hire him. The length of a player’s existing contract is a good example, as Liverpool’s already pulled the standard negotiating ploy of wondering within earshot of a reporter why they shouldn’t wait till January and sign Thiago on a free for next summer.

Clubs have their own constraints, starting with money. A lot of teams could probably scrounge up €25 million for the right signing, but that market dwindles when you factor in senior players’ salary and bonuses and the low probability of recouping any costs on a 29-year-old midfielder by selling him on in a few years. Transfer budgets are capped not only by owners’ eagerness to spend but also, at least for some clubs, by Financial Fair Play rules meant to inhibit them from spending beyond their means. Throw in a pandemic-induced financial shock and all the weirdness that’s wrought in the transfer market and honestly who knows who’s got what kind of money right now aside from Roman Abramovich, who has all of it.

And yet transfer fees do have something to do with soccer, don’t they? We’re aware that the boring realities of contracts exist but we still expect some sort of relationship between the euro figure in the tabloids and the player on the field. There’s a reason for that. Plenty of folks have shown that you can do a pretty good job predicting transfer fees even if you’ve got no information about player contracts or club balance sheets.

One fun entry in this line of research is a blog post by Shubham Maurya that explored the relationship between Fantasy Premier League prices and Transfermarkt values. “Market value,” he concluded, “is a function of ability, position and age.” Makes sense, right? But Maurya went on to fit a pretty good transfer value model using no real information about ability at all, only players’ position, age, nationality, and Wikipedia page view counts. That jibes with a 2017 paper by some European academics who found WhoScored-type player stats could predict transfer fees about as well as crowdsourced Transfermarkt estimates could, but only after they mixed in Wikipedia, YouTube views, and Reddit posts to get a feel for a player’s popularity. Even then, the model only beat the wisdom of the crowd at guessing transfer fees under about €20 million. Once you got up in the Thiago-Wijnaldum range, humans did better than performance data at telling you what a player would go for.

Transfers, in other words, are kind of a popularity contest. Top players are a luxury good in a limited market. Branding matters. It’s hard to say exactly which way that cuts for Thiago and Wijnaldum. The Bayern midfielder’s got the edge in Instagram and Twitter followers, but the Liverpool player has the edge in playing for Liverpool, the reigning champions and popular darlings of the world’s most suffocatingly omnipresent league. Maybe Barcelona’s willing to pay a little extra in hopes that old Jürgen Klopp charm might rub off on them. Just look at the guy’s salesmanship. Asked last year about Wijnaldum’s low profile, Klopp called him the “perfect midfielder” and added, “It is not my fault he goes under the radar. I don’t set the radar!”

Jürgen, buddy, I’ve seen Gini’s radars and let’s just say “perfect midfielder” is a little bit of a stretch. If we were pricing these guys on a stats-and-clout model it wouldn’t be close. I’ve peeked at some advanced possession value stuff for Thiago and Wijnaldum and the comparison doesn’t get any prettier. To be fair, event data only scratches the surface of what Liverpool and Barcelona’s cutting edge analytics departments can tell their sporting directors about player performance (though whether the sporting directors listen is another story, as those clubs’ very different transfer hit rates would suggest). There’s a whole cottage industry of articles trying to explain away Wijnaldum’s meager statistical output by reference to his tactical nous or positional je ne sais quois, and who knows, maybe that shows up in the tracking data. Anyone who plays as many minutes as Gini for a team as good as Liverpool has to be doing something right.

To hear most people tell it, though, the case for Wijnaldum commanding a fee on par with Thiago’s starts and ends with all those minutes. Maybe he’s not the best, they say, but he’s always available! Which calls to mind that joke about the old ladies kvetching in a restaurant: “The food here is terrible, and in such small portions!” I mean, if a player’s not great, do you really want him playing all the time? These are superclubs! They’ve got whole benches full of €25 million players waiting to step in. It’s not unreasonable to worry that Thiago’s phone-book-length injury history might cut his career shorter than Wijnaldum’s, even though he’s half a year younger, but while plenty of studies show players can be injury prone, I haven’t stumbled across one showing an effect on longevity. One recent paper on modeling transfer fees tossed out injury data because, the authors explained, “early stages of the analysis showed that the player’s injury history was not contributing to his transfer fee.”

But if it’s not performance or fame or injuries driving these guys’ transfer fees, what is it? Maybe it’s the contractual stuff we talked about. Maybe it’s club finances, like that time Barcelona paid Liverpool about double what people thought Philippe Coutinho’s contract was worth because Josep Bartomeu had Neymar money burning a hole in his pocket. A few weeks ago there was a brief flurry of concern that a Wijnaldum transfer might be off the table because Liverpool had inserted a €90 million Barcelona premium into all of its buyout clauses after the Coutinho deal. Officially they’d done it to keep Barça from poaching any more Liverpool players, but maybe they just knew an easy mark when they saw one. ❧

Thanks for reading space space space! Sign up here for more.

Further reading:

• Miriam Quick, How does a football transfer work? (BBC)
• Shubham Maurya, Can Linear Models predict a Footballer’s Value? (Medium)
• Oliver Müller et al., Beyond crowd judgments: Data-driven estimation of market value in association football (European Journal of Operational Research 2017)
• Narek Sahakyan et al., Analyzing soccer's transfers and predicting footballers' transfer price (2020)
• James Yorke, Liverpool: Season Preview 2020-21 (Statsbomb)
• Rich Jones, Truth about Liverpool's Barcelona transfer premium and what it means for Gini Wijnaldum (The Mirror)

Image: Claude Martin, Black Stork in a Landscape