Are attackers the most valuable players in soccer? Or do we just think they're the most valuable players because we're better able to measure everything they do? The answer is obvious and impossible -- at the same time.
The establishment of expected goals, or xG, showed, among other things, that the best goal-scorers score lots of goals mostly because they take lots of high-value shots. Finishing skill matters, but only at the margins. The ability to find space or create space near the goal, over and over again, is the unifying skill among all the best scorers on the planet.
From there, we can look at the players who created those expected goals with their passes, and we get a pretty good sense of who the best creators are. But once you take a step back from the pass that led to the goal, you'll quickly find yourself subscribing to goals-only nihilism.
All efforts to quantify the things that happen farther away from the goal have led to similar conclusions: The stuff that happens between the boxes doesn't have much of an effect on whether or not a goal is scored.
An elegant turn through pressure by a midfielder in his own half might take way more skill than a center-forward barreling over a defender to get his head on a cross, but the latter is what directly affects the score line. And goals win games, so congrats on your press-resistant manipulation of the cover shadow, but our big man just walloped one in with his beefy forehead and we've got the three points now.
This feels wrong and right, somehow. We know midfielders matter because we've seen so many teams change their midfields and totally change the way they play. But we also know a midfielder can't single-handedly win games in the same way an on-fire attacker can.
With the proliferation of tracking and movement data, there are all kinds of PhD-level approaches that can be applied to these questions of player value: Particle physics, biomechanics, rocket science all can reveal new knowledge of how the sport works.
And I hope it does, but I am not a particle physicist, a biomechanist nor a rocket scientist. So, for today, I want to focus on a much more specific and universal definition of value: money. How much do Premier League teams pay the players at different positions, and what does that tell us about how the richest league in the world values each one?
- Why Strasbourg becoming Chelsea's "farm team" is a big deal
- Premier League Future Power Rankings: All 20 teams projected
- Ranked: The top 50 women's soccer players of 2025
The values of the average Premier League players
To start, let's look at the average Premier League salary per position.
For this, we're using the data from the site FBref. It's a combination of confirmed and estimated data, but it's broadly accurate and makes it more useful when looking at larger aggregate numbers. I made a somewhat arbitrary decision to cut off the list at players who, per FBref, make $500,000 per year. And that gives us 557 players: 145 forwards, 150 midfielders, 199 defenders and 63 goalkeepers.
We, of course, could get more granular than those designations, but the more you cut it down, the smaller the sample gets for each position. And while there's even overlap between defenders and midfielders and forwards, the distinctions get even murkier once we start talking about wingbacks and fullbacks and box-to-box midfielders. For the actual designation for any given player, we're going with whatever FBref lists as his primary position.
So, here's what the average Premier League player at each of the four positions makes:
Forward: $5.27 million/year
Midfielder: $5.31 million/year
Defender: $4.38 million/year
Goalkeeper: $3.29 million/year
Based only on that, we'd say that forwards and midfielders are the premium positions in the Premier League, then there's a gap down to defenders, and another gap down to goalkeepers.
That makes some intuitive sense, perhaps: Forwards and midfielders aren't that different from each other, and given my arbitrary cutoff point, I think the gap between the two is mostly meaningless. should also note that attacking midfielders such as Phil Foden, Martin Odegaard and Florian Wirtz all get classified under the midfield designation here, which helps boost the value of the position group. So, the further you move from goal, the less players cost.
Still, it's surprising that midfielders match, let alone, exceed attackers here -- even with those caveats. My theory is that midfielders tend to be more interchangeable than attackers. So, there's a bigger pool of relied-upon midfielders whom teams value, and that boosts the overall average of the position.
On top of that, teams are less willing to experiment with less experienced and therefore lower-paid midfielders in the same way they might with attackers. A mistake made by a young attacker is nowhere near as penalizing as a mistake made by a young midfielder, and so might result in a higher salary floor for midfielders than it does for attackers.
The values of the average Premier League starter
Continuing with that idea: Defenders are rarely getting subbed out in a match, and in an ideal world, most teams would use only one goalkeeper for an entire season. And so, the players beyond the starter level at those positions tend to make significantly less money than the starters, and it pulls down the averages.
So, what if we looked at starters only?
To define "starter," I'm not going to look at who is actually starting matches but rather which players are being paid like starters. The 4-3-3 or 4-2-3-1 is the most popular formation in the Premier League, and there are 20 teams, so we can just multiply those numbers and come up with what quantity of players at each position are being paid as starters.
It obviously doesn't work exactly like this in the unequal financial landscape of the Premier League, but I'm defining a starter as the 20 highest-paid goalkeepers, the 80 highest-paid defenders, the 60 highest-paid midfielders and the 60 highest-paid attackers.
The results:
Forwards: $9.53 million/year
Midfielders: $9.30 million/year
Defenders: $7.48 million/year
Goalkeeper: $6.55 million/year
Put another way, here's the premium you have to pay to go from average player to starter at each position:
Forward: 80.6%
Midfielder: 75.1%
Defender: 70.9%
Goalkeeper: 99.3%
This, again, makes sense. Since only one goalkeeper can play at a time and one goalkeeper tends to keep playing unless he makes a bunch of mistakes or gets hurt, the starters make way more money than the average player at the position.
Starting forwards make a lot more than the average forward, which starts to confirm some of the ideas we talked about at the start. There's less of a divide between midfield starters and average midfielders because more midfielders tend to play. And there's less of a divide between defenders because, well, there's simply fewer defenders who aren't starters because more defenders start.
On top of that, I think there's a bit of a risk premium in both of these areas, too: A backup goalkeeper or forward might be a lower-paid prospect, but teams tend to want players they think are more reliable in midfield and defense.
If we take the 4-3-3 as the base, here's what the average "starting unit" in the Premier League earns:
Goalkeeper: $6.55 million
Defense: $29.9 million
Midfield: $27.9 million
Attack: $28.6 million
It's interesting, I think, that there's not really a huge gap in how teams are valuing each unit as a whole. And if we add goalkeepers to the defense, then we could even say that teams are spending the most money on "goal suppression": $36.5 million.
Obviously many defenders, especially fullbacks, contribute to attacking play, too. But I actually think a sharper trend is coming into place.
If we accept that a good chunk of the midfield spending is coming from attacking midfielders whose value mainly comes from what they contribute near the goal, then we can boost up that "attack" value and lessen the midfield value. That would then mean teams are aligned with the analytical idea that everything happening near the goals, whether defending their own or attacking the other, is what's most valuable.
The value of Premier League stars
We know that the cost of a point increases the higher you go in the table. In other words, it's a lot easier to go from 44 points to 45 points than it is to go from 89 points to 90 points. But does that mean the best teams in the league are spending their money in the same way as everyone else -- just devoting higher sums to the same positional distributions?
Here, I'm defining a star player as a starter on a top-four team in the league. So, with the framework from the starter section, that would give us the four highest-paid goalkeepers, 16 defenders, 12 midfielders and 12 forwards. Here's how they average out:
Forwards: $18.7 million
Midfielders: $15.7 million
Defenders: $13.5 million
Goalkeepers: $11.6 million
And if we look at the premium you have to pay to go from "starter" to "star," here's what it looks like at each position:
Forward: 96.1%
Midfielder: 69.2%
Defender: 80.8%
Goalkeeper: 76.3%
To build a star-filled starting lineup, you'd be spending $11.6 million on your goalkeeper, $54 million on your defense, $47.1 million on your midfield and $56 million on your attack.
The Premier League is telling us a bunch of different things with where the teams are putting their money. The league places a high value on competence in the midfield, but there's also a ceiling on the value a midfielder can provide, and that ceiling gets lower and lower the better the team gets. I hadn't even thought of this before, but the way money is allocated in the Premier League suggests midfielders are actually more valuable to bad teams than they are to good teams. And I think I agree.
As a friend put it to me, adding a good midfielder to a bad team is like putting a new engine in your broken-down 1985 Nissan Sentra: All of a sudden it can get you from point A to point B. But the best teams in the leagues already have their engines; they need to add the higher-end details.
With goalkeepers and defenders, the higher up you go on the food chain, the more valuable they become. Perhaps that's because defensive competence comes from teamwide organization lower down the table, but as you go higher, teams have to score more goals, so their defenses are frequently left unorganized and therefore more reliant on the individual talent of goalkeepers and defenders who can no longer be protected by the system around them.
As for attackers, they earn expensive salaries everywhere, and as you go higher up the competitive ladder, their relative cost actually increases. To go from an average Premier League starter in attack to a star attacker, it's going to cost you twice as much as what you were already paying.
We started off by wondering what the comparative difference was between midfielders and attackers, and the richest teams in the league have given us one answer: Star attackers are more valuable, by about 20%.
