James Ward-Prowse is statistically the best direct free kick taker in the world
James Ward-Prowse is statistically the best direct free kick taker in the world
So we can all stop debating it now and just be smug
The Ward-Prowse free kick debate has always struck me as a bit unnecessary. The question of the greatest player, or greatest striker, Gerrard or Lampard, whatever, always remains open because any method of calling it is unavoidably subjective. There is much in the greatest ‘something’ in football that cannot be measured or compared, and much argument over the worth of different achievements and qualities.
This isn't the case for the question of the greatest free kick taker. The quality of a goal, a pass, a player, or a team might involve many intangible things that aren't easily squashed into a number. But the best free kick taker is the person who bangs them most reliably, from the most difficult positions, against the best shot-stopping. These things are all now relatively easy to measure. So the debate about Prowsey shouldn't be that hard to settle.
Maybe a year or two ago, on another occasion we were doing the Ward-Prowse GOAT convo, I realised this and asked a mate who works in football analytics to check where the skipper ranks right now. How reliably a player scores free kicks is easily given by free kick goals divided by number of direct free kicks taken. The remaining ambiguities are the difficulty of each one and the quality of the attempt, which we can now measure quite confidently with expected goals and a variant of it. I'm not going to add to the long list of xG explainers on the internet but if you need a quick refresher look here. Expected goals doesn’t measure the probability of scoring an individual shot perfectly, especially over small numbers of observations, because so many different and obscure factors can confound or enable a shot. But it's pretty accurate where n > 100, and more importantly direct free kicks are the most homogeneous class of shots in football. This means we can expect xG to the track the true difficulty of a free kick very well.
If you've read this far, cheers. If you scrolled straight down, fair fucks, I probably would have. I can’t produce any charts because I don’t have direct access to the data and my friend is mindful of his proprietary obligations. But he has pulled out a few relevant figures. Sample details in the footnotes.1 Anyway here's the sauce.
James Ward-Prowse has scored 15 of 70 free kicks since the start of the 2018 season, a conversion rate of 21.4%. His total of 15 is higher than his nearest neighbours, James Maddison and Lucas Zelerayán, in fact by some distance.
The rate at which he converted attempts muddies the waters a bit. Scoring 8 of 31 free kicks, James Maddison actually has the best conversion rate in the world over the same time period (25.8%). But, muddier still, he did so from a post-shot xG of 3.81.
Post-shot xG (PSxG) measures the probability of a shot resulting in a goal after being taken. It tracks the quality of a shot itself, rather than a chance that results in a shot (xG), and is thus mostly used to rate goalkeepers (You measure their save rate against quality of shots they face, given by PSxG). But because it’s a measure of shot quality we can use it to rate free kick attempts incredibly well. Maddison's cumulative PSxG for his 8 free kick goals is 3.81, suggesting that with average goalkeeping/luck he should have scored about half the goals that he did. Ward-Prowse, by contrast scored his 15 free kicks from a much more representative post-shot xG of 13.2. Put differently, on another day, we would expect JWP to have put away about 13 free kicks to Maddison's 4 (having had twice as many chances), based on the quality of their strikes.
The final nail in the coffin is the differential between their post-shot xG and their xG, in this case the differential between the likelihood of their strikes going in (once taken) and the difficulty of their shooting positions. I don't have the exact numbers in this category, but am given to understand that JWP comes out on top again. Trent Alexander-Arnold also scores very highly.
All in all, Ward-Prowse's dominance along most relevant metrics, and the fact he has sustained them all over a larger sample of attempts, makes it very difficult to build a case that JWP has not been the best free-kick taker in the world over the last five years, at least. When I first asked about it two-ish years ago, I was given the top 5 players for average PSxG, and JWP was in the top spot, a mere 0.01 above Messi (I think it was 0.109). This is yet another way to cut the cake, but it means I suspect he’s been on top for a while.
The GOAT question is more difficult. Juninho scored over 70 direct free kicks in his career, so even pricing in the differences between leagues, eras, opposition, technology (which isn't much), the Brazilian’s title looks pretty unassailable. He is also credited with developing the ‘knuckleball’ technique (which JWP replicated against Wolves), which arguably gives him an extra unquantifiable GOAT credit.
What we can do for certain, is bandy these decimals in face of anyone who dares question JWP's current supremacy. The naysayers will tell you data in football needs to be used responsibly, understood in context, and employed to support a story but not replace one. But we know that actually, statistics are a bludgeon, a stick to beat other fans into the 3G dirt. So next time someone pipes up, show them these digits, give it a big JWP golf swing celebration™, and ride off into the statistical sunset before they can reply. Ping it like Prowsey motherfucker.
Data from top European and American leagues, start of 2018 season to time of publication (15.02.2023). I've agreed not to name the data source.