An ode to Will Smallbone

When you do your work quietly, the critics are much easier to hear

Not really through any fault of his own, Will Smallbone has had a rough week.

Explaining why Saints are where they are is a hard thing. Look at me. When we were flying I wrote nothing. Now we're in a funk, I'm churning out conspiracy theories week on week. It's an unfortunate thing that in lieu of a good explanation, some fans choose to dump their grief on Will Smallbone.

Some of you will have seen a now deleted social media post that crossed a moral line, and may or may not be related to Smallbone having deactivated his social media account this week. The most important point to be made is that players deserve a degree of basic human respect regardless of how well they play. But this point has been well made elsewhere, so I’d like to take the time to make the point that Smallbone is also a genuinely talented and useful midfielder, more than he is probably given credit for. And I too brought receipts.

It’s a point worth making because even before this week, Smallbone has come in for some stick. It’s difficult to say why. Will is a likeable and hardworking academy product who we all want to succeed, but seems to have a moment or two in every game that evokes groans, even when his 90-minute contribution is very strong. Cruelly, he was jeered off at QPR by a handful of our own fans. Some players are just lightning rods.

Anyway, on the subject of 90-minute contributions, let me show you what I mean.

A few notes on my data. It’s all publicly available on FBRef. But the graphics are my own, so the green data bars work differently: they show a player’s stats as a proportion of highest scoring player’s stat in that row. So above, Stuie’s ‘Min’ bar is full because he played the most minutes of the four players. Aribo’s is just over a third full, having played 38.5% of Stuie’s minutes.

I’ve chosen these four players on the grounds that they have played consistently at no. 8 this season and for enough minutes on which to judge them fairly. I would love to compare Shea Charles but it’s not fair unless I can isolate the games he wasn’t playing in the pivot, and that is a chore I just ain’t doing. Flynn is obviously doing the Lord’s work there too, so he’s out. Similarly, bear in mind that Charly played more than a few games as a false nine, which will have had some bearing on his numbers. Finally, Aribo and Charly have played more off the bench, and thus during different gamestates (e.g. defending a lead). Keep this in mind when you interpret how they do what they do.

Most importantly, everything, except percentages, are adjusted per 90 minutes.

Goal threat

So first of all, it probably won’t surprise anyone to discover that Smallbone doesn’t spend loads of time directly threatening the goal or raiding the box. Fine. Not his game, or something for which he has been criticised. Stuie is by far our most important no. 8 for this kind of labour, and he and Smallbone are likely our most played pairing precisely because they do different things.

However, note that despite his lack of take-ons and touches in dangerous areas, Smallbone’s expected goals (0.17) is as good as Armstrong’s (0.18), and having outscored all his peers, in fact has the best differential between his actual (0.16) and expected goals (0.17) per 90. Nothing wrong with having few touches in the box if you know when to arrive there, and you can hit the target.

Smallbone also has the joint-highest expected assists (xA)¹ for completed passes, which suggests that although his passes do not lead to expected goals (0.13 xAG) as efficiently as Armstrong (0.2 xAG), the kinds of passes he plays are typically incisive and dangerous. So although he doesn’t spend nearly as much time in the box or driving at defenders as Aribo, he nonetheless matches him for shot-creating actions (2.83, 2.86).

In short, Smallbone is our most efficient goalscorer in the no. 8 position and probably the best passer, insofar as Stuie’s passes are improved by his ability to get into dangerous positions. More on this below.

He also has the most touches, so, y’know. He’s busy.

Passing and progression

In the whole Saints team, Smallbone’s pass completion (89.3%) is only surpassed by our three centre backs, and Flynn Downes (a Godly 94.0%) and Kyle Walker-Peters (89.4%).

He is the best of our no. 8s, and by a margin that matters. There’s a technical point here. A variance of 9.0% seems insignificant, but consider it like this. A pass completion rate of 90% means that for one in every ten attempted passes, you give the ball back to the opposition. At 80%, that increases to one in every five attempted passes.² Think about what that means in Martin’s system. That’s the difference between Armstrong and Smallbone.

This should be considered in light of the fact that Stuie provides more key passes, more passes into the box and the final third, and his passes generate more xG (xAG), than any of his no. 8 peers. But we also know that the profile of completed passes Stuie plays are about as likely to assist a goal as Smallbone’s (0.2xA). So there’s probably a part of the pass completion gap that is explained by Stuie being the more ambitious passer, and a part of it explained by his being a sloppier one. Just by the eye test, I personally feel Stuie is guilty of giving away balls in dangerous second-half periods more often than I would like.

Should probably note at this point that Aribo offers Southampton something entirely different here. He nearly matches Smallbone’s pass completion (88.6%), while still making a comparable number of progressive passes and passes into the box, but is a monstrous ball carrier to boot, more productive than even Stuie over 90 minutes.

Defence

The differences in ball retention between Armstrong, Smallbone and Aribo don’t seem so bad, or at least mostly explained by style and role rather than quality, until you factor in their defensive statistics.

I was really surprised to see how often big man Aribo is dispossessed, especially because this stat doesn’t even include attempted take-ons. The fact that Stuie miscontrols a ball almost three times a game is an issue, only partially mitigated by how much time he spends in dangerous positions. Remarkably, he performs worse than Alcaraz, who has essentially gone on loan in no small part because of his ball retention. It would maybe matter less if Stuie was turning out tackles, interceptions and blocks in the other direction, but he’s not doing that either (Actually big praise for Charly here).

Let’s suppose that it’s nonetheless a reflection of all the ball-carrying and penetration that Armstrong and Aribo do. Only Stuie is actually generating more shot-creating actions (3.86) as a consequence, and arguably that team as a whole will generate more threat if they can more reliably retain the ball. Smallbone is categorically the best at this.

I don’t think any of our no. 8s are perfect. The Liverpool game left me feeling that should we go back to Premier League we need a glow up in the middle. But it is my strong belief that Will Smallbone will be a Southampton player for a long time. His relationship with the club runs deep, and whilst I don’t see him progressing to an elite club, he is finding evermore ways to add value at Southampton (notably he is starting to show promise at no. 6). I expect his career trajectory to be a bit like Ward-Prowse’s, in that there will be no revolution in his game, but one would expect him to slowly hone his craft, season on season, becoming an evermore accomplished midfielder.

That, and I think he’ll captain Saints one day.

---

1. The likelihood each completed pass becomes a goal assist given the pass type, phase of play, location and distance. Think of it as a measure of the incisiveness or threat of a player’s passing.

2. Almost anyway. Not all incomplete passes lead to turnovers.