xV: Villa are playing their worst football of the year at an inopportune time

Welcome to Expected Villa (xV), a column that looks at the numbers behind Aston Villa’s results! Villa are struggling, and shouts for Dean Smith’s job are growing louder. This week, we look at those Villa struggles, but I nevertheless make a case in support of Smith.

Aston Villa are playing their worst football of the year

We’ve talked before in this space about why I don’t really like using single-match expected goals (xG) calculations — there’s a lot of inherent variance in any xG model, which doesn’t know the quality of the pass to create the chance, where defenders are positioned, or what the game state is. Here’s a good primer on it if you’re interested in reading more.

This variance can largely be accounted for with appropriate context — if you watched the match, you know the difference between a shot that’s rolled into an empty net and one that has to be squeezed between three defenders to score — but it’s sometimes hard to communicate that nuance. Where xG becomes very valuable for me, though, is when it’s aggregated over a long period of time, like a full season or a 10-match run. Rolling 10-match xG calculations, then, are one of my favorite metrics — they allow us to combine the predictive power of xG with insights over a team’s recent run of form.

FBref, an arm of the broader Sports Reference network, has started publishing StatsBomb’s xG calculations for certain leagues, the Premier League included. The overall numbers aren’t good for Villa — their -16.5 expected goal differential (xGD) is the worst in the division — and a dive into those rolling 10-match xGD calculations show a concerning trend for Villa (data pulled from FBref’s Villa team page).

In a word... yikes.

There’s some rounding error here owing to FBref only publishing xG totals to the tenth, but Villa’s downward trend is noticeable no matter how it’s rounded. Through their first 10 matches, Villa kept an xGD pace that was respectable, and one that would have them sitting out of the bottom three in the xGD table today.

Over the last 10 though, Villa have gone in the tank, with the performances in their last five matches — where they’re a combined -8.1 in xGD — their worst run of the season.

I name-dropped game states earlier, but if you don’t remember from past columns, game states center around the idea that we should expect a trailing team to have the lion’s share of attacks because they need to score. Villa’s 2-1 loss to Bournemouth are a good example of game states here — the Claret and Blues out-attacked the Cherries in this one, but that would be expected, given Bournemouth scored twice in the first 15 minutes and simply needed to defend well to win.

Game states, then, are the lens that really make Villa’s recent struggles difficult to stomach. In most of the games this season where Villa have conceded a high number of chances, it can be explained at least in part by the predominant game state. For example, Villa surrendered more than 20 shots to each of Tottenham, Arsenal, Norwich (A) and Liverpool — but this is generally expected, since Villa led for the majority of those matches (Arsenal’s 10 men non-withstanding).

What’s concerning for Villa’s recent struggle, though, is that the game states haven’t aligned this way, as Villa have spent the (slight) majority of their time on the pitch trailing; in the last five matches, Villa have trailed for 227 minutes and led for just 26. And while this run includes a heavy defeat to second-placed Leicester City, it also includes much more worrying pastings from Southampton and Watford, two of Villa’s chief rivals in the survival fight.

Nevertheless, a plea on Dean Smith’s behalf

Villa are playing their worst football of the year at an inopportune time, and naturally, the manager’s position has come under scrutiny, and I get it; right now, Villa are persisting with a “Plan A” that just isn’t working. I still believe, though, that Dean Smith is the right man to be given the opportunity to make changes and see this rough patch through. My case:

1. Villa aren’t cut astray by any measure. They’re one point from 17th, and a strong run of form over the next month could see them rise to the fringes of mid-table. Bournemouth are just 3-3-7 in their last 13 matches, West Ham are 2-1-9 in their last 12. Southampton and Watford both looked utterly hopeless a few weeks ago.
2. As detailed above, this is Villa’s first truly awful run of the season. Yes, it’s at a very inopportune time, but up until a couple weeks ago, Villa got every result they needed. Sacking Smith because he lost to Southampton and Watford (without John McGinn and Tyrone Mings) isn’t a good precedent, and isn’t the sign of a stable football club.
3. Dean Smith’s Villa followed their worst run of form last season by winning 10 consecutive Championship matches. He’s shown the ability to turn poor form into great form, and while he’s not getting Jack Grealish back this year, he should get Tyrone Mings back soon to help stabilize the struggling defense.
4. Philosophically, it’s important to remember that Smith still suits the club’s long-term vision. I don’t think the club’s leadership wants to be in an annual relegation battle, and they need a stable, forward-thinking manager if they’re going to reach their long-term goals. That’s not meant to downplay the short-term need for survival, but if taking a risk now makes it more likely the club can return to the top half of the table in the future, it’s probably worth it.

Again, I understand it. But I think we need to remember how difficult this league is, and how long of a time frame 18 matches are. This league has been a frustrating mix of up-and-down clubs — Villa just need to weather the storm and come out the other side stronger.