What happens next? How to win the Championship.

In Part Three of the series, Riley Roberts follows the data trends to unlock the Championship's cheat codes.

Apr 12, 2024

Riley Roberts

What does it take to be good in the Championship?

We’ve all heard commentators describing the second tier as “the most competitive league in the world”. And if you scan the Championship table right now, there are only five teams on the beach with only a handful of games to go. Every team from 1st place to 10th has a shot at either the automatic spots or the play-offs places. Every club from 16th to 24th has cause to fear relegation. Just look at it — it’s a magnificent thing:

Now, Sheffield United are all but mathematically heading back to the Championship. Before we reach our destination, I want to embark on a journey through historical data — the past three Championship seasons — to better understand whether there is a cheat code to winning the league. At the very least, I wanted to unlock the correlation between statistics and success so that, ahead of next year, we can say: if we can manage to do X, we’re more likely to improve our chances of going up.

Nothing is certain—the stats are a predictor, not a guarantee. And the data I’ve chosen reflect the three seasons post-COVID. These are, I think, the most relevant given that styles, tactics, teams and managers change over time. In some respects, scanning further back would be less relevant. What follows are six key markers of successful Championship teams; six data points that have a strong correlation with success (plus a few weaker correlations at the end).

(1) Wages

Let’s start with a cheap shot (not literally). The best predictor for success in the Championship — and in football more generally — is the wage budget. Per the Capology website, the top 6 in the Championship this season (as it stands) comprise the top 5 wage bills in the division plus Ipswich Town.

This is good news for Sheffield United Football Club who, barring an unfathomable reduction in player salaries, will almost certainly have a top 5 wage budget next year and who, supposedly, have no cause for a fire sale in the summer.

Having one of the biggest wage budgets is a kind of safety net for us: we have the allure of a promotion challenge and a substantial financial advantage over other teams in the league, meaning we can theoretically build a higher-quality squad. Whether we do or do not is another matter. And -2 points deduction won’t help.

(2) GK Goals Prevented

Key Terms:
Expected Goals (xG): the quality of a chance by calculating the likelihood that it will be scored
Post-Shot Expected Goals (PSxG): expected goals based on how likely the keeper is to save a shot
Post-Shot Expected Goals minus Goals Allowed (PSxG-GA): a calculation of the value of expected goals based on how likely the keeper is to save a shot minus the actual number of goals allowed

Budgets were an obvious starting point, but let’s get into the good stuff. How much does your goalkeeper’s shot-stopping ability affect your chances of winning promotion? The answer will probably surprise you.

I’m sure we’ve all heard the tropes. You know the ones: “A good goalkeeper gets you x points a season”. But what does x stand for?

It may look like a suggested password, but the PSxG-GA metric helps to give us an answer. Post-Shot Expected Goals minus Goals Allowed — more commonly referred to as “goals prevented” — shows us how much a goalkeeper needs to overperform, on average, for a team to get in the top 6 and top 2. Here’s how that looks:

League Average: -1.43 (-0.03 per game)

Top 2: +1.567 (0.03)
Top 6: +1 (0.02)

There it is! A top 2 quality goalkeeper in the Championship, only saves about 1.5 more goals than expected and prevents about 3.0 more than the average Championship goalkeeper. For the top 2, that number is closer to 3 goals. But even then, the greatest number of points a goalkeeper with +1.567 could possibly save a team is 6. Anything above that would mean your keeper isn’t just good, but exceptional.

The average top 2 team concedes a PSxG of 39.4 over a season. That means, in preventing 1.567 goals, a keeper needs to perform 4% above expected to be classed as top 2 quality.

Last season Wes overperformed his PSxG against by 2.5%, which is typically only top 6 level. Whilst this season, in the Premier League, he has underperformed by 2.6%. Grbic on the other hand, has so far underperformed by a giant 68% in the PL — small sample size alert — and over his career in the big leagues has underperformed by around 18%. Back during his time in his native Croatia though, he overperformed by roughly 19%.

With all of this in mind, I’m not sure how to rate Grbic going into next season. I don’t think he’s a top-level GK but he has shown, as recently as last season when he had to fill in for Jan Oblak at Atlético, that when he’s on form, he’s about as good as you could ask for.

Another example of that was in his final season at Lokomotiva Zagreb (which prompted the move to Atletico) when he conceded just 34 from an xG against of 55.78. Therefore, I’m going to go out on a limb and say Grbic’s future could well be in the hands of Wilder and the backroom staff. If they can get him bedded in, back him and ensure that his confidence isn’t blown to bits following this season, I think he’ll do just fine. Unfortunately though, if Grbic keeps up recent poor displays and can’t regain his confidence, I do worry he won’t be at the level of a top Championship GK.

We don’t need a world beater. But the data suggests we do need a better-than-average goalkeeper to thrive in the Championship.

(3) GK Launch %

Key Terms
Goalkeeper Launch %: the percentage of passes that a goalkeeper launches (not including goal kicks)

This is where I inevitably lose some of the “gerrit forward” fans.

When GK makes a pass, not from a goal kick or set piece but from open play, they have a decision to make. Over the past three seasons, there has been a moderate correlation between the decision a GK tends to make throughout a season and the points their team collects. In short, the more a GK chooses to make a long pass in open play, the fewer points their team collects.

This is not to say that playing long from the keeper cannot work. Of course, it can. Luton were promoted last season with almost 62% of their keepers’ passes being long. Nevertheless, the average for a top 6 team across the last 3 years is 34.6 %, and the average for a top 2 team is 28.5%. That means most top 6 teams play the ball short from the goalkeeper around 65%, and top 2 teams, around 70% of the time.

Interestingly, the negative correlation between points and going long on goal kicks — remember, GK Launch % doesn’t refer to goal kicks — isn’t quite as big. I think the reason for this is probably that, despite some of the stereotypes, going long from goal kicks still has a strong tactical element and can be used to form consistent patterns of play. The GK going long from open play is often done because of an inability to build up, or in panic when under pressure. Grbic was hesitant to go long with the ball at his feet vs Liverpool, and whilst it cost us in the game, it may be a good sign for next season given we should have the player quality to control games and dominate the ball.

(4) Long Pass Completion %

Key Terms
Pass Completion: a ball passed from one player to a teammate directly without a touch from an opposition player
Long Pass Completion %: a pass of 30 yards or more from one player to a teammate directly without a touch from an opposition player

Tying in quite nicely with GK launch, the strongest correlation I found, and probably the one that surprised me most, was long pass completion:

League Average: 47.97%

Top 6: 51.24%
Top 2: 53.92%

The gap between an average Championship team, and a promotion-challenging one, is BIG in terms of the rate at which they complete long passes. Why is this?

Well at first, I struggled to draw a conclusion. But I think I’ve found out how to interpret this. My reading is that teams with a lower long pass completion % are often inaccurate in their long passing for 2 reasons:

They’re just aimlessly hoofing it up the pitch.
They’re playing long passes under a lot of pressure due to an inability to build up effectively (how many times have we done that this season against elite opposition?)

With this in mind, it makes a lot of sense. Teams with a worse long pass completion % are generally worse teams since they’re unable to progress the ball by playing through the thirds and so are forced into playing lots of low-percentage passes.

Teams with a better long pass success rate (like Southampton and Leicester) are often better teams as their long passes are less often hopeful hits up to a striker to win duels, and more often carefully picked and precise through balls and switches of play.

A common denominator in the higher performing teams is that they tend to have fast, direct wide players who can cause problems for the opposition by offering that threat in behind and stretching the defence, allowing teams to mix it up in possession when they must.

This profile of player is something we’re lacking. and I think it will be interesting to see if we stick to playing 5-3-2/3-5-2 and, if we do stick, how we accommodate for not having wingers or wide forwards.

(5) Carries into the box

Key terms
Carries: the number of times a player controls the ball and moves 5m or more

Another area where I have concerns about us is how the number of carries into the penalty area has a moderate correlation with points in the Championship. The average for a top 2 team per game is almost 5.3 and the newly relegated teams this season are currently combining for over 7 per game.

This season we’ve averaged just a touch over 3 per game, but when removing loan players + Cameron Archer, this falls to 1.4 — quite the distance away from the numbers expected in a Championship promotion-challenging team.

During last season’s promotion, we were fortunate enough to have Iliman Ndiaye and James McAtee who were both excellent at driving us into the box. But next season, I struggle to see where that penetration comes from.

The fewest box carries per game by a promoted team in the past 2 years is 4.15, so one thing we desperately need ahead of next season is a bit of individual quality in the final 1/3. In Wilder’s teams, the wingbacks are historically the players with the most carries into the box, so signing a very attacking LWB, who is a good ball carrier, seems like a must in the summer if we stick with 3-5-2.

(6) Goals, Goals, Goals

Any promotion requires goals. But is there some kind of secret formula for creating and taking chances? Actually yes, or so it seems anyway. The winning criteria is combination of these three statistics:

60+ xG.
0.11+ xG per shot.
33%+ of shots on target.

The final (or current) positions of teams who meet all these points are 1^st, 1^st, 1^st, 2^nd, 2^nd, 2^nd, 3^rd, 4^th, 4^th, 4^th, 7^th, 8^th, 12^th.

That is thirteen teams, including all 6 teams that finished (or who are currently) in the top 2, plus an additional 4 occupying a play-off spot. Coincidentally, the 7^th place team was Wilder’s Middlesborough, who underperformed their xG by more than 15 — that’s gonna cost ya.

One thing that this shows is that you don’t need bucketloads of chances. Instead, teams need chances of decent quality ones — 60 xG from an average shot quality of 0.11 per shot only works out at 12 shots a game, which means 33% on target amounts to 4. All it takes then, is 4 shots on target per game with an average xG of 0.11, and that gets you into the top 6 more than 75% of the time.

This tells us two things: (1) teams need to be patient, and (2) teams need to be direct. Those two things appear to be opposites but I promise it makes sense.

Poorer-quality teams will defend deep against higher-quality teams. So, when attacking a low block, a high-quality team needs patience and players who can manipulate an opposition defence. When attacking an unsettled defence on the counter you need direct, fast players in behind. I think this is probably why there is an ever-increasing trend of wide forwards becoming the main protagonists in not just the Championship but the EFL as a whole — see this article from Not The Top Twenty for a discussion on the rise of Wide Forwards — rather than traditional strikers. Wide players with pace, the ability to go both ways, and create chances are becoming the go-to. And I think their prominence comes down to their adaptability as this profile suits the different kinds of situations you encounter in the Championship.

I think this could be our biggest challenge next season (aside from set pieces!). How will we create consistent high-quality chances, especially if we don’t play wingers/wide forwards? Will we see traditional wingers playing at wing-back? How about a return of the overlapping centre back? Or will we change our formation altogether?

The weaker correlations

As a final note, I want to add a few other correlations I found. These are not as strong as the ones that came before, but they pull out some interesting themes.

Both short and medium pass completion correlate with success in the Championship. It pays to keep the ball, but I didn’t think this was worthy of its own section as, in the second tier, it is not always a must. Teams have and always will find success without having 60% possession every week. Our last promotion came with only about 50% average possession, and I think the ability to mix it up when needed has become underrated, or at least, that possession dominance has become overrated.

Progressive passes and progressive carries both have a moderate correlation with success in the Championship. Of course, this feels quite obvious. Teams who are better at getting the ball closer to the net more often get better results, not exactly groundbreaking stuff.

In conclusion

Football is never as simple as saying, if you meet these statistical checkpoints, you will go up. It just doesn’t work like that. But some of the points I’ve touched on here definitely qualify as KPIs — if Sheffield United meet them, there is every chance of success. However, if we don’t meet them, it is still not impossible that we find another way to win.

If you have a slightly above-average shot-stopper in goal, can play the game on your own terms, have players who can drive into the box, and create high-quality chances, you’ll give yourself the best chance possible to get out of the “most competitive league in the world”.