Jack Butland's worrying Rangers numbers and Kyogo defence

The 31-year-old made a crucial save away at Rugby Park which helped catalyse a comeback for Philippe Clement, an 18th win in his first 20 league games, to stay top of the league with a 2-1 victory in Ayrshire that felt significant. The League Cup had been won, difficult fixtures navigated - a significant title win was now very much in grasp.

Since that night in Ayrshire, Clement has struggled to regain the momentum that defined his start in Glasgow, with Butland’s form both a by-product of that drop-off and a cause. The Englishman’s concession during Sunday’s 3-0 Old Firm defeat, which sent Celtic five clear after four games, triggered a level of criticism generally avoided for a popular player who arrived at the club last summer and adapted quickly to demands.

The numbers support a belief that Butland hasn’t quite hit the levels seen over winter since that meeting with Kilmarnock - but why?

To assess Butland’s form in a Rangers shirt we can use a statistic our data provider StatsBomb calls Goals Saved Above Average (GSAA). GSAA is defined as: “How many goals did the keeper save/concede versus expectation (post-shot xG faced)? This is representative of how many goals the goalkeeper's saves prevented within a season.”

Post-Shot xG is a metric that evaluates the likelihood of a shot turning into a goal after it has been taken. It uses elements such as shot velocity, the positioning of defenders, a goalkeeper’s starting point and more to judge the probability of a shot finding the net. GSAA takes this data, deducts the goals a goalkeeper does concede and tells us if they’re over or underperforming based on the quality of shots faced.

Context does need to be applied to some numbers. For example, Daizen Maeda’s deflected effort that James Tavernier failed to clear in last season’s 3-3 Old Firm draw at Ibrox had a PSxG of 0.04 (a 4 percent chance of going in). Butland understandably wasn’t primed for a shot in the below frame, but GSAA still marks him down heavily (-0.96) for such a moment.

(Image: StatsBomb)

Butland conceded a few avoidable goals once the league was over in high-scoring fixtures last season (a 5-2 win vs Dundee and a 3-3 draw with Hearts) that skewed his overall record somewhat. Furthermore, raw numbers don’t always tell a full story. We’ll come on to provide some context as to why the No.1 was in a far-from-ideal starting position to save Kyogo’s effort on Sunday, much like the above Maeda goal. On the other hand, the 11-cap England stopper is yet to have a standout Old Firm game and has conceded a few from distance he ought to have fended with better.

The below trendline outlines Butland’s GSAA on a game-to-game basis (blue bars) and 10-game rolling average (red line). We’re only using data from the Scottish Premiership.

What is this showing me?

• Overall, the red 10-game rolling average line agrees with general consensus. Butland was performing at around average this time last year, before shooting upwards in line with the excellent start under Clement, tailing off around the spring and split and bouncing back to an expected average this season.
• There are more single-game ups than downs, as outlined by the blue bars. The big, alarming match at the end of last season (a 3-3 draw with Hearts) was a dead rubber. Everyone had an off day in Michael Beale’s final game (a 3-1 defeat against Aberdeen) and the loss which started last season’s slip (a 2-1 defeat at home to Motherwell).
• Butland’s had some really important, genuine point-winning performances in moments his team have needed him. The opening day of the season (0-0 vs Hearts) could’ve feasibly gone in the hosts’ favour if not for a few important interventions while the aforementioned comeback against Kilmarnock was made possible by an outstanding save with Rangers 1-0 behind.
• But there are reasons for concern too. Only once has Butland finished an Old Firm with an ‘average’ GSAA number (last season’s 1-0 defeat under Beale). He made a few excellent stops in the first half of last season’s 3-3 draw with Celtic before a late error, allowing Adam Idah’s shot to squirm under his body. In Sunday’s Old Firm defeat and the 2-1 defeat at Christmas, the former Stoke stopper conceded long-range goals against Kyogo the data suggests he should’ve done better with. That was similar to the effort from distance scored by Kyogo in Celtic’s 2-1 Old Firm win at the tail end of 2023.

Butland is clearly a very good goalkeeper who’s been forced to take upon a very big burden at Ibrox - both in terms of his performance and leadership. What’s more, last season he played far more minutes (5220) than in the previous four seasons combined (4815). Is there a reality that Butland simply tired from the spring onwards - again like many of his teammates? Is there a suspicion that in the Old Firm, there is a collective inferiority complex that Rangers just cannot seem to overcome? While some have cited the fact that Butland did not receive a rumoured England call-up in March as a reason for his poor form, would such circumstances not spur him on further?

If you were to remove the final two games of the season against Hearts and Dundee last season Butland’s GSAA in the league came in at 1.01 - he would still be in the green despite Rangers’ drop-off from March onwards. At the end of February, that number stood at 3.23 and in the Europa League Butland finished the campaign having saved 1.82 goals above expectation.

(Image: StatsBomb)

The No.1 made a costly error to concede a shot from Kyogo with a PSxG of 0.26 (a 26 percent chance of finding the net) on Sunday. As mentioned it wasn’t the first time he’s seen a long shot rifle past him at Parkhead. According to StatsBomb Kyogo’s second goal at Christmas (in a 2-1 win for Celtic) had a 28 percent chance of becoming a goal once it left his boot, the exact same value attributed to Matt O’Riley’s effort in a 2-1 victory for Brendan Rodgers’ team that sealed the title after the split.

(Image: StatsBomb)

Again, while the Rangers Review doesn’t profess to be anything close to an expert in situations like the one below Butland takes the flack for the goal conceded but likely expects Barisic to block the near-post shot given the space available to O’Riley to curl an effort to the back post.

Instead, Barisic attacks O'Riley leaving a gap for the midfielder to shoot through. 

(Image: StatsBomb)

Newcastle defender Dan Burn explains this concept further: “We say ‘no big shapes’ - the strikers look for it, they’re that good at it, the second you make too big of a step they just fire it back through your legs. Normally what we do, say it’s me and Popey [Newcastle keeper Nick Pope] is I’ll take the back half of the goal [and not leave gaps to shoot through] so he can’t fire it in and Popey will take the front half of the goal.

"It’s something we practice and it doesn’t look great sometimes if it looks like half the time I am not trying to block the shot but I have an understanding with Popey that anything in the front half of the goal he has got, and I’ll cover the back half. It’s quite hard for him [the goalkeeper] if he is doing that and the ball is getting fired between your legs to the back of the goal.”

So should Butland do better here? Likely so, but from his point of view, the keeper also expects one of Barisic and Ben Davies to help him out and block one side of the goal. Otherwise, he risks leaving one side of his goal totally free.

(Image: StatsBomb)

Butland offered some context as to why he was beaten too easily from distance again by Kyogo after Sunday’s 3-0 defeat.

“I'm in a position which is suitable for where we were at in the game, being in possession of the ball,” he said.

“In giving the ball away, I'm trying to get back to my optimal position where I want to be in that moment. And perhaps a little bit I didn't quite get back to where I wanted to get to, ideally... As I said, my optimal position was where I wanted to get back to. I didn't get the opportunity to get there with us turning over the ball.”

Again, Butland is not void of criticism in this situation but he’s thinking of supporting possession from an angle, as had been the case all game, first and foremost rather than defending a shot from distance. Rightly, the goalkeeper doesn't expect his team to lose the ball and concede a chance so quickly from a position of initial control. 

(Image: StatsBomb)

(Image: StatsBomb)

StatsBomb’s freezeframe demonstrates that the keeper wasn’t in his ‘optimal position’ which made it too easy for Kyogo to beat him from distance again.

(Image: StatsBomb)

Two things can be true in unison - Jack Butland is not in the form shown over winter and has let in goals he’d expect to save. The inability to deal with Paulo Bernardo’s late effort in the Scottish Cup Final, for example, put a game-winning chance for Idah on a plate. Even if Butland was in a position to build play leading up to Celtic’s second on Sunday his recovery to the middle of the goal was slow. The numbers clearly suggest that he is yet to perform in an Old Firm fixture.

It is no coincidence, however, that a numerical drop-off ties in with a wider context. Tavernier’s inability to clear a ball and hitting it off Maeda into the net, an ever-changing defence last season failing to cover ‘their side’ of the goal as O'Riley found the net, a team losing the ball in bad areas and leaving the goal vulnerable.

Butland’s form is a byproduct and cause of the drop-off seen at Ibrox since March - Clement needs both to recover to top-level sooner rather than later.