Celtic are in a title race.

If I say this is ludicrous, I can expect the usual ‘entitled’ tropes. However, pure economics suggest that in football, the wealthiest club usually prevails. The champions, with £70 million in the bank, are certainly that.

It is also true that at Ibrox they are stretching every financial sinew to the limit despite their inbuilt stadium capacity and match-day income inferiority season over season, not to mention relative commercial performance. Incredibly, their wage bill for players exceeds Celtics. The performance of the Celtic board in setting strategy and budgets is the subject of many a podcast but not this piece. Here we deal with the world as we find it.

And so back to the title race. What will be the differentiators in the run-in and, crucially, where must Celtic improve?

Goalkeeper

There is no doubt that the Rangers custodian is a media darling, with promptings for an England call-up a regular feature. The 31-year-old arrived similarly to Celtic’s Joe Hart in 2021, looking to rebuild a career that was sloping away into a life on the bench. Hart is eyeing retirement at the end of this season at 37. He has the motivation to end on a high with another title.

How do their performances compare in this vital position?

Despite what the papers may tell you, neither is the top (or bottom) of the key metrics e.g. goals saved above average. SPFL teams generally recruited well last summer in this department. Dimitar Mitov (St Johnstone), Will Dennis (Kilmarnock) and Zach Hemming (St Mirren) have impressed.

The top two are on either side of the average, however. Hart is letting in 0.05 goals less than average per game whilst Butland is saving 0.07 above. Indeed, the Rangers keeper outperforms his senior Englishman in most key categories.

His save percentage is 80 to 72 compared to an expected save performance of 77 and 74 respectively. So, again, just under the expected for Hart and slightly over for Butland. Hart is in the 54th percentile as regards how far on average he is from the optimal position when facing a shot whilst Butland is in the 99th percentile.

Hart is not in the territory Allan McGregor was last season of being statistically the worst keeper in the league and arguably playing on one season too many. He is in the ‘just below average’ category for the SPFL. However - in the Champions League - Hart ranked bottom on similar metrics to above.

This means Butland is an extremely competent custodian whilst Hart is probably getting out before his performance levels fall off the proverbial cliff. For the run-in, we must hope Hart dredges up from muscle memory the final vestiges of the form that made him England’s number one whilst at the same time hoping Butland suffers some negative variance.

Carter-Vickers

Celtic’s centre-back position has seen a never-ending circulation of personnel due to injury. The undisputed first pick Cameron Carter-Vickers has been visibly hampered and missing in fact with a niggling hamstring problem. He has contributed meaningfully to just over half the league matches but has often looked hindered in his movements.

Remarkably, without Carter-Vickers Celtic has conceded 0.64 goals per game yet 0.81 with him. However, as we know, expected goals are a more accurate indicator of performance, and in this regard, the average xG against with him in the team is 0.76 yet without him this rises significantly to 1.06.

The defensive action success rate of the team goes down from 52 per cent to 49 per cent when he is missing. The opposition volume of shots goes up from 6.12 to 8.57 without the American and alarmingly shots in the Celtic box rise from 4 to 6.57. The opposition gets possession of the ball in the Celtic box 14 times when he is missing against only 10.38 with the right-sided defender present.

With an international break coming up, Celtic needs to give him every chance to recuperate for the run-in. They simply don’t have the reserves to cover the no-frills, no-risks defending the American international offers.

Attacking efficiency

Ruthlessness in front of goal will be key in the decisive matches to come. According to StatsBomb, there isn’t much between the sides.

Celtic generates more xG with 2.06 to 1.88, according to their model per match. Also, one relatively recent trend is that the average xG per shot for Celtic is now ahead of their rivals at 0.1 to 0.09. That does not seem a huge difference but given the average shot volume per game, it equates to an advantage to Celtic of 0.13 xG per game. Surprisingly, Celtic generate slightly more xG from set piece situations with 0.45 to 0.42 per match.

They also generate more shots from high pressing which is less surprising although Rangers have more clear shots. Opta has Celtic generating 3.9 big chances per match compared to 3.66 across the city. Rangers are more effective on the counterattack with 1.41 shots generated over 1.03 by Celtic.

Celtic generates better quality chances as measured by post-shot xG with 2.65 to 2.31 per match according to Opta. And their differential from post-shot xG for and against is 1.91 to 1.62. Unfortunately for Celtic, they are underperforming their xG by 0.26 per game whilst the Blues are underperforming by a lesser 0.15.

There is not a lot in it as regards attacking efficiency. Celtic seems to have slightly the better functioning attack but needs to see a more positive distribution of variance for the run-in to make that count.

Summary

As evidenced in the Champions League, Celtic were always playing with risk in maintaining Hart for one more year. That did not work out well in that tournament and in Scotland, they are playing catch-up, relative to their rivals. That isn’t going to change in the short term and Celtic must hope most shots aimed at their goal are within Hart’s diminishing zone of spatial capability.

Carter-Vickers is mission-critical for Celtic as the defence simply does not function as well without him. But they need to do so without further risking serious injury. Their opponents tend to play two, three sometimes four defensive midfielders and the overall defensive statistics bear this out. Celtic needs their defensive leader to at least match the relative frugality of their rivals. In attack, Celtic do have the edge but are on the wrong side of the xG variance patterns. To change this, they need to have optimal cohesion in attack meaning a settled team and consistency from the wide areas especially.

All data points lead to a very tight finish and small margins will be decisive.