Can Lustig and Simunovic be Defended?

Celtic have conceded almost a goal a game this season overall (with a bias towards conceding in Champions League matches) and the defence continues to attract criticism. In particular, Lustig and Simunovic are perceived to be below par. What do the performance numbers say?

Remember, the Glossary closes down the space between doubt and enlightenment.

“Wot You On About?”

This may seem an odd time to be talking about defensive frailty. On the back of a 2-0 hard-fought win at Aberdeen, and only conceding 3 goals in the last ten league matches since the debacle of Tynecastle. Indeed, only 18 league goals have been scored against the Hoops. With command of the Scottish domestic scene, Celtic aspire to be judged at the Champions League level. And 25 goals against in 14 European outings tells another story.

As discussed previously, defensive organisation is a team responsibility and measurement of this is in its infancy. But there are individual data points to supplement the defensive shape we see with our eyes.

That being said, I focus here on Lustig, a stalwart for six years now and a league champion in every one of his Celtic seasons so far. And Simunovic, whom Celtic went near their maximum outlay for and considered a top young defensive prospect.

This article builds on another by the excellent The Backpass Rule – Some Quick Numbers on Mikael Lustig utilising data by the equally excellent Dougie Wright. I recommend you read it as a companion piece. The thrust of that article – Celtic predominantly give away chances from the right centre of their defence. Lustig covers that station, and Simunovic is often the right sided central defender.

Note: The analysis below does not include data for Zenit St Petersburg away nor Aberdeen away. Against Zenit, Simunovic lost Ivanovic for the first goal and Lustig was generally skittish (I don’t really have a metric for that). Against Aberdeen, Lustig was sent off following an almost identical disconnect between brain and limbs as afflicted him against Partick Thistle. Simunovic conceded a string of free kicks to invite Dons pressure. Both players under the microscope will be grateful those omissions, therefore.

Lustig

Now into his 30s, the Swede has had to switch from a more orthodox right back, to being the right sided defender in a back three throughout the season. Last season Lustig played 32 times at full back and 19 at centre back. This season it is 26 at full back and 12 at centre back. So he played around 67% of his time at full back in each season. Therefore, I am comparing like for like, in terms of time in position.

Defensive Action Success Rate (DASR%)

DASR is the aggregated metric of defensive actions I use as a simple overall score for defenders. You can see in The Problem Position that it is a differentiator but that the margins are small between elite (Van Dijk) and merely competent (everyone else recently). The article above explains the metric as does the aforementioned Glossary.

For Lustig, his DASR has reduced from 72% last season to 70%. As mentioned, 2% may not seem much, but appears to be a significant drop given the historical data collected over four seasons. The DASR will be higher for central defenders than full backs, hence the comparatively low score overall.

Possession

Possessions won and lost is another useful defensive indicator. Clearly defenders would be expected to win back possession significantly more times than losing it. And across the squad this is the case with the reverse being true for the attacking players.

He is winning possession back more times than last, but also losing it 55% more times per 90 minutes than last season. Overall, Lustig is securing 5% less possession than last season, a significant drop based on historical data.

Defensive Errors

I subjectively award key defensive saves and errors. All I can say is that it is me judging it so there is the consistency of the observer. Broadly if a mistake leads to a goal, or shot, or dangerous moment, I’ll usually call it.

Defensive saves tend to be last ditch blocks / tackles / interceptions and ‘keeper saves that are beyond routine. Ultimately it is a guide but as the data set increases, a useful comparator.

How is Lustig doing?

Full back errors are usually lower than centre back errors. Centre back errors tend to have greater negative consequence due to central positioning. As shown above, Lustig’s minutes per position at full back / centre back are almost the same over the last two seasons.

His error rate per 90m has gone up 77%. 0.3 last season was quite low to be fair. His error rate this season is the same as Simunovic and Boyata. However as will see, theirs has not risen to that extent.

Only fair to point out his save rate has also gone up hugely. Now I am slightly ambivalent on this point. The reason being that my historical data tends to show that the defenders with the highest error rate often have the higher defensive save rate.

I interpret this as follows. Good defending is predominantly the product of good positioning, communication and decision making. Good defending is best “evidenced” by the lack of something happening – shots or chances. This is difficult to measure and attribute to the individual. Defensive saves are, for outfield players, usually last gasp challenges, blocks etc. There is inherent danger to be snuffed out by sudden action. In that sense it is remediation NOT prevention. Therefore having a high defensive save rate is a double edged sword and can indicate recovery where prevention in the defensive system has failed. Lustig has by far the highest defensive save rate of any Celtic outfield layer this season. Next highest is Ajer 0.23. So, well done on saving a goal but would good positioning, for example, have prevented the need for it?

Lustig – A Summary

There is clear evidence of performance regression with Lustig across the key defensive action indicators. The regression is a few percentage points, but significant nevertheless. Margins are fine at this level. The reasons why are for Rodgers and co to address. My opinion is that, as seen against Aberdeen concentration is failing compounded by poor decision making to redress. Also, more consistently, Lustig does not seem to move as fluidly as of old. I don’t have data as to whether he is slower in reaction and ground speed, but the circumstantial evidence is there.

Lustig has been one of the most utilised players in terms of playing time. He is one of the squad “backbone” players. He displays leadership, is a highly experienced internationalist, and has what Rodgers would refer to as “character”. Perhaps he can be managed more effectively to optimise game time whilst developing the obvious talent that is Ralston. Also, Gamboa has been a solid replacement when called upon defensively whilst lacking Lustig and Ralston’s attacking threat.

I love the Swede as much as any Celt, but the Celtic By Numbers mission statement includes “no sacred cows”.

Simunovic

The Croatian-born Bosnian has won a league champions medal for every completed season of his professional career. Rumoured to have cost around £3.7m, he represents a significant investment by Celtic standards. Regular injury has hampered his time in Glasgow. Over the last two seasons, nine completed 90 minutes is his longest consecutive run in the first team.

This season he has been utilised 47% of the time, last season it was 53%. He has 13 times been the right sided centre back, 4 times the middle of a three, and 7 times left sided centre back this season. I am not sure what the tactical thinking is here but would have expected greater consistency in defensive alignment.

Defensive Action Success (DASR)

Simunovic will tend to have a higher DASR than Lustig as he always plays at centre back. The data shows no improvement on last season although before the Christmas break he was up at 78%. So, a degradation on earlier season form, but not on last season. Another point is that unlike Lustig he is engaged in less defensive actions on average than last season. This may be a product of playing with Boyata and Ajer who are more aerially commanding.

Possession

Despite winning and losing less possessions events than last season, Simunovic’s overall Possession Success rate has increased 2%. He and Boyata lose possession the least of any players in the squad.

Defensive Errors

Simunovic’s defensive error rate has increased but slightly – up 8%. As seen against Aberdeen, he is an aggressive / optimistic defender who sometimes thinks he can win challenges he probably can’t. It is a common fault of young defenders to not realise sometimes you have to let the forward have the ball, and defend from there. Rather than going through players to get the ball. It’s coachable.

His defensive save rate has gone down significantly. As discussed above, this is a difficult metric to interpret.

Simunovic – A Summary

Simunovic is engaging in less defensive actions than last season and there is scant evidence of significant improvement. As best the data suggests slight stagnation, which in a young player (he is 23) is a concern. On the other hand, his appearances have been more fragmented than last, allied to the continual switches in partners and from back threes to fours.

I think the jury is out for me on Simunovic. I would like to see development and improvements in the data at this age. However, development of young players is not linear. He is a long way from the Van Dijk bench mark but further improvement is certainly possible.