Celtic Strikes Back
Article by James Dailey
The signing of Patryk Klimala has generated the typical frenetic response from Celtic support, as some immediately think he and the Slovakian league are rubbish, some watch his video highlights on Youtube and extol on Twitter that he is the “next Lewandowski,” while still others take a wait and see approach. Being a proud member of the stats and analytics fiefdom, I’ve run off and tried to accumulate performance data and then do some comparative analysis. The following data is via Wyscout and breaks out performance data metrics per 90 minutes for the following:
Klimala in the Polish top league this season;
Bayo broken out between his 2018-2019 season in Slovakia, then his limited minutes at Celtic this season;
Andraz Sporar who also played this season in Slovakia and was linked to Celtic; and
Edouard’s metrics for his first season at Celtic and then this season.
Rational
My reason for these selections and comparisons are as followings:
Age is an important component of player development, as players ascend and then descend physical aging curves. Players typically peak physically between 22-25, slowly start to decline into their late 20’s, and then begin to drop off more quickly in their 30’s. However, physicality is not the only aspect of an athlete, and the cognitive development players experience accumulates and largely offsets the modest physical decline into their late 20’s. The impact is a player peaking with the relative balance of wisdom/experience overwhelming the modest physical decline into ages 27-29. At that point, the physical decline speeds up and begins to overwhelm the accumulated wisdom. Each position has its own spectrum of physical vs cognitive skills – wingers and central midfielders lean towards the former, while centre halfs and strikers lean towards the latter. I’ve broken each player out by season/age to provide context of where they may be in their physical and cognitive development.
I’ve selected the current roster of Celtic strikers with the exclusion of Leigh Griffiths, and then included Sporar because of the value offering further context for Klimala’s performance as a 21-year-old in the same league. Griffiths has so little performance data recently and is now at an age where physical decline is a major risk. I believe there is a significant risk that Griffith’s time at Celtic’s level is over.
Other than issues with sample size, relative strength of league and opposition are important to consider. One imperfect way to have a starting point for comparison is the UEFA club coefficient by country, which places Scotland 16th, Poland 28th, and Slovakia 30th. However, those coefficients can be impacted heavily by 1 or 2 teams being dominant domestically and accruing points with success in Europe. I would argue Scotland fits this characterization, whereas it is possible that the competitive balance and depth in Poland and Slovakia could be comparable or even better than Scotland. What do I mean by this in layman’s terms? Basically, their versions of Hamilton and Aberdeen may be better than ours, but overall, I think it is reasonable to consider that the competitive levels in the three leagues are in the same neighborhood.
Player Comparison
The following table compares each player by season:
Klimala
I’ve highlighted various cells to draw attention to metrics which stand out positively (green) and negatively (red). Within the context of these strikers, Klimala has generated the lowest xG per 90 minutes, but the second highest xA. Edouard’s 0.24 xA is exceptional for a striker and Klimala is not far behind. Klimala’s extremely low # of passes per 90 may be more a product of his team’s style of play. His aerial stats are ok and his interceptions are low, but again, that could be due to playing style. His total xG+xA of 0.67 is a very good level of offensive production, as his excellent xA stands out.
Most importantly for Klimala is his age and some of the qualitative reporting which has surrounding his transfer. Apparently, he’s had some attitude issues in the past and the past six months have seen him “grow up” some and become a more dedicated and coachable professional. That is completely normal within the context of cognitive development for players his age, as people come from a huge spectrum of background and life experiences. This mental maturity is variable and can often emerge in jumps, which also often coincides with big improvements in development. Players’ biggest jumps in development are typically in their late teens and early 20’s precisely for this reason – mental maturity. Klimala’s reported elite pace, quickness and playing style running off the shoulders of centre halfs could be a potent combination if Lennon decides to play him with Edouard, whose creativity is elite.
Edouard
I want to again highlight how elite Edouard’s creativity is, and how much his Dribbling stands out. If we look at his heat map for the season, we see how left sided he is and how much he creates outside the box – almost like a left sided #10:
As Edouard has developed and played more, he is unquestionably an incredible player who offers tremendous value through his creativity and dribbling. However, his scoring and finishing stats are good but not great. It is his well roundedness which makes him elite. Combined with his left sidedness, and Celtic’s already imbalanced tilt left with McGregor being a superior creative player on that side vs Brown on the right, I think Celtic really does need a striker(s) who dominates in the box. Despite rumors of a loan deal to get playing time, I really think Bayo could be that man.
Bayo
I started this research looking to gather more information on Klimala, but it has also provided me renewed optimism for Bayo, as he dominates the table with green. Bayo’s stats standout as him having a presence inside the box, with superior aerial stats- both in number but as importantly his % of wins. I am mystified by the widespread narrative among our support that he’s been relatively poor when he’s played this season – I believe he suffers from issues of confirmation bias similar to Bolingoli, as a few poor touches have created an inaccurate mental anchor for many. For example, Edouard has averaged 13.11 losses of possession per 90 minutes this season vs 11.89 for Bayo. People understandably remember Bayo’s really ugly touches, but he averaged 12.86 in his last season in Slovakia, which is still below Edouard’s rate this season, and suggests he may be better on the ball than many perceive.
Bayo’s heatmap also offers why I believe he offers something different:
This is from his 18/19 season in Slovakia – notice the dark red right around the penalty spot. How many times have you screamed at a game when a Celtic cross goes through the box and “no one” is there? I know I do it all the time! A comparison of this Bayo heatmap with Edouard’s is stark. I’d also like to highlight that Bayo’s last season in Slovakia was at the age of 22, yet still compares favorably to what Sporar has done this season in his age 26 season. It is likely that Bayo still has further development experience.
Impact on Shape
I also am growing increasingly concerned that Lennon’s preferred 4-2-3-1 does not optimize the talents of our players, including Edouard as a stand-alone striker. Underlying performance metrics suggest McGregor has been better as an attacking 8, Brown better as an centrally played 6, Ntcham and Christie have also been better as attacking 8’s, and based upon my analysis, I think Edouard would be better leveraged by either playing as a left sided forward, or as part of a 2 along with a box striker. Klimala appears to offer something different which could compliment Edouard if Lennon is open to a change in shape and/or role for Edouard.
Is Klimala the Answer?
However, Klimala is 21 and just “broken out” in his mental development in the past 6 months. Setbacks and a period of acclimation would be the norm. Can Celtic “afford” patience, and is it fair to the young man to have such pressure to perform immediately given the title challenge and an opportunity for a run in the Europa League? I believe we need 3 strikers – especially if playing 2 is an option Lennon desires, and hope Bayo is one of the three. His aerial superiority also addresses two huge areas of competitive weakness for Celtic this season – both scoring and defending set pieces/corners. Particularly as we look forward to the two remaining derbies, which in aggregate may determine the league, a shift in tactics and addressing clear competitive weaknesses vs Rangers should be a priority.
Klimala may very well assist in that regard, but it seems to me the risks of that not occurring are significant given his age and stage of development, and Bayo’s skillset offers a clear fit in formulating a badly needed plan for those upcoming derbies. A scenario where Klimala does eventually develop to become the next “Vardy” or “Lewandowski” could unfold even if he doesn’t contribute enough to help win the league this season. I believe that putting all our chips on Klimala while loaning out Bayo would be a significant and unnecessary risk to 9 in a row, as well as placing unnecessary pressure and risk on Klimala’s long term development.
Finally, the “Mystery Man” in the table above is Lawrence Shankland. His data needs to be tempered by the fact that the Scottish Championship is clearly a lower level, and his stats are not nearly dominant enough for me at that level. Pass!