"Hoof it, Son!" Center Back Passing Behavior in the the SPFL Premiership
Written by Matt Rhein/@thebackpassrule
Among the many discussions resulting from the untimely exit of Celtic from Champions League Qualifying, the responsibilities of center backs became a topic of debate. Like many well-known managers, Brendan Rodgers usually asks more from his central defenders than just keeping that leather circular thing away from the net. Gone are the days of the hoof it out of the area defender. The modern center back is often required to be as proficient passing the ball as he is at stopping the opposition. Celtic's back-line struggling to do both of these things against AEK Athens is one of the reasons why they will be attempting to qualify for the Europa League next week instead of the Champions League.
Outside the various verbal sparring that Celtic's Champions League exit caused, Cheuk Hei Ho of American Soccer Analysis (which is one of the best football analytics accounts out there. Regardless of your interest in MLS, they are a must follow) tweeted an interesting graph regarding central defenders in MLS and their passing ability. He put the central defenders in America's top league pass success rate per pass and their gain of shooting probability per pass on a graph to see which MLS center backs were the best passers. It seemed an interesting way to see a center back's passing ability and what type of passes they were making. Were they just making sideways passes to their teammates or trying to create chances?
With all the discussion about the abilities or lack thereof surrounding the Celtic center backs, I wanted to apply this to the center backs of the SPFL Premiership. While we do not have the robust data the ASA team does, thanks to the fine folks of InStat we can get close. Since we are only two matches into the 2018-19 Premiership campaign, it is a bit early to put much stock into this year's numbers. We can look at last year's numbers to see which SPFL center backs can pick a pass and which are still looking to clear their lines at all opportunities.
InStat provides us with data on every pass each player makes. We can see how many passes each center back completes and how many lead to a shot (also known as a key pass). We then can find the average pass completion percentage and percentage of total passes that are key passes for a center back in the SPFL. If we compare each central defender's metrics to these averages, we get an idea of which of these defenders are the best on the ball in the league.
Using a minute threshold of 900 minutes, of the 50 defenders that met this minute’s minimum only Darren O'Dea, Ross McRorie, Paul Hanlon, John Souttar, Jordan Turnbull, Bruno Alves, and Efe Ambrose were above league average (average being 76.03% of total passes; 1.06% of passes completed leading to a shot) in both categories. Now obviously, I would not suggest signing these players based on these stats alone. You would want some defensive ability to go with passing skills, so do not think I am advising anyone to go out and write a check for Darren O'Dea. However, there certainly are some interesting names on that list.
The three players in Edinburgh are of particular interest. John Souttar was part of a Hearts back line that was frequently the topic of conversation last year. Hearts put together that string of clean sheets last season, though perhaps there was some good fortune linked with those results.
Yet, there is no denying that Christophe Berra and Souttar were often impressive for the Jambos last season. Pairing Souttar with a veteran defender like Berra allows him to bring the ball up and distribute it to his attacking counter parts at Tynecastle. It will be interesting to see how the young Scottish defender gets on with Berra likely missing six months due to injury. However, after looking at these numbers Souttar seems to be a promising young defender with impressive ability on the ball.
On the other end of the capital, we see Hibernian's Paul Hanlon and Efe Ambrose on the list of above average passers at the central defender position. Efe Ambrose is a player who most fans across the country would associate with defensive frailty, we see here Efe is pretty good on the ball compared to his defensive peers in the league. Of the 7 CBs who are above average in both of these metrics, Efe has the highest overall pass completion percentage and is above average in the % of his passes that lead to shots. I'll never claim Efe Ambrose is a rock solid defender, but given that his defensive weaknesses could be overblown combined with his ability on the ball and his ability to travel with the ball, (whisper it) could Efe Ambrose be among the better center backs in the SPFL?
Both Efe and Paul Hanlon were mainstays in the Hibs backline that was third in xG against last season. Like his Nigerian colleague, Hanlon had some of the better passing metrics among center backs in the league. He is among the 7 players with an above average pass completion percentage and percentage of passing leading to shots, of which he had the 9th highest percentage of. With both Hanlon and Ambrose ability on the ball and their solid defensive metrics, Neil Lennon has two defenders he can trust both defending and passing.
With Bruno Alves and Jordan Turnbull in pastures new, I just wanted to briefly discuss Ross McCrorie. An often unfounded criticism of statistical analysis in football is that those prescribing its benefits suggest that traditional scouting methods of watching the players is not necessary. It is a strawman argument that no one even remotely involved with football analytics would make. Ross McCrorie showing up on this list is a great example of the idea you still need to watch the player despite having good numbers.
McCrorie has impressive metrics here, but that's because he should be playing midfield! He's not a center back and anyone watching him sees that. The kid can pass, there is no doubt, but he does not seem to have the skills to defend. Stevie G would be wise to avoid the folly of his predecessors and play the young Scot in midfield rather than as a center back.
Finally, since the inspiration for this post partly came from the play of Celtic's center backs in Champions League qualifiers, I wanted to comment on the metrics for Jack Hendry, Kristoffer Ajer, Jozo Simunovic, and Dedryk Boyata. Boyata seems to be not long for the Celtic squad, but the other three are easy to spot on the graph. On the graph we see that all of these Celtic center backs complete a much higher than league average percentage of their passes overall but a lower than league average percentage of passes leading to a shot.
Now, there is a plethora of creative players in the Celtic squad, so it may not be necessary for these players to play a major role in chance creation for Celtic. Yet, this could indicate that these players are more than capable of making the simple pass, but if more is asked of them they have more trouble. The metrics telling this narrative would certainly match up with what we have seen in this recent Champions League qualifying campaign from some of these players. Despite what the red-hot memes out there say, Ajer, Hendry, and Simunovic are all on the young side of the age curve for center backs so these issues could be resolved as they gain more experience. It certainly seems the numbers back up what we have seen that there are issues there though.
With the emphasis on ball playing center backs growing each year, this type of data can help to identify a player who can meet a club's needs. If you are on the hunt for a ball playing center back, running these numbers can help identify targets. From there you have a list to give scouts to view (again I want to emphasize I am not suggesting you sign Darren O'Dea. No. Don't. Stop it.) and see if they have the defensive abilities necessary to compliment their passing ability. These stats are another example of how analytics can help a club make sure your resources are being used efficiently.