Statistical trends in English Premier League Center Back Pairings

written by – Rangers Report    photo courtesy of –Reuters

Being a Rangers supporter means you have a built-in admiration for good center back play.  I came of age during the time of Terry Butcher & Richard Gough.  As I watched matches, I paid extra special attention to how they played because I knew the importance that they had to Rangers success.

That fascination of center back play has stuck with me.  Last summer, I did a write-up on using stats to identify the best & worst ball playing center backs at Euro 2016.

A few months ago, I began going through the rabbit hole known as the internet – looking for ideas for another statistical study for center backs.  I stumbled upon an article entitled – Analysing the Importance of a Left Foot-Right Balance in Central Defence.  It was written by Sam Tighe for Bleacher Report in 2014.

Tighe talked about how managers are constantly striving to create, “a relatively simple, yet difficult, balance to attain in central defence: a right-foot, left-foot combination to essentially split the pitch in two.”

The concept made sense & his analysis was logical.  “As a centre-back in a two-man partnership, much of what you do will be instinctive.  A left-sided centre-back (LCB) will regularly be dropping off to his left shoulder, and should he chase a wide man down the line will be looking to block/tackle with his outside foot.”

“Because of the instinctive nature of the vast majority of the decisions you will make, you naturally need to feel comfortable playing the ball first-time and clearing/playing out instantaneously. If you’re leading with your weaker side, there’s a much higher chance of something going awry.”

That last part annoyed me. The author made a subjective statement but made it seem objective.

“There’s a much higher chance of something going awry.”  Based on what data?

While I agreed with the concept of having that left/right balance – I began to wonder – what if this was just one of those beliefs that gets hyped up in a manager’s mind & it in reality, may not actually matter?

What if you just played your two best central defenders, regardless of whether they were left or right footed?  Were coaches crowbarring left-footed center backs into the lineup to the detriment to the team’s overall defensive capabilities?

For the record:  Butcher & Gough were both right footed.

courtesy of Getty Images

I wanted to find actual statistics (also known as facts) to see how important this concept of ‘balance in the central defense’ actually was.  I discovered that Matt Cane conducted a similar study for hockey back in 2014.  This post, from his Puck++ hockey analytics blog, provided the foundational ideas for my own study.

Cane wanted to see if you could measure a defenseman’s ability to suppress shots based on the side of the ice they are usually responsible for defending.

“How do we turn our knowledge of who’s playing on what side of the ice into a measurable result for defenceman? The simplest method is to just divide the defensive zone in half up the middle and assign each defenceman responsibility for defending their half. All the shots taken from the left side of the ice are the responsibility of the left defenceman, while all the shots taken from the right side are a negative mark against his partner. It’s a simple model, but as we’ll see later, it does provide a half decent view of individual defensive zone play.”

Hmmm…what if I applied this same concept to football?

Now obviously football, like hockey, is a fluid sport.  A center back is often called upon to support his partner & they both end up defending in the same area.    Also, sometimes a center back had is man marked & a breakdown somewhere else has created the opportunity for a shot to be taken.

I acknowledge, as Cane did, that this will not be perfect.  But over hundreds & thousands of minutes…some definite trends will develop.  Sometimes these trends will support common perceptions & sometimes they will contradict them (see the results for Laurent Koscielny in the next post of this series).

In this first part, I will share the results on a macro level & then in the next post will share a new defensive rating that I created for individual center backs based on the data from the study.

I went back through the 2016-17 English Premier League season & tracked shot locations & organized the results based on whether it was a left footed/right-footed partnership, or if it was two right-footed center halves playing together.  I also tracked goals & scoring chances (more on that later).

Note:  I used FourFourTwo’s StatsZone to collect the data.

I split the pitch in half & recorded which half of the pitch the shot came from.  I only included shots that started within the frame.  Notice the blue shot (a shot that was on target) & the yellow shot (a goal) originated from outside this frame & were not counted.  Shots from that far out were less likely to be impacted by a center back.  Make sense?

So in this example, there were two shots counted & both came from the attacking right side.  Given that this is the point of view of the attacking player, realize that those two shots came under the zone of influence of the left-sided center back.

Before sharing the results, let’s revisit some of the questions that helped originate this project:

  • To what extent, is it necessary to have a left-footed center back paired with a right-footed one?
  • Does this kind of ‘balance’ lead to better results?
  • Are managers blinded by a left-footed center back?  Will he leave better options on the bench simply to have a balance in the heart of defense?
  • Can a righty-righty pairing defend just as well – even if one of them has to play on  their weaker side?

The results below include when a left-footed center back is paired with a right-footed center back & when two right-footed center backs are paired.  They also include the rare occurrences when two left-footed center backs were paired & also when the pairing was inverted (i.e. a left-footed center back played the right side, while a right-footed center back played the left side….aka the Alfie Mawson experiment).

Also, you may be wondering…what about when teams played three center-backs?  There are no ‘pairs’ per se.  In this case, I “paired” the central center back with the player on his stronger foot.  In most cases, the central defender was right footed, so he was ‘paired’ with the center back on the right (also usually right footed).  If the central defender was a lefty, he was paired with the center back to his left.

The results (again the sides that shots are coming from are from the p.o.v. of the attacking team):

Given the sample side for the more rare pairings, let’s just focus on the left/right vs. right/right pairings.

  • When we are only looking at shots, there is zero difference between the shot rates for the ‘ideal’ pairing of a left-footed center back with a right-footed one & ‘being forced’ to play two right-footed center backs.  Literally, no difference.  In both cases, 56% of shots are coming from the right side against the left sided defender.
  • Ummmmm…holy shit.  That’s a pretty big difference:  56% & 44%.  I assumed the results would be closer to 50/50.

Let’s see if the results are similar for scoring chances.  The term is borrowed from hockey.  The commonly accepted definition of a scoring chance is that, “A scoring chance is a clear play directed toward the opposing net from a dangerous scoring area.”  Basically a shot taken from a location that is statistically more dangerous then other locations.

For me, it’s any shots coming from inside that red area.  Well, let me clarify that…any kicked shot coming from that area.  Headers are different.  I only consider a header a scoring chance if it comes within the vicinity of the six yard box.  Headers that come from a more central area, away from the six-yard box, do not score at a high rate.  Neither do shots inside the box, but from an angle.  The shooter has less of the goal to shoot at as he is pushed further away from the central part of the penalty area.  Shots from outside the box are clearly not scoring chances, they are glorious when they are goals but usually 95% of the time they aren’t.

So the results below have eliminated shots that are not deemed to be scoring chances based on the shot location & what kind of shot it was:

  • The overall rate has not changed much.  55% of scoring chances come on the attacking right side vs. left sided defenders.
  • The rate actually increases back to 56% when a left-footed center back is paired with a right-footed center back & dips slightly to 54% when two righty’s are paired together.

How about goals?  I’m not going to spend too much time moving forward including goals as part of the study.  But, I’m sure you’re curious to see if the trend continues of more action occurring from the right side of the attack.

  • Damn.  The rates actually went up.  62% of goals coming from shots on the right vs. left sided defenders is a rate I did not see coming.  To me that is highlighting the need to put your best defensive center back on the left side…regardless of he’s left or right footed.
  • Again, there is little to no difference in the results from the ‘ideal, balanced’ pairing of a lefty/righty & the righty/righty pairing.

Why is this?  Why are 56% of the shots, 55% of the scoring chances & 62% (!) of the goals coming from the right side of the attacking third?  I posed this question on Twitter & here were some of the responses:

The world is dominated by the right handed (or this case:  right footed) people.  Maybe this is just a natural tilting of the pitch?

This was my original thinking as well…again highlighting the importance of having your best defensive center back being positioned on the left to provide support for when a left back has disappeared up the pitch.

Well, the stats have proved this theory to be invalid.  Left footed center backs are just as weak on the left side as right footed defenders.

Jonny Evans in 2012, courtesy of Getty Images

Some quotes from Jonny Evans back in 2012 may highlight some of the reasoning for these results.  In the interview, Evans (who is right footed) was explaining why he felt more comfortable now that he had been switched from the left side to the right side.

“Whenever I played, it was always on the left side of defence.  Growing up I was probably always a bit more capable of using my left foot than whoever was alongside me in the centre, so I ended up on the left side.”

“It’s harder bringing the ball out of defence on your left side, though.  When you’re passing the ball five yards you’re fine using your weaker foot but if you’re driving into space and dribbling forward, you ideally want the ball on your good side.”

From a player’s point of view, Evans makes sense.  Being forced to play on your weaker side will likely form a mental block about your abilities & will likely set limitations in your head of what you can & cannot do.

However, the facts suggest that it doesn’t matter if it is a left-footed center back or a right-footed one.  The results are basically the same.  Right-footed center backs played 72% of the minutes on the left side of defense & had the same results when a left-footed defender played that role.

courtesy of AP

Now what?

One of the goals of analytics is to impact team strategy in order to gain an advantage over the opponent.  Whether that be with player selection or tactics:  how can the facts that come from statistical output influence the decisions made my managers?

It is safe to say that managers can step away from any fascination that they may have had with finding a lefty/righty  balance in the heart of the defense.  Rather, given the trends for shots, scoring chances & ultimately goals – the priority for managers should be two fold:

  1. Put your best defensive center back on the left side.  That’s where shot suppression is more needed.  Attacks are more dangerous & teams are more likely to get shots off when attacking up the right (regardless if a center back is left or right footed), so put your best defender on the left to help limit the impact of the attacking team.
  2. Tactically, can you develop a defensive system that forces teams to play on their left side (again from point of view of the attacking team)?  We’ve seen that less shots & less scoring chances come from the left side vs. the right side of defense.  If you can force teams to play from that side you may be more able to dictate the results.  Will attacking teams instinctively look to switch play &/or will it be easier to stifle attacks?

For the record, here are results for when teams played with three center backs:

Note that the shot rates were, again identical.  However, scoring chances were way more balanced, while the goal % did come down slightly.

This is part one of this series looking at the results of my center backs study.  The next part will highlight my first attempt at developing a defensive rating for center backs.  To add to that post, please go here to identify who you think the best center backs in the EPL are.

All of the data from this study will be released once all the posts have been written.

One thought on “Statistical trends in English Premier League Center Back Pairings

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s