Defensive Third Exits: Peter Haring & Hearts clean up v. Celtic
written by: @TheGersReport
Defensive Third Exits are a stat inspired by the work of Dimitri Filipovic in his coverage of the National Hockey League & may be an area that I spend some more time with moving forward.
Why defensive third exits? Because so often a controlled exit can lead to transition into the attacking half. As Filipovic explained, “It can be easy to lose sight of the fact a [defender] generally acts as the first line of attack for his team.” Obviously, the playing surface is more compact in hockey, but in football a defender who cleanly plays the ball out of his third can trigger the attack by getting the ball to a midfielder who then can get the ball into the attacking half. Whereas, a defender who clears the ball up the pitch with a long ball will often see the play come right back at him as his team fails to gain possession in the attacking half.
Before we continue, let's define what the hell this weird stat is. A Defensive Third Exit is any time the player is clearly trying to get the ball out of the defensive third via a controlled pass, by dribbling out, or with a clearance. I only include plays in which the player had the ball at his feet, meaning I do not include headed clearances.
There is a real sense of euphoria surrounding Hearts right now. After dismantling Hamilton in week one, they ousted Celtic buoyed by a (low percentage) wonder-strike from Kyle Lafferty.
Let's just say, Hearts supporters are loving taking their digs at us "stattos"....
Ninth? What a joke...I had them at seventh & I still kinda stick by that!
Guess what, Hearts have 36 more matches to prove us right (or wrong).
Kind of reminds me of when the 2017-18 version of St. Johnstone went into a fixture with Celtic...tied for first & then earned an impressive 1-1 draw.
Then...went on to finish 8th.
Yes, after four matches the Saints had earned 10 of a possible 12 points, only dropping those two points to the eventual champions. Over the next eleven matches, St. Johnstone earned.....eleven points.
So, don't start planning that victory parade just yet Hearts fans.
But...before I totally lose half our fanbase in Edinburgh. Let me point out, the rest of this post is going to be very positive about the Jambos. Peter Haring looks really good, while it feels pretty safe to say that one young center back looked way better than his counterpart for Celtic.
This is the third post I've written this season about Defensive Third Exits. In the last one, Ryan Jack came out the hero, while in this one a certain 25-year old defender (playing out of position in midfield) who spent last season in Austria's second tier put on a show that was equally impressive.
Here are the results, organized by the different ways in which a defender can look to get the ball out of their defensive third. The first two ways are classified as Controlled Exits: dribbling the ball out & passing the ball out via a direct pass to a teammate.
Haring was very involved in supporting the back four in a defensive midfield role. He's played mostly as a center-back in his career, so he clearly was comfortable in a defensive role. He tended to be decisive when on the ball & that's supported by the fact that four of his seven exit attempts were controlled plays. When he did look to get the ball out of the defensive third via a controlled pass, he was successful 100% of the time & that led to Hearts transitioning the ball into the attacking half 50% of the time.
Even though John Souttar's controlled exit numbers seems low, it does reflect 40% of his exit attempts (remember I don't include headers). Compare that to the two matches in which I've tracked Connor Goldson from Rangers, who has only had a controlled exit on 11% of the 19 exit attempts I've tracked.
In this example, compare Souttar's approach to the examples I focused on for Goldson last week.
After that initial bad touch, many center-backs would have played it safe & thumped the ball up the pitch. Instead, Souttar quickly composes himself & picks out a pass that should have resulted in a dangerous transition into the attacking half.
Now for a look at the times players couldn't get the ball out of the defensive third on a controlled play & looked for a clearance instead. These do not include headed clearance attempts.
Again, the Hearts players I spotlighted had more to do in the defensive third. Given that, it's way too small of a sample size to make any conclusions about Hendry here. He had a total of three attempted Defensive Third Exits & two were via clearances.
It should be noted that it took Celtic four full minutes to establish possession in their attacking half following this clearance. When you're losing 1-0, you can't afford to defend for four straight minutes. It’s time wasted...
If Hendry had passed to Brown, Tierney or Sinclair I have a feeling they could have easily triggered a transition in Hearts' half.
Again, these are small sample sizes (i.e. one friggin' match) - but they do offer entry points for further analysis. What if Souttar really is a player who will take the time to make a controlled pass out of the defensive third, while center-backs like Hendry & Goldson look to take the easy clearance? It's only one aspect of their game & shouldn't fully define them as defenders. However, if this data can find center-backs who can regularly make controlled exits, you have identified players who truly can be the "first line of attack" rather than the guy who makes the "safe play" & lets the the team get ready to defend some more.
This seems like a good time to note that I will also be tracking shot suppression numbers this season & will publish results sometime in the next couple of months. For a preview of what kind of data to expect, here's the results from the 2016-17 English Premier League season.
Keep an eye on Peter Haring this season. Hearts have a good track record of finding random ass players from random ass leagues (the Austrian second tier????).
It was actually Haring's defensive instincts that triggered the play that led the Hearts goal v. Celtic:
Given the amount of data we are receiving this season from StrataBet & InStat Football, I kind of feel the need to challenge myself with tracking stats that aren't just showing up in the Modern Fitba inbox. Defense is obviously a part of the analytics world that is still a work in progress. I first played around with Defensive Third Exits back in 2015-16, but never had the chance to consistently track them...so I figured this season I would.
I'm also really looking forward to seeing what comes from the shot suppression stats. They certainly will be interesting conversation starters. Check out the team-by-team EPL results from the last time I did this.
I'm still taking requests for players (especially center-backs) to track Defensive Third Exits for. Dundee has been tabbed as a team to track...who else should I be looking at?
This was written under the influence of The Cramps & The Band.