Becoming a Better Fan Series: Goalkeeper passing
In the leadup to our season recaps here at the SaturdaysOnCouch media behemoth, we wanted to understand the game better ourselves. A pile of data is always more useful than no pile of data but we’ve wanted to break the game down a bit, phase-by-phase the way we often hear managers and experts talk about it to try and understand new things and share that understanding with the masses clamoring for updates here. Previously we have looked at:
Phases of play leading to goals
This and the next piece will look at buildup play, and when you look at the other side of buildup you see mid and high pressing.
Buildup is a huge category of passes, broadly not the most important, goal-scoring or creating types but basically 75%+ of all passes slide into these two categories. The categories are
-Goalkeeping passes to start a sequence
and buildup passing, where there is some slight overlap with goalie passing in a way. If a team brings the ball out and then plays back to the goalie, that gets put into this category.
Here is a map showing all the different types of buildup and goalie passes we are looking at.
Today we are looking just at goalkeepers passing as always with these pieces once you dive in, it’s hard to come back up for air. Kind of like a large Domino’s pizza, you order it thinking you will save some for tomorrow and then at 11 PM you see 3 pieces left and think, ah, well a sandwich will be fine tomorrow. Anyway, dive on into your Domino’s pizza, which is today a lot of charts about how teams start and defend sequences from goalies.
Goalkeeper Passes
These are sequence-starting goalie passes. All the orange ones above are what we are talking about here. This category of pass takes up about 3.5% of all teams passes and teams range from a high of 86.6% completion rate (Brighton) to a low of 55.7% (Sheffield United).
Teams range from a high of 8.5% shot generation (Liverpool) to a low of 2% (Wolves).
Liverpool are incredible standouts here in terms of how often they start with the ball at their keepers feet and within 30 seconds turn it into a shot. Everton and Bournemouth are also standouts in that way, particularly because their original pass is often incomplete.
The one thing I would love to be able to do and think would add a ton to my analyses is to check how often the opponent turns around a ball loss or incomplete pass and turns it into a threatening situation in 10/15/30 seconds or whatever. Right now, I am not confidently able to do that but am searching for a solution.
Brighton, Spurs and Man City complete the highest % of passes, but is that just a sort of artifact of the passes they choose to make? Let’s look at how often each type of pass is completed or turns into a shot and then see how teams perform over or under expectation:
Not really. Both City and Spurs complete passes at a rate well above what you would expect from looking simply at their pass types, with Brighton above average as well.
Crystal Palace’s woes bringing the ball out from the back this year were truly awful. I would guess that they were pretty decent at preventing opponents from turning around an incompletion and generating threat, while Burnley were not. They played an inexperienced 20 year old the whole year because he was good with the ball at his feet and wound up 19th in completion% above expectation. They definitely didn’t wind up playing a ton of short passes either
compare that to Spurs
The average team was
-29% long balls (low of 9% Brighton to a high of 53% Sheffield United)
-19% GK playing it from wide (low of 14% Villa to a high of 25% Palace)
-18% spread to FB/wide players (low of 12% Palace to a high of 23% Man City)
-35% very short central passes (low of 19% Sheffield United to a high of 61% Brighton).
Looking at a team like Liverpool who were so good at getting shots off, they were basically very good at getting shots off every type of entry pass
+8% on spreading to wide players (2nd place was 2.3%)
+3.4% when the goalie played from wide (2nd place was 2.8%)
+3.5% on long balls (2nd place was 2.2%)
+2.7% on short passes (3rd place)
Notable weaknesses
Manchester United and Spurs were both really poor at winning the ball or generating anything off long balls.
Manchester City and Chelsea perform very well when the goalie moves the ball to a wide player.
Newcastle performed very well when building up with short passes, while Burnley and Palace were terrible at it.
The Other Side
This isn’t exactly how I pictured it in my mind. Brentford allow the lowest completion% off opponents sequence-starting goalie passes? Once you dig into their phase of play profile, you find some really interesting stuff. Can’t wait to do their end of season recap.
Newcastle, Arsenal and Liverpool are among Brentford in seeing opponents complete a low share of these passes, I expected Spurs to be over there as well.
Palace were great at denying shots from teams in these situations despite allowing a relatively high completion rate.
Of course a lot of this is the pass selection these teams force opponents into
Opponents bomb long balls vs Brentford more than any other team while Everton are 16th in that category with the opposite results in very short passing.
So let’s look at once you force a certain type of pass how the teams do over/under expectation there.
Here we see Brentford are good at breaking up passes, but not elite like Liverpool are who both force you into hard pass types and make all your pass types hard to complete.
Chelsea and Spurs are not in a great category here, alongside Fulham, Wolves and Luton as teams that are quite easy to play through from the goalie.
On both ends of goalkeeper passes to start sequences, Liverpool stand out. They make it very hard to complete passes against them:
-long balls in particular are almost 11% below what you’d expect, dominant aerial presences there vs Spurs at 8% above expectation) but they also press extremely high and make spreading play wide much more difficult than normal. They take away your short options and make you hit long balls, and they are great at controlling those long balls. While their defense was broadly leaky for a top team, they were fantastic in this situation.
Summarizing Both Sides
Trying to put this all together we will compare 5 teams defensively and offensively.
Brentford basically never allow opponents to go very short off goal kicks and are very stingy against them. Liverpool similarly force opponents to go long a ton, and couple that with being extremely strong defensively against these passes, while Brentford are roughly average.
Aston Villa seem to set up to allow opponents to spread play successfully out to the fullback, specifically the left fullback for some reason.
Spurs have broadly a similar distribution of pass types against as Liverpool but are just not very good at defending those pass types while Liverpool are.
Everton are strong at defending long balls but opponents very rarely hit those, instead playing very short and spreading to the fullbacks, where opponents find success.
Liverpool are just incredibly successful at every type of these passes, even the rarely used long ball. Spur struggle at long balls and were not great at spreading play to the fullbacks.
Brentford struggled out of goal kicks, outside of going long, but they kept a wide mix of types.
Everton almost never played the ball short from goal kicks but were oddly successful when they did.
Let me know if you have any comments, suggestions, improvements, confusion or ideas. I would love to hear from you in the comments if this helps at all, it definitely does for me. I would have almost never picked up on Aston Villa or Brentford’s pressing off goal kicks, but now it’s such a bizarre standout I will look for it during the game.
Buildup will be next.