Goalkeeper Distribution Analysis System

Written By Andy Elleray, Performance Analyst, Cheltenham Town

This system was initially brought into the football World by Andy Grant, Mark Williams, Martin Thomas and Ray Clemence in their year 2000 Insight article ‘Goalkeeper Distribution Patterns in the Premier League’. This work was continued in the World Cup 2002 where Martin Thomas was accompanied by Jim Lawlor, Patrick Riley, James Carron and Mark Isaacson in their article titled, ‘Goalkeeper Distribution’.

After researching this and realising the potential of measuring the distribution of a goalkeeper in a match and the benefits this can have not only to their coaching, but along with the team tactics and performance, I decided to modify the system and tweak a few areas to get the most out of what I was looking for. This was to build a record of not only where the ball went from the goalkeeper but how, if the ball was won and whether possession was won back or retained by their team.

After becoming First Team Performance Analyst for Cheltenham Town Football Club I decided to implement this system using our First Team goalkeeper Scott Brown and goalkeeping coach Barry Richardson. After trialling the system with the Youth Team and Centre of Excellence sides I was able to modify any areas that need improvement to take this to professional level.

Luckily I was given this opportunity and the system has thrived being utilised for every match and allowing all coaching staff to access the information and data available. Not only does it show if the goalkeeper is adhering to the team tactics, it allows for analysis into what distribution method seems stronger, which one is used most frequently and what areas of the pitch most balls are being won or retained in. The outfield players can now be looked at and research can be done into what areas are being successful from the goalkeeper’s control of the ball and if the team setup and formation is inline with where the ball is going.

For example if the team tactics are to play the ball short and build up play from the defensive third and the analysis shows that a high percentage of balls are going long straight into the attacking third, the goalkeeper and staff must be notified and this rectified.

The other side of this is for example, is if the goalkeeper is continuously playing the ball long, inline with the tactics, and the forward players aren’t winning the ball then they must be notified of these statistics. Also if they are winning most of the balls but possession is not retained then the players around them might be slow in reacting to their knock downs or passes, or may not be in a good position initially to receive the ball from the forward players.

There is a variety of different information available to be calculated from the raw data produced, including percentages of how many balls are won, lost and retained in each individual area along with a breakdown of which distribution method was most successful overall and into area as well.

This system is going to continue throughout the season and studies are in place to do the same from the World Cup 2010 in South Africa, focusing on each team and their goalkeeper’s distribution patterns and methods.

Below is an example of the system and an explanation to how it works, along with a breakdown of some statistics that were requested for this particular match in the season.

Example: Hereford United vs Cheltenham Town

Date: 15/08/2009

Venue: Edgar Street, Hereford

Goalkeeper Analysed: Scott Brown – Cheltenham Town

Raw Data

AREA ACTION METHOD

Action C

OUTCOME RETAINED POSSESION
R1 A 6, 12, 13, 17, 22, 25, 27, 28 FK 4, 5, 14, 16, 21 W 1, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 15, 19, 21, 24, 28 YES :

1, 2, 3, 4, 6, 8, 9, 10, 11, 12, 15, 16, 17, 20, 21, 23, 24, 28

R2 B 2, 11, 18, 19, 24, 26 T L 2, 5, 14, 16, 17, 18, 20, 22, 23, 25, 26, 27 NO:

5, 7, 13, 14, 18, 19, 22, 25, 26, 27

R3 18 C 3, 4, 5, 10, 14, 15, 16, 21, 23 HK 3, 10, 15, 23
R4 15 D 1, 7, 8, 9, 20
R5 23, 26 E
R6
C1
C2
C3 10, 13, 24
C4 3
C5 27, 28
C6
L1
L2
L3
L4 1, 6, 11, 12, 14, 16, 17, 19, 20, 21, 22
L5 2, 4, 5, 7, 9, 25
L6 8

Overall there were 28 instances in where the goalkeeper distributed the ball, 19 in the 1st Half – 9 in the 2nd Half.

Explanation

Area:

L = Left          R = Right        C = Centre      N = Numbered Box

Actions:

A = Goal Kicks

B = Free Play (Back Pass and Throw In)

C = Ball in Hand (Drop, Volley, Half Volley Kicks, Dribble & Drive and Throws)

D = Free Kicks

E = Header

Method:

FK = Floor Kick (Dribble & Drive Strategy)

T = Throw

HK = Hand Kick (Drop, Volley and Half Volley)

Outcome:

W = Ball won

L = Ball lost

Retained Possession:

After the initial challenge, whether possession of the ball was retained/won back or was then lost. This has to be comfortable retained possession where the team can look to pass and build up an attack.

Other:

Balls directly from the goalkeeper that goes straight out of the field of play will be stricken from the record but stated at the start of the Frequency of Distribution. The method, number of times and area will be noted in this section.

Along with this, balls that bounce out of play and do not give the player a fair chance of winning and retaining the ball will also be stricken and the method number of times and area will be again noted.

Frequency of Distribution

Blank areas mean that no ball was directly played into there by the goalkeeper.

First Half

Second Half

Statistics

Observations:

  • Overall 60% of the distribution was played in areas L4 and L5. This showed that the team tactics could be to play the ball long into these areas. Either targeting their Right Back or playing to our Forwards aerial strengths.
  • In the 2nd half the balls up to the Forwards were much more varied showing a desire to mix the long ball plays up.
  • No short throws or passes were performed into the defensive third, showing a lack of building up play from the back, possibly inline with the tactics or the inability for the Full Backs to pull out wide when the goalkeeper gains control of the ball.
  • No balls went directly out of play and no balls went out of play not giving their team a fair chance of winning the ball.

Overall:

  • 57% of the GK’s distribution was initially won – with 81% of the amount won then retained.
  • Of the 43 % of the GK’s distribution that was initially lost – 30 % was then won back or possession was then retained.

Specifics:

Right

  • 25% of the GK’s distribution was initially won – with 100% of the amount won then retained.
  • 75% of the GK’s distribution was initially lost – with 33% then won back or possession was then retained.

Left

  • 56% of the GK’s distribution was initially won – with 80% of the amount won then retained.
  • 44% of the GK’s distribution was initially lost – with 50% then won back or possession was retained.

Centre

  • 83% of the GK’s distribution was initially won – with 80% of the amount won then retained.
  • 17% of the GK’s distribution was initially lost – with 0% then won back or possession was retained.

L4:

  • 45% initially won – with 64% of the balls won possession was retained.

L5:

  • 33% initially won – with 50% of the balls won possession was retained.

Distribution Actions

Goal Kicks

  • 50% of the GK’s distribution was initially won – with 75% of the amount won then retained.
  • 50% of the GK’s distribution was initially lost – with 25% then won back or possession was retained.

Free Play

  • 50% of the GK’s distribution was initially won – with 66% of the amount won then retained.
  • 50% of the GK’s distribution was initially lost – with 33% then won back or possession was retained.

Ball in Hand

  • 63% of the GK’s distribution was initially won – with 100% of the amount won then retained.
  • 37% of the GK’s distribution was initially lost – with 33% then won back or possession was retained.

FK = Floor Kick (Dribble & Drive Strategy)

  • 40% of the GK’s distribution was initially won – with 100% of the amount won then retained.
  • 60% of the GK’s distribution was initially lost – with 33% then won back or possession was retained.

T = Throw – No Actions

HK = Hand Kick (Drop, Volley and Half Volley)

  • 75% of the GK’s distribution was initially won – with 100% of the amount won then retained.
  • 25% of the GK’s distribution was initially lost – with 100% then won back or possession was retained.

Free Kicks

  • 80% of the GK’s distribution was initially won – with 75% of the amount won then retained.
  • 20% of the GK’s distribution was initially lost – with 100% then won back or possession was retained.

Summary

All the above information can be worked out using the raw data shown previously. Using different calculations you can really take an in depth look into how successful a goalkeeper’s distribution has actually been. In terms of hitting target areas, adhering to tactics etc, along with this, coaching staff can identify areas of the pitch where the ball isn’t being won or retained.

This system is designed to dissect and explore an individual match and hopefully the example has given an insight into just how many areas their distribution can be broken down in to. The aim is for this system is to aid coaches and players alike and the feedback so far has been extremely positive this season with information given used to aid the teams performances.