Possession and goals | Bolton – Man City (2-3) EPL 2011-12

Continuing my analysis of the Bolton-Man City match, this time I look at possession, currently a hot topic among football analysts.  My objective is to analyse this metric during a number of fixed time (min.) intervals of the match, and find out how this alternates between the teams.  By looking when critical events like goals happen, I shall then draw some conclusion about the relationship about possession and goals.

The data

For this analysis one only need to use two variables in the #MCFCAnaytics Opta data for this match. These are Time (min, sec) and Team.

The analysis

  1. I analysed each half separately, and divided the min in 10 intervals that vary from 3 to 6 min.  Why not equal ones?  To avoid large differences in the number of passes for interval.  Details of these intervals are shown in image below.MC-Bolt, poss intrvl
  2. The performance by the teams in each interval is shown in the chart below.  One can see that Man City dominated possession except during four time intervals (highlighted in pink).MC-Bolt, Poss chartA clearer picture of Man City’s dominance is shown in the chart below, which plots the difference in possession by the two teams, and the shifts in momentum.MC-Bltn, Poss Balance
  3. I then compared the number of passes of the teams at each interval in sequence, and aggregated sequential intervals where there is no statistically significant difference between them.  The result is shown in the line chart below.  Note that 1st and 2nd half intervals are analysed separately, and then combined in the chart.Bolt-MC poss KS


    a) The first half divides into three intervals, with Man City dominance increasing during the middle part [12, 35), and then diminishing after scoring the second goal.   This is followed by a reaction by Bolton that narrows the gap, scores a goal, and ends the half strongly.

    b) In the 2nd half possession divides into four interval, and alternates between the two teams, albeit with a narrow difference.  Bolton is stronger in the first interval, but after Dzeko’s goal (50’) start to deflate. Bolton’s second goal (65’) comes during a period when Man City is narrowly on top. This is followed by a short reaction from Bolton which bears no fruit. Afterwards, Man City continues to enjoy more possession than Bolton, presumably holding the ball to prevent a Bolton’s come back, and end the match in control.


    Is there a relationship between possession and scoring goal?  In other words, are goals more likely to be scored during a period of time when a team dominates possession?  This appears to be true for the first two Man City goals, and marginally for the first Bolton goal.  Man City’s third was against the run of play, as was Bolton’s second.  Both these goals, though, were scored when the difference in possession was slight.   Therefore we cannot draw any conclusion from these results.  Perhaps a better way to verify such relationship would have been to analyse possession (or completed passes) during the few minutes before each goal.  Something I may do in the near future, perhaps, or for someone else to try.


About soccerlogic

Data analyst/miner of 23 years experience. Pretty sure I was first (1998) to apply Statistical Analysis and Machine Learning to study performance in soccer. I probably invented Soccer Analytics or, as I called it then, Football Intelligence. Haven't stop learning since, and experimenting new analysis that can help teams improve performance.
This entry was posted in Soccer analytics and tagged , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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