Fifty fifty ball

Exploring soccer analytics and data visualization.


Ask me anything  
An enhancement of my earlier post trying to plot all union games in an interesting way. Pairs of games with a given team are now linked. Text size is proportional to goal differential (tiny text means the Union lost by two goals—which only happened once all season) and large text means they won big. Blue is a home game; pink is an away game (as before), and that’s about it.
What jumps out at me? First: high possession games with many chances only came at home, but we didn’t do very well in them. Secondly, there seems to be a pattern with good opponents: they let us have the ball at their places, but they don’t let us shoot. See RSL, LA, Dallas, Seattle.
Plots are in R.
ggplot(union.games,
       aes(x=Union.possession,
           y=Union.attempts,
           color=H.A,
           size=Goal.differential,
           label=OPP)) +
  geom_text() + #puts the text on the graph
  geom_line(aes(group=paste(OPP,":",Playoffs)),size=0.5,color="gray",alpha=0.5) + #links the paired H/A games
  scale_x_continuous(name="Union possession pct.") + 
  scale_y_continuous(name="Union attempts") + 
  scale_size_continuous(legend=FALSE) + 
  scale_color_discrete(name="",breaks=c("H","A"),labels=c("Home","Away")) + 
  opts(title="Philadelphia Union, 2011 season") + 
  theme_bw()

An enhancement of my earlier post trying to plot all union games in an interesting way. Pairs of games with a given team are now linked. Text size is proportional to goal differential (tiny text means the Union lost by two goals—which only happened once all season) and large text means they won big. Blue is a home game; pink is an away game (as before), and that’s about it.

What jumps out at me? First: high possession games with many chances only came at home, but we didn’t do very well in them. Secondly, there seems to be a pattern with good opponents: they let us have the ball at their places, but they don’t let us shoot. See RSL, LA, Dallas, Seattle.

Plots are in R.

ggplot(union.games,
       aes(x=Union.possession,
           y=Union.attempts,
           color=H.A,
           size=Goal.differential,
           label=OPP)) +
  geom_text() + #puts the text on the graph
  geom_line(aes(group=paste(OPP,":",Playoffs)),size=0.5,color="gray",alpha=0.5) + #links the paired H/A games
  scale_x_continuous(name="Union possession pct.") + 
  scale_y_continuous(name="Union attempts") + 
  scale_size_continuous(legend=FALSE) + 
  scale_color_discrete(name="",breaks=c("H","A"),labels=c("Home","Away")) + 
  opts(title="Philadelphia Union, 2011 season") + 
  theme_bw()

Notes

  1. 5050ball posted this