Shot Maps In R using StatsBomb Data

Im not sure if anyone is following these, but I will do one more and see what happens!

I have covered some passing based stuff, I thought it might be useful to look into shots.

Therefore, the rough plan for this piece:
1) Total player xG in the WSL for this season
2) Find the top 9 players based on xG
3) Plot all shots taken including xG
4) Add labels
5) Plot the shot map of the 9 players against one another

As always, my coding is in the learning stage so this isn't a definitive way...just something that works for me and might help others!

Anyway, load in this seasons WSL data as we have previously.

We want to extract 3 things from the data - the number of shots, numbers of goals and total xG (initially including penalties)

To start - tallying player shots:

player_shots<-StatsBombData%>%
  filter(type.name == "Shot")%>% ##filter all shots in StatsBombData
  group_by(player.name)%>% ##group by player
  tally(name = "total_shots") ##tally

Once you've run the above you should have two columns - one with the player name, another with the total shots that player has taken. You essentially run the same as above for goals and xG:

player_goals<-StatsBombData%>%
  filter(shot.outcome.name == "Goal")%>% ##filter goals
  group_by(player.name)%>% ##group by player
  tally(name = "Goals") ##tally goals

player_xg<-StatsBombData%>%
  filter(type.name == "Shot")%>% ##filter shots
  group_by(player.name)%>% ##group by player name
  tally(shot.statsbomb_xg, name = "total_xg", sort = TRUE)%>% ##tally xG for each player and sort
  top_n(9) ##pick top 9 players

You should have something like below:

Next challenge, to combine the three tables into one and calculate xg_per_shot:

summary <- left_join(player_xg, player_shots, by = "player.name")%>% ##join player_xg and player_shots
  mutate(xg_per_shot = sprintf("%0.2f",total_xg/total_shots)) ##calculate xg_per_shot

Finally, left_join() the goals:

summary<-left_join(summary, player_goals, by = "player.name")

view(summary) and we should get:


Lovely stuff, this table already gives us some reasonable insight...Bremer with 10 goals off 5.3xG!

Now that we have the top 9 players by total xG plus some additional numbers, we can filter the original StatsBombData data frame by the 9 players and all shots they have taken:





As always, give the data frame a check and have a look at the player.name and type.name columns - hopefully nothing unexpected in there!

With the above completed, we can move on to the plotting. Again, using the ggsoccer package we can build a pitch from the ground up allow us to plot shots on top.

ggplot()+
  annotate_pitch(dimensions = pitch_statsbomb)+
  theme_pitch()+
  coord_flip(xlim = c(55, 120),
             ylim = c(-12, 105))

Equates to:




Good start...now to add all shots with geom_point():




We appear to have a couple of shots from corners....fun...or something has gone wrong! Will check this later.

We can adapt the size of the geom_point() by using the size function within the aes(). Size = shot.statsbomb_xg will result in:


This tells us what we know about xG...the closer you are to goal and the more central you are = higher xG. Unsurprisingly, the shot attempts from the corners don't score too highly on xG! Now to colour the points to highlight those shots that were goals:

colour = shot.outcome.name == "Goal"



Using scale_colour_manual we can alter the TRUE/FALSE legend:

scale_colour_manual(values = c("#ff4444", "#5e9a78"), labels = c("No-Goal", "Goal"), name = "Shot Outcome")

I have just picked random red and green colours to highlight no-goal or goal. You can obviously get playful altering this as you wish.

We can glean a decent amount of information from the above, where goals are scored from, the xG values of those goals, but would be handy to compare across the 9 players we filtered. Much like the passing, and an advantage of using ggsoccer we can use facet_wrap() to compare players:




This now gives us loads of insight. If you are a scout/coach you can establish some decent information...just from a quick scan:
- Both Kelly and Williams scored from corners
- Miedema and England are the stand out for attempts in the six yard box
- Kelly favours the inside right on the edge of the 18 yard box - pretty unsuccessfully

Etc etc!

Finally, to add some labels to give further information. You can use geom_text() or geom_label(), but you can use the 'summary' table we compiled earlier to plot this. Using geom_label():

geom_label(data=summary, size = 3, colour = "black", aes(x = 65, y=65, label = paste0("Goals: ", Goals)))

You can add any of the statistics that were calculated earlier and position wherever you like - mess about with it and see what works.

Finally, I will add labs() to rename the xG legend and add a title/subtitle, then you can go mad on the theme() setting to change colours, fonts etc!

The final plot should look like:





Hopefully this was a logical and sensible walk through of how to produce an xG shot map. As always, feel free to message me if you have any questions or queries! (Equally if I can do better - I'm still learning myself!)

To take this further you can remove penalties and create a NPxG shot map also!

The final code: