Dato-Futbol / passing-networks

A couple of functions to create customized passing networks with event data by Statsbomb and tracking data by Metrica Sport

Passing networks using R

Here you can find a couple of R functions to create customized passing networks with both event data by Statsbomb and tracking data by Metrica Sport.

First, i would suggest to read this article in order to get some general context and more details of this work.

Then, you can follow this steps:

1.- Download the files contained in (or pull) this repository, including the codes (*.R files) and the field background images (*.JPG files from "fields" folder).

2.- Get the data

Statsbomb event data

• Don't forget to read and agree to the User Agreement

• Get the data using the code of the file "get_SB_event_data_WC2018.R" (save the file "SB_WC_events_clean.RDS" into a folder named "data")

• If you have any problem with this process you can also review these excellent resources about working with Statsbomb data in R: 1 & 2

Metrica Sport data

• Download the CSV files from this repository

3.- Create your passing networks graphs

• Open the "main_code.R" file, load the necessary packages and call the respective function to create your customized graphs.

• There are a lot of arguments to configurate the creation and outputs of the passing networks, both data selection and aesthetics parts.

• Here a code example using Statsbomb data with the function "soccerPassNetEventing.R"

soccerPassNetEventing(gameID = 7584, TeamName = "Japan", poss = T, pass_dir = T, convex = T,
                      minPass = 5, node_pos = "origin", nodeFill = "blue", edgeAlpha = 0.5,
                      label = T, shortNames = T,  maxNodeSize = 15, maxEdgeSize = 2.5,  
                      Flipx = F, field = 1)

• And here a code using Metrica Sport data with the function "soccerPassNetTracking.R"

soccerPassNetTracking(gameID = "1", TeamName = "Home", pos_source = "track", node_pos = "origin",
                      context = "attacking", half = "own", field = 2,
                      pass_dir = T, minPass = 2, convex = T, 
                      label = T, shortNames = F, labelSize = 2, 
                      maxNodeSize = 12, maxEdgeSize = 1.5, edgeAlpha = 0.8)

---

This work is in some way a mix of these things:

• A complement to the soccerPassmap function from the {soccermatics} package by Joe Gallagher

• An adaptation to R from the Python code contained in the Friends of Tracking repository about passing networks

• An implementation of different ideas from the football analytics community and myself

Future/pending work:

• Mainly to enable a passes value metric option, like VAEP, xT, g+, EPV or similar