2018 FIFA World Cup Match Explorer by EXL
The 21st Men’s FIFA World Cup was held in Russia from 14 June to 15 July, 2018.
In September 2018, StatsBomb - a leading football analytics company – open sourced an industry-standard dataset capturing event level details for all 64 matches from the World Cup. All in-game events like passes, crosses, shots, saves, tackles, dribbles, interceptions, etc. were supplemented with timestamps, location coordinates and other qualitative football specific metrics to map out each game with a high level of granularity. The primary motivation behind open-sourcing the data was to encourage the burgeoning football analytics community to come up with innovative new research in the field.
This interactive dashboard – FIFA World Cup 2018 Match Explorer – contains a set of visualizations showcasing the finer details of each game which basic summary stats fail to uncover. The navigation guide walks the viewer through the main features of the dashboard. Here are a few general pointers about the visualization:
1. The Information Bar at the top can be clicked to access the Match Selection dropdown, Light/Dark theme switcher, match details and brief explanation of xG.
2. All individual plots contain mouse based click, hover and drag functionalities. Click and Drag to zoom into the views and double-click/click on the home button next to the visuals to reset to original.
3. Selecting a time range in the xG Chart (by click and drag) allows the viewer to filter the two charts below so that they show shots and team performance metrics for the selection game duration only.
4. The passing network maps show the team on the left attack from bottom to top, and the team on right from top to bottom. Clicking on player bubbles pops up details about that player’s performance in the Player Profile right above them.
5. Players’ passing angles and directions are illustrated through the radar plot, and are synced up directionally with the passing network map. Color of the wedges indicates the average pass length – Green to Red: Short to Long pass.
6. The size of the player bubbles and thickness, opacity of passing lines in the network maps are proportional to the number of passes a player, a pair of players were involved in.
This entire project has been developed using Plotly’s Dash framework which enables programmers to build beautiful web-based interfaces in Python. The underlying code is available on GitHub and more details regarding the functionalities and development process are available in an accompanying blogpost. Links to both are available in the ‘References’ section of the visualization.