statsbombpy 1.15.0

Add missing events and formations fields to sb.lineups()

statsbombpy 1.14.0

This release adds functionality to interact with the new team match stats endpoint, which provides aggregated data at the team-match level. It is only available to customers with API access.

statsbombpy 1.13.1

This update fixes a bug with certain 360 frames where the variable distances_from_edge_of_visible_area is None. Previously, the sb.frames() function couldn't handle such instances and would return an error for that match.

statsbombpy 1.13.0

This release adds substitution info to the event functions.

Variables added:

• substitution_outcome_id
• substitution_replacement_id

1.12.0

This release adds pass_recipient info to events.

statsbombpy 1.11.0

This release upgrades the API version numbers for the following endpoints:
matches -> v6

What's new?

Collection Status added to Matches in Competition API.
To help customers find out where a match is in the collection pipeline we have added a new field collection_status to the “matches in competition” API. There are 16 collection status options available, which are listed in detail in Appendix 3 in the “Matches in Competition” API documentation

statsbombpy 1.10.1

This patch adds a status check to the public API calls.

Previously when a user entered the wrong match or competition and season id into the open data functions you get the following error:
JSONDecodeError: Extra data: line 1 column 4 (char 3)

Instead, we've added a status check to all the public API calls so that if, for example, the user passes through an invalid match_id we get a HTTPError 404 returned to indicate that they've tried to access a resource that doesn't exist.

statsbombpy 1.10.0

This release upgrades the API version numbers for the following endpoints:
events -> v8
player-match-stats -> v4
player-season-stats -> v4

What's new?

Expected Pass (xP) & Pass Clustering models added to events data.

• The xP model data is accessible via the pass_success_probability field.
• This is the estimated likelihood of an attempted pass being completed successfully based on the location of the pitch and context the pass is attempted under, and the location of the target location of the pass.
• The pass clustering model is accessible via the following fields:

  • pass_cluster_id: A numerical identifier that indicates which of the 60 distinct clusters a pass is most associated with

  • pass_cluster_label: A text description of the cluster most strongly associated with the pass.

  • pass_cluster_probability: The probability that the pass belongs to the cluster described in the pass_cluster_label and pass_cluster_id.

Defensive actions metric added to player stats data.

• Player Match Stats
  • player_season_defensive_action_90: Tackles, pressure events and fouls recorded per 90.
• Player Season Stats
  • player_match_defensive_actions: Tackles, pressure events and fouls recorded

statsbombpy 1.9.0

In further efforts to speed up statsbombpy this update removes the slow language engine responsible for pluralising event types.

• We now use a mapping dictionary to change event types from their singular to plural versions (e.g. Pass -> passes) - this is much faster than letting the language engine do it.
• As a result, we have removed inflect and joblib from the requirements as they are no longer dependencies.

statsbombpy 1.8.1

• This patch, adds the dependencies inflect and joblib which were incorrectly removed in the previous release.
• nose2, which is required for running the tests was moved out of the dependencies and into the extra requirements.