Skip to content

bigdatacup/Big-Data-Cup-2021

Repository files navigation

Big Data Cup Submission Guidelines


Submission Rules & Guidelines:

Maximum 6 pages, including figures (size limit 10GB on submission)

Format of submission:

  • Define your thesis
  • Provide a summary of your methodology
  • Give an overview of your findings
  • Identify key action points from your analysis
  • Code may be included as an appendix

Dataset

Summary

The dataset is comprised of Stathletes-tracked women’s hockey event data from the Olympics and NCAA, as well as player tracking data generated from broadcast video. The included events have been translated from Stathletes’ raw data to enhance accessibility and interpretability. The various event types include shots, plays, takeaways, puck recoveries, dump ins, dump outs, zone entries, faceoffs and penalties. Event definitions may slightly differ from other sources. For each event, expanded details are provided and the relevant skaters and teams involved are indicated when necessary.

NEW: Tutorial on play-by-play, plotting, and tracking in R can be found here


Tracking Data from Broadcast Video

Tracking data and documentation is here.

NEW: Tutorial on merging tracking and play-by-play data can be found here.


Contextual (Event) Data

International Games | NCAA Games | NWHL Games

  • Date (e.g. ‘2020-12-23’. Format = ‘yyyy-mm-dd’)
  • Home Team (e.g. ‘Toronto Maple Leafs’)
  • Away Team (e.g. ‘Boston Bruins’)
  • Period (range from 1-3 for regulation, 4+ for overtime)
  • Clock (e.g. ‘19:34’. Format = ‘mm:ss’)
  • Home Team Skaters (range from 3-6 for home skaters currently on the ice)
  • Away Team Skaters (range from 3-6 for away skaters currently on the ice)
  • Home Team Goals (current goals scored by the home team at the time of the event)
  • Away Team Goals (current goals scored by the away team at the time of the event)
  • Team (name of the team responsible for the event)
  • Player (name of the player responsible for the event)
  • Event (type of event, e.g. ‘Play’, ‘Shot’, …)
  • X Coordinate (x-coordinate of where an event occurred on the ice, between 0 and 200)
  • Y Coordinate (y-coordinate of where an event occurred on the ice, between 0 and 85)
    • Coordinates are always from the perspective of the eventing team

  • Detail 1-4 (up to 4 supplementary details for each event, varies by event type)
  • Player 2 (name of a secondary player involved in an event, varies by event type)
  • X Coordinate 2 (x-coordinate of a secondary event detail, varies by event)
  • Y Coordinate 2 (y-coordinate of a secondary event detail, varies by event)

Events

Shot

Shot attempts that are unsuccessful (block, miss or save)

Players Involved

  • Player: Shooter

Coordinates

  • X,Y Coordinate: Release location

Event Details

  • Detail 1: Shot Type (Deflection, Fan, Slapshot, Snapshot, Wrap around, Wristshot)
  • Detail 2: Shot destination (on net, missed or blocked)
  • Detail 3: Traffic (true or false)
  • Detail 4: One timer (true or false)

Goal

Shot attempts that are successful (goal)

Players Involved

  • Player: Shooter

Coordinates

  • X,Y Coordinate: Release location of the puck

Event Details

  • Detail 1: Shot Type (Deflection, Fan, Slapshot, Snapshot, Wrap around, Wristshot)
  • Detail 2: Shot destination (on net, missed or blocked)
  • Detail 3: Traffic (true or false)
  • Detail 4: One timer (true or false)

Play

Pass attempts that are successful

Event Types

  • Direct (e.g. a tape-to-tape pass)
  • Indirect (e.g. a pass that is rimmed along the boards)

Players Involved

  • Player: Passer
  • Player 2: Intended pass target

Coordinates

  • X,Y Coordinate: Pass release location
  • X,Y Coordinate: Pass target location

Event details

Detail 1: Pass Type

  • Direct (eg. a tape-to-tape pass)
  • Indirect (eg. a pass that is rimmed around the boards)

Incomplete Play

Pass attempts that are unsuccessful

Event Types

  • Direct (e.g. a tape-to-tape pass)
  • Indirect (e.g. a pass that is rimmed along the boards)

Players Involved

  • Player: Passer
  • Player 2: Intended pass target

Coordinates

  • X,Y Coordinate: Pass release location
  • X,Y Coordinate: Pass target location

Event details

Detail 1: Pass Type

  • Direct (eg. a tape-to-tape pass)
  • Indirect (eg. a pass that is rimmed around the boards)

Takeaway

Steals, pass interceptions and won battles that lead to a change in possession

Players Involved

  • Player: Skater credited with the takeaway

Coordinates

  • X,Y Coordinate: Location where the skater gained possession when taking the puck away

Puck Recovery

Possession gains initiated by retrieving a loose puck that was created by a missed/blocked/saved shot, an advance (e.g. dump-out/dump-in), a faceoff or a broken play

Players Involved

  • Player: Skater who recovered the puck

Coordinates

  • X,Y Coordinate: Location where skater gained possession

Dump In/Out

Actions in which a skater intentionally concedes possession by advancing the puck up ice

Players Involved

  • Player: Skater who dumped/advanced the puck

Coordinates

  • X,Y Coordinate: Location where skater released the puck

Event details

  • Detail 1: Possession Outcome (Retained, Lost)

Zone Entry

Attempts to move the puck into the offensive zone from the neutral zone

Players Involved

  • Player: Entry skater
  • Player 2: Targeted defender

Coordinates

  • X,Y Coordinate: Point of release for dumps/advances, point where puck crossed the blueline for passes and carries

Event details

  • Detail 1: Entry Type (Carried, Dumped, Played)

Faceoff Win

Faceoffs

Players Involved

  • Player: Skater who won the draw
  • Player 2: Skater who lost the draw

Coordinates

  • X,Y Coordinate: Location of faceoff dot

Penalty Taken

Infractions

Players Involved

  • Player: Skater who took the penalty
  • Player 2: Skater who drew the penalty

Coordinates

  • X,Y Coordinate: Location of infraction

Event Details

  • Detail 1: Infraction Type (e.g. Slashing, Tripping, Roughing, Hooking, ...)