Curling is a sport where two teams take turns sliding stones down a sheet of ice towards a target. The most common form of the game features teams with four players, with each player getting two shots per round, known in curling as an “end”. When one player shoots, two of their teammates follow the stone down the ice, sweeping its path with brooms to direct it to the desired position. Teams score by having the stone closest to the center of the target at the completion of the end, one point for each stone inside of the opponent’s closest stone. After a set number of ends (usually 8 or 10), the team with more points is the winner.
A primer on curling rules (both 4-person and mixed doubles) can be found here.
Mixed Doubles Curling presents a unique analytics opportunity described as the "Wild West of Curling" due to its distinctive format and strategic complexity. Unlike traditional curling, Mixed Doubles features:
The full rules of all forms of curling can be found here.
This Data Challenge focuses on power play optimization - when to use it, how to execute it, and how to defend against it - using data from international competition results that show final stone positions.
Since mixed-doubles curling is fairly new, there are a variety of topics that have not previously been studied. Below are a few example topics. However, we note that this list is far from exhaustive. Participants should feel free to study any aspect of mixed-doubles curling that interests them, provided that the analysis would be useful to a coach, curler, or some other member of the curling community.
Student Deliverables Could Include:
Shot-by-shot stone positions from Curlit are available on Github for the following competitions:
These data correspond to the shot-by-shot images that are found e.g. in this 2025 World Mixed Doubles Championship Results Book. See Curlit for results from other competitions.
There are 26,000+ observations in the data which correspond to one throw during one end of a game in one of the competitions listed above. The data are stored in six CSV files and the columns contain the following information:
Competition.csv:
CompetitionName – Name of the competition.Place – City/Country where the competition was
competed.Venue – Venue where the competition was
competed.Competitors.csv:
CompetitionID – ID for the competition (from
Competitions.csv).TeamID – ID for the team. Note: Team IDs across
different competitions are associated with nations, not
necessarily the individuals on that team.NOC – National Olympic Committee, the nation of
the competing team.Reportingname – Name of the competing athlete
(LAST NAME First Name) .Ends.csv:
CompetitionID – ID for the competition (from
Competitions.csv).SessionID – ID of the draw/round for the
match.GameID - ID for the match. Note: A unique ID
for a match can be made from a combination of Competition,
Session and Game IDs.TeamID – ID for the team (from
Teams.csv).EndID – ID for an end of the match.Result – The points scored by that team in the
end.PowerPlay – A flag for when a team uses their
power play in an end. A value of 1 corresponds to pre-placed
stones being moved to the right side, 2 being pre-placed stones
are moved to the left side. An empty value means the team is not
using their power play.Games.csv:
CompetitionID - ID for the competition (from
Competitions.csv).SessionID - ID of the draw/round for the
match.GameID - ID for the match. A unique ID for a
match can be made from a combination of Competition, Session and
Game IDs.GroupID – In competitions where teams are split
into groups for play, the ID of the group for the teams in this
match.Sheet – The name of the sheet on which the
match is played.NOC1 – National Olympic Committee of the first
competing team.NOC2 - National Olympic Committee of the second
competing team.ResultStr1 – Final Score of Team 1.ResultStr2 – Final Score of Team 2.LSFE – Last Stone First End. This column
indicates which team threw the last stone in the first end of
the match. In curling parlance, this is called starting with
“the hammer”. A 0 value means that NOC2 threw the last stone in
the first end, a 1 means that NOC1 threw last.Winner – Indicates the winning team. 0
indicates that NOC2 won the match, 1 indicates that NOC1 won the
match.TeamID1 – ID for the team corresponding to NOC1
and ResultsStr1 (from Teams.csv).TeamID2 - ID for the team corresponding to NOC2
and ResultsStr2 (from Teams.csv).Stones.csv:
CompetitionID – ID of Competition in which
match was competed (from Competitions.csv)SessionID – ID of the draw/round for the
match.GameID – ID for the match. A unique ID for a
match can be made from a combination of Competition, Session and
Game IDs.EndID – The end in which the shot is taking
place.ShotID – ID for the shot. Within a particular
match and end, shots occur in ascending order of the ShotID
values. (i.e. lower ShotIDs are thrown first)/TeamID – ID for the Team (from
Teams.csv).PlayerID – ID for the player taking the shot (1
or 2).Task – The type of shot that the player is
throwing, or the objective of the shot. Details to come.Handle – The turn of the stone as it is thrown.
A 0 value means the shot is turning clockwise, a 1 indicates
counterclockwise.Points – An assessment of the execution of the
shot, ranging from 0 to 4. A 4-point shot is one that has been
ascertained to have been perfectly executed to the player’s
intention, while a 0 is a shot that totally failed in its
intended result. Note that these points are not the same as
points awarded to teams after ends, but simply an evaluation of
a shot’s effectiveness.TimeOut – Binary variable for whether a time
out was called before the shot.Stone_{i}_x – The x position of stone i on the
sheet, after the stone for the row has been thrown. Stones can
take x values in (0,1500). The value 4095 is a sentinel value
indicating that the stone has been knocked off the sheet and is
no longer in play. The value 0 indicates that the stone has not
yet been thrown in this end.Stone_{i}_y – The y position of stone i on the
sheet, after the stone for the row has been thrown. Stones can
take y values in (0,3000). Again, the value 4095 indicates the
stones have been knocked off the sheet and aren’t in play, and
the value 0 indicates the stones have not yet been thrown.A Note on stones: Stone i does not necessarily correspond to ShotID i. The first six stones are the stones of the team that is throwing first in the end, while stones 7-12 are for the other team in that end. Stone_1 and Stone_7 are the pre-placed stones, and teams throw the rest of their stones in order. For example, a team going first in an end would have stones 2-6 and throw them in ascending order.
Teams.csv:
CompetitionID - ID for the competition (from
Competitions.csv).TeamID - ID for the Team (from
Teams.csv).NOC - National Olympic Committee, the nation of
the competing team.Name – Name of the country corresponding to NOC
abbreviation.More data from additional competitions may be given at later time if it becomes available.
Given the challenge's popularity, we would like to know the number of participating teams to adequately prepare and recruit the necessary judges. Please register for the data challenge by December 15, 2025 at:
Data challenge registration form
This registration is for the data challenge only. Registration for the conference will be available at a later date.
Students must submit a zip file containing:
The CSAS Data Challenge is open to students only. You must be enrolled as a high school, undergraduate, or graduate student at some point during the current academic year. Participants must register using their school email address.
Teams must enter one of the following two tracks:
To be eligible for the High School / Undergraduate ALL members of the team must consist of either high school and/or undergraduate students. Each team can have up to 3 members. The team captain, if 18 years old or over, should fill out the registration form for the entire team using their school email addresses. If all team members are under 18, a faculty advisor needs to be the point of contact and register for the team.
A panel of judges from across academia and the sports industry will judge your submissions based on the following:
Finalists (Six teams: three high school/undergraduate and three graduate) will be invited to present their work at the 2026 CSAS and will receive some travel support and have their registration fees waived. Winning teams will have the opportunity to showcase their team’s work to data scientists and practitioners at the 2026 U.S. Olympic & Paralympic Performance Innovation Summit (location and date TBD but likely mid/late 2026). The winning teams will also receive a cash prize and a plaque.
An introductory workshop led by members of the CSAS organizing committee and members of the curling industry will be planned in late October or early November. We will give a short introduction and spend most of the time on Q+A.