Data

Sample datasets

• College Football 2024
  • Description: Lists team stats for all 135 D1 College Football teams for the Fall 2024 season. I assembled this summary data from more detailed data obtained from Massey Ratings.
  • CSV
• Colors
  • Description: This is a small, manufactured data set of RGB values that are close to primary colors; there are also a few missing values values. The intention is to illustrate imputation with, for example, a KNN imputer.
  • CSV
• NCAA Basketball data (multiple files for sports analytics)
  • The Big South 2025 Regular Season
    • Description: Game by game results of the Mens’s 2025 Big South regular season.
    • CSV
  • Big South Games with Eigen ratings
    • Big South games with score differences and Eigen ratings for the 2023/24 Season
    • CSV
  • Big South Games with Massey ratings
    • Big South games with score differences and Massey ratings for the 2023/24 Season
    • CSV
  • Paired Tournament Games
    • Lists all Men’s NCAA Tournament games from 2010 through 2023. Each game lists the difference of the Massey ratings between Team 1 and Team 2, as well as the seed difference between the teams. There’s also a Boolean label indicating whether Team 1 defeated Team 2. The intention is illustrate a machine learning approach to sports prediction and each game appears twice to maximize the data set.
    • CSV
  • Processed Kaggle data.
    • There are four files here - two for the men’s tournament and two for women. Prior tournament games record actual past tournament games dating back to 2010 up through the 2024 tournament. Each game has a slew of team stats and a label indicating who won the game. Potential tournament games lists all possible pairs of tournament games from the 2025 tournament with the data but no labels. The idea was to generate a Kaggle submission file the 2025 submission.
    • Men’s priors
    • Men’s potentials
    • Women’s priors
    • Women’s potentials
• Palmer Penguins
  • Reference
  • CSV
• Wine
  • Description: This is a modified version of the data obtained from SciKit Learn’s sklearn.datasets.load_wine function. I simply renamed the “target” variable to “variety” and placed it all in one CSV file to obtain a nice classification example.
  • CSV