metadata

license: mit
task_categories:
  - video-classification
tags:
  - baseball
  - sports
  - video-classification
  - computer-vision
size_categories:
  - 100K<n<1M

Dataset Card for Picklebot130k

130 thousand video clips of balls and strikes from MLB games from the 2016 season through the 2023 season.

Dataset Details

Dataset Description

The dataset consists of roughly 130 thousand video clips of balls and strikes in .mp4 format, resized to 224x224 resolution.

Curated by: Henry Freed
License: MIT

Dataset Sources [optional]

Repository: The original project that this dataset was compiled for can be found here on github.
Demo: The demo for a neural net trained on this dataset can be found here on huggingface spaces.

Uses

The dataset was originally collected to call balls and strikes using neural networks. There are many other potential use cases, but they would almost certainly require relabeling. For more videos and more complete information about each pitch, see Picklebot-2M.

Dataset Structure

The dataset is structured as a .tar.xz file of all the videos. The labels are contained in three .csv files (train, val, test).

See the structure of the .csv files in the dataset viewer, a few notes on that:

Zone is the pitch location. The zones on Savant are presented as 1-9 for inside the zone, and 11-14 for outside the zone. We renumbered them by subtracting 1 from the inside zone values and 2 from the outside zone values, resulting in a continuous range of 0-12, where 0-8 represents inside the zone, and 9-12 represents outside the zone.
The dataset includes the following pitch labels: FF (Fastball/4-seam), SI (Sinker/2-Seam), FC (Cutter), CH (Changeup), FS (Split-finger), FO (Forkball), SC (Screwball), CU (Curveball), KC (Knuckle Curve), CS (Slow Curve), SL (Slider), ST (Sweeper), SV (Slurve), KN (Knuckleball), EP (Eephus), FA (Other), IN (Intentional Ball), and PO (Pitchout).

Source Data

The source data were scraped from Baseball Savant's Statcast Search. It's a pretty powerful search page, and a lot of fun to play around with.

Data Collection and Processing

After downloading the videos, they were cropped from 1280x720 to the middle 700x700 pixels and downsampled to 224x224 resolution at 15 fps (this can all be done using one ffmpeg command). Some of the longer clips where there was a lot of noise (shots of the crowd, instant replays, etc.) were trimmed (mostly by hand) down to a more manageable length.

Who are the source data producers?

Baseball Savant and MLB/the broadcasters (whoever it is) originally created the videos.

Bias, Risks, and Limitations

It's important to note that only balls and called strikes were collected. No swinging strikes, foul balls, hit by pitches, or anything else are included in the dataset. Additionally, most pitchers and batters are right handed, and nothing was done to try and balance that in this dataset.