All Collections
Using Uplift Capture
Tips for capturing and using bat data
Tips for capturing and using bat data

How to use Uplift's bat tracking to enhance your hitting captures.

Written by Matthew Kowalski
Updated over a week ago

In addition to body keypoints and kinematics, Uplift Capture can now track the movement of the bat in baseball and softball hitting captures. This data can help identify the relationship between the bat and the body during swing assessment as well as calculate helpful coaching and performance metrics like bat speed, attack angle, and various other metrics.

Capturing the bat

Tracking the bat accurately requires that the full bat is in frame for the entirety of the capture. If the bat moves out of the camera frame at points along the recording (such as during extension), Uplift Capture’s computer vision algorithms will not be able to track the tip of the bat accurately.

To ensure you’re able to capture the bat through the entirety of the capture:

  1. Move the cameras back. Make sure you set up the cameras at a sufficient distance and at an adequate angle to see the entire swing motion.

  2. Do some practice swings. During setup (and before completing calibration), have someone do a few practice swings while checking both devices to make sure the framing leaves plenty of room for the bat path.

  3. Give yourself some buffer. You may have a perfect setup for the first athlete you’re capturing, but may run into issues with subsequent taller athletes.

Using bat data

Currently, Uplift generates 3D keypoints for 2 points on the bat - the base, and the tip. This data is available in our raw data CSV files (available via webhooks or via direct download in the web application). Bat XYZ data are located at the end of the set of 3D keypoints, and just before kinematic metrics (starting with head_extension).

There is also an extra column titled “ApprovedByBatQA”. While the existing “ApprovedByBiomechQA” column deals with just body keypoints (and passes and fails based on whether we detect each keypoint through the entire movement), “ApprovedByBatQA” deals with the bat keypoints, and checks things such as whether the bat is in frame for a sufficient proportion of the capture and if the bat is tracking realistically in 3D. If this column indicates that bat keypoints did not pass QA, this means that some characteristics of the bat data were insufficient for us to include in our automated reports. However, the bat keypoints and metrics will always still be computed and will still be the user’s decision whether those data are still appropriate for their purposes.

If you do not currently have bat tracking added to your account but are interested in learning more, contact us at

Did this answer your question?