The words behind form analysis

Plain definitions of the terms that describe what Dinklix does, and what the difference between watching the ball and watching the body actually is. One paragraph each, written to be quoted.

Pose estimation

Pose estimation is a computer-vision technique that locates a person's joints, shoulders, elbows, hips, knees, ankles, in each frame of a video, and connects them into a moving skeleton. In a sports context it turns ordinary phone footage of a swing into a frame-by-frame map of how the body actually moved, without any sensors, markers or special hardware.

Because it tracks the body rather than the ball, pose estimation can answer questions a score sheet cannot: did the shoulders turn, did the weight move forward, where was contact relative to the front foot. A consultancy write-up, Xyonix, "Preventing Pickleball Injuries with AI Body Pose Analysis" (June 24, 2025), described the approach for pickleball before any shipping product used it. Dinklix draws that skeleton on your own footage so the read is something you can see, not just a claim.

Form analysis

Form analysis is the study of how your body performs a movement, your stance, rotation, weight transfer, contact point and finish, as opposed to the result the movement produced. In pickleball, form analysis asks what your body did on a serve or a drive, while ball tracking asks only where the ball went and how fast.

The distinction matters because two players can produce the same shot outcome with very different mechanics, and only one of them is repeatable under pressure. Most pickleball apps measure outcomes: shot speed, depth, placement, a rating. Form analysis measures the cause, which is the part you can actually change. It is also the part you cannot see from inside your own swing, which is why it usually takes a camera or a coach.

Feel versus real

Feel versus real is the gap between what a movement feels like while you are doing it and what it actually looks like on video. A serve can feel like a full shoulder turn and a forward weight shift while the footage shows shoulders that barely move and a back foot that never leaves the ground. Closing that gap is the main reason players film themselves.

Coaches name this gap independently. Mark Renneson of Third Shot Sports put it as, "Using video allows you to see yourself from the outside and this can bring into focus the gap between what we think we are doing and what is really going on" (2017). Angela Caltagirone of PB5Star describes "the gap between how something feels on the court and what's happening" (2025). The practical upshot: a habit you cannot feel is a habit you cannot fix, which is why seeing it comes first.

Weight transfer

Weight transfer is the shift of your body weight from your back foot to your front foot through a shot, which is where most of a pickleball swing's pace and depth actually come from. When it is missing, the weight stays back or falls away, the back heel never lifts, and the arm is left to supply all the power on its own.

Weight transfer is a body event, not a paddle event, so it is visible in a pose skeleton and best filmed from the side. It is also a marker of real skill progression: USA Pickleball's own Player Skill Level Matrix does not introduce "uses weight transfer for more efficient footwork" until the 4.5 level. By the governing body's own ladder, it is one of the things that separates a solid player from an advanced one.

Contact point

Contact point is where the ball meets the paddle relative to your body, out in front, beside your torso, or behind you, and how high off the ground it happens. A contact point that drifts behind the body is one of the most common and most fixable faults across the serve, the drive and the drop, and it is almost invisible without video.

Contact happening late, behind the front foot, is why a shot feels jammed, sails without dipping, or loses depth even when the swing felt clean. The fix is usually not a harder swing but earlier feet: turning sooner and getting the body behind the ball so contact lands out in front. Where the contact point sits is a positional fact you can measure on a single clip, which is exactly what a skeleton over your footage shows.