Hand Path Grading — FAQ
What is Hand Path grading, in plain English?
Hand Path grading is a soft visual aid, not a definitive verdict on your swing. It looks at the path your hands traveled on the way down compared to the path they traveled on the way up and asks one simple question: does the down-path stay above the up-path (over-the-top) or drop below it (into the slot)? It assigns a traffic-light grade — 🟢 / 🟡 / 🔴 — so you have a quick "is this swing worth a closer look?" signal on the swing summary.
A 🔴 doesn't mean your swing is broken. It means the data on this one swing looks like the hands came over the top. Take it as a prompt to scroll through the video, look at the cyan (backswing) and yellow (downswing) lines on the Hand Path overlay, and decide whether it matches what you see. You're the coach; the grade is a heads-up.
How is the grade calculated?
For each frame on the way down (top of backswing → pre-impact), MotionEdge looks at where your wrist is and finds the matching point on your backswing curve at the same horizontal position. It measures the vertical gap between them.
- If the downswing wrist is higher than the matching backswing point, that frame is "over the top" by some amount.
- If the downswing wrist is lower than the matching backswing point, that frame is "in the slot."
The grade is based on the average gap across the downswing, expressed as a percentage of your shoulder-to-ankle body height measured at address. Dividing by your body height normalizes the metric so a 6'6" player and a 5'4" player are graded on the same scale — what we care about is the gap as a fraction of your own body, not in absolute pixels or inches.
The number itself is internal; you don't see it on the swing summary. The traffic light is what shows up.
What affects the grade?
The grade is a measurement, and like any measurement, it depends on the input. Things that can move the grade up or down for the same swing:
- Camera angle. Hand Path grading is designed for Down-the-Line (DTL) captures — camera placed behind you, looking down the target line. Face-on captures don't get a Hand Path grade because the geometry the calculation depends on isn't visible from that angle.
- Camera position relative to "true DTL." If the camera is set up a few feet off the target line, or tilted, your hand paths in the video aren't the same as your hand paths in the real world. This can push a borderline swing across a grade boundary in either direction.
- Wide-angle and ultrawide lenses. Wide-angle distortion bends straight lines near the edges of the frame. Your hands often travel through the distorted edge region during transition. That can stretch or compress the apparent path in ways that have nothing to do with your actual swing. If you film with the iPhone's ultrawide (0.5x) lens, expect the metric to be noisier than with the standard (1x) lens.
- Pose detection accuracy. The grade depends on the app reliably tracking your wrists frame-by-frame. Difficult conditions (poor lighting, busy backgrounds, motion blur on a fast shutter, the wrist hiding behind the body at the top) can produce noisy data. When the noise is severe enough, MotionEdge will abstain rather than show a misleading grade.
- Trimmed vs untrimmed video. A clean trim that starts at address and ends near impact gives the calculation the right frames to work with. If extra setup/aim time is included before the takeaway, the math still focuses on the swing window — but cleaner inputs give cleaner outputs.
- Small natural body motion. Tiny shifts in your stance, breathing, balance, and lateral hip motion all show up in the pose data and can shift the metric by a few percent in either direction. There's a real noise floor — a textbook-clean swing might still land near the green/yellow line.
This is why the grade is a soft aid and not a hard verdict. It's an indicator that summarizes hundreds of frames of pose data into one light — useful, but never a substitute for watching the swing.
What do the three grades mean?
| Grade | Meaning |
|---|---|
| 🟢 Green | Neutral or in the slot — your hands tracked roughly the same path up and down, or dropped slightly into the slot on the way down |
| 🟡 Yellow | Mild over-the-top tendency — the down-path sits a bit above the up-path |
| 🔴 Red | Strong over-the-top — the down-path sits clearly above the up-path through the transition |
The grade is shown next to "Hand Path" on the swing summary panel, in the Hand Path stats table, and on the Hand Path overlay panel header. You won't see the raw percentage — the grade is the actionable signal.
I don't agree with the grading on my swings. Can I adjust it?
Yes. The thresholds that turn percentages into 🟢 / 🟡 / 🔴 are configurable per device.
Path: Settings → Training → Posture Grading
In Posture Grading you can see the numeric thresholds for each graded metric, including Hand Path, and adjust where the green/yellow and yellow/red boundaries fall. If a swing you consider clean is reading yellow, you can widen the green band. If you want a more sensitive grade — flag yellow earlier — you can narrow it.
This is also where you'll see the raw percentage numbers for reference, since you need to see numbers to set thresholds intelligently. We don't show those numbers on the swing summary because the percentage by itself isn't actionable — but on the settings screen, where you're tuning, they're useful.
The thresholds you set apply to all subsequent grades and to historical swings as they're re-graded. Defaults are tuned to a corpus of labeled swings, so the out-of-box behavior is reasonable for most golfers — but the system is yours to tune.
What does "insufficient data" mean if I see it on a swing?
Sometimes the pose data for a single swing is just too noisy to produce a meaningful Hand Path measurement — for example, when the wrist is lost behind the body for several frames, or when motion blur smears the wrist position at impact. In those cases MotionEdge intentionally shows "Hand Path: insufficient data" rather than displaying a wrong grade.
If you're seeing this often, look at the capture conditions: lighting, camera angle, distance from camera, and whether your wrists are reliably visible through the full swing. A cleaner capture almost always fixes it. The Filmstrip & Video Quality FAQ has tips on this.
Why isn't the Watch hand path score shown?
The Apple Watch hardware can't achieve the spatial precision required for a reliable hand path score.
Hand path is a spatial measurement — where your hands are in 3D space at each moment of the downswing. Getting that right requires tracking position, not just acceleration. The Watch's accelerometer measures forces on your wrist. Turning those forces into a position requires double-integrating the signal over time, and that process accumulates error rapidly. Within 300–400 milliseconds (roughly the length of a downswing), the position estimate has drifted enough that the resulting score is not trustworthy enough to display with confidence.
The measurement is computed and stored — it may become useful for relative comparisons over many swings — but we won't show a number that could mislead you about your swing mechanics.
Why isn't the Watch hand path score shown?
The Apple Watch hardware can't achieve the spatial precision required for a reliable hand path score.
Hand path is a spatial measurement — where your hands are in 3D space at each moment of the downswing. Getting that right requires tracking position, not just acceleration. The Watch's accelerometer measures forces on your wrist. Turning those forces into a position requires double-integrating the signal over time, and that process accumulates error rapidly. Within 300–400 milliseconds (roughly the length of a downswing), the position estimate has drifted enough that the resulting score is not trustworthy enough to display with confidence.
The measurement is computed and stored — it may become useful for relative comparisons over many swings — but we won't show a number that could mislead you about your swing mechanics.
What causes an over-the-top (OTT) move?
OTT happens when the upper body starts the downswing instead of the lower body. The hands get thrown forward and outward rather than dropping into the "slot." Common causes:
- Rushing the transition — starting down before the backswing is complete
- Starting the downswing with the shoulders instead of the hips
- Trying to generate power from the arms rather than the kinematic chain
How do I use the video-based hand path overlay?
In video review, enable the Hand Path overlay from the overlay settings panel. When pose detection has run on the swing:
- MotionEdge draws the path your hands traveled from the top of the backswing to impact
- The arc shows whether your hands dropped down (shallow) or came over (steep)
- You can compare your hand path arc to a reference player's path
- The Hand Path % grade (above) puts a number on what the overlay shows
This is a direct visualization from camera data — what you see is what your hands actually did.
How does this compare to TrackMan club path?
TrackMan's "Club Path" measures the actual direction the clubhead moves through impact relative to the target line. The hand path overlay measures something different — the trajectory your hands traced through the downswing, not where the clubhead was at impact. They're related (shallow hand paths tend to produce in-to-out club paths) but not interchangeable. Think of hand path as measuring the cause (transition and downswing shape) rather than the effect (club direction at impact).