What a camera can see in a pistol squat
A pistol squat is a full squat on one leg with the other held out in front. From the side the camera resolves the working knee angle at the bottom and how far the trunk pitches forward to stay balanced, which is the pair of numbers that decides the score. The identity gate is unusual here: it requires a large left-to-right knee difference. In every other squat rubric that difference is penalised, and in this one its absence means the clip is really a back or front squat and should not be scored as a pistol.
How ScoreRep scores a pistol squat
The rubric declares 4 scored criteria. The strongest single camera angle is the left side view. Every criterion below is available from that angle.
The score is the points earned divided by the points the recorded angle could support, so a clip filmed from a limited angle is never quietly marked down for what the camera could not see. At least 60% of the applicable weight has to be measured before a result is ranked at all.
| Criterion | Points | Measured from the athlete’s pose | Full marks when |
|---|---|---|---|
| Depth | 35 | smallest knee angle in the rep | 85° or less |
| Torso stability | 25 | largest torso lean away from vertical | 30° or less |
| Control | 25 | ratio of lowering time to lifting time | 1 or more |
| Consistency across the set | 15 | drop in rep speed between the fastest and the slowest rep | 10% or less |
Criteria that depend on the camera angle
These criteria are excluded, not estimated, when the recorded angle cannot support them. That is why the same set filmed twice from different positions can produce two valid scores out of two different totals.
- Depth — needs a left side, right side, left oblique and right oblique view.
- Torso stability — needs a left side, right side, left oblique and right oblique view.
Does this pistol squat rep count?
ScoreRep answers that in two separate steps. First, the clip has to read as a pistol squat at all: every check in the table below has to pass, and a reading the camera could not take counts as a fail, never as a pass. A clip that fails is returned unranked with the reason attached, rather than scored low against the wrong rubric, and no template can override a failed check.
Second, the scoring criteria award points. Falling short of a full-marks threshold costs points, not the rep. Every number here is read from version 1 of the pistol squat rubric, the same file the scoring engine runs.
- Compare the pistol squat standard with the other 29 movements on the movement standards page.
- How the rep counter decides between a valid rep and a no-rep, station by station.
| Reading | Has to be | Otherwise rejected because |
|---|---|---|
| Vertical travel of the hips across the rep | 0.08 or more | no squat was detected |
| Whether the hip crease passes below the knee | yes | the hip crease never passes below the knee |
| Whether the hands finish above the shoulders | no | the rep finishes overhead, so it is not a pistol squat |
| Largest left-to-right difference in knee angle | 15° or more | both legs bend together, which reads as a back or front squat rather than a pistol |
What the recording has to contain
Eligibility decides whether a result can enter a leaderboard. Failing it does not block the feedback — the clip is still scored and returned, but marked unranked with the reason attached, so a bad camera position never silently becomes a bad score.
- A pose has to be detected in at least 55% of the frames.
- The camera angle has to be one of: left side, right side, left oblique, right oblique, front and back.
- At least 1 complete repetition has to be detected.
What is deliberately not scored
Identity, repetition counting, eligibility and the score are derived only from the athlete’s pose. Measurements that come from tracking an external object — the path of a barbell, for example — are drawn on the review overlay but are deliberately outside the scoring contract, so they never affect the number.
Reading the result
Pistol depth is limited by balance far more often than by strength, and the two look different on camera. If the torso score collapses in the same reps where depth collapses, the athlete is falling forward to find the floor, and the fix is usually a counterweight or a box rather than more single-leg volume. Consistency across the set is scored separately for the same reason: the fifth rep of a pistol set is a different movement from the first.
A technique score is educational feedback about visible movement. It cannot tell you whether a load is safe for your joints, whether pain means something, or whether a programme suits you. Those judgements need a coach or clinician who knows your history.