Understand the Metrics

Tap any metric to see what it means, the target range, and the research behind it. Colors match the boxes on your video.

Green = Tour range Yellow = Playable (about a 7 handicap) Red = Work on it Not graded at this position

How to read the numbers

0 = your address position. Turns and Hip Sway are measured from where you set up.

+ = back, away from the target. − = toward the target (open).

Face-on turns: 0° = square to the camera. 90° = the two shoulder joints line up one in front of the other (a full turn); past a full turn the number goes above 90°.

Turns: shoulders go 0° → about +90° at the top, count back down in the downswing, pass 0, and are about −10° to −35° (slightly open) at impact. Hips do the same at about half the size: about +45° at the top and −20° to −50° at impact.

Hip Sway: stay centered in the backswing (within about 1″). Starting down, a small move toward the target is good, and at impact the hips are about 1–5″ toward the target (−1″ to −5″).

Head arrows: Head ↕ = up (+) / down (−), best seen down the line. Head ↔ = side to side, face-on: + = back, − = toward the target. Head ↔ Ball = toward (+) / away from (−) the ball, down the line. The small word under each number (up, down, back, to target) tells you the direction.

Hands: where your hands are vs address, down the line: + = out toward the ball, − = in toward your body.

Wrist angles are not measured yet: phone video can’t see the club and the wrist joints are too small to track reliably. The research on wrist hinge (Fedorcik 2012) measured the angle between the club and the forearm, so it needs the club. Wrist angles will come once club tracking is added.

Angles like Forward Bend, Lead Arm and Knee Flex are plain body angles (180° = straight), not measured from address.

Ask the AI Coach

How accurate is phone video?

Swing AI measures from flat phone video. Research shows 2D angles follow the same pattern as lab 3D systems but can differ by several degrees (Smith 2016). Best-practice single-camera AI can come within about 2° of lab sensors (Sportsbox AI). Film in good light, with the full body in frame, camera at hip height.

Sources

Meister DW, Ladd AL, Butler EE, Zhao B, Rogers AP, Ray CJ, Rose J (2011)

Rotational biomechanics of the elite golf swing: benchmarks for amateurs

Journal of Applied Biomechanics 27(3):242–251 (Stanford University)

Key finding: 3D motion capture of 10 male professionals. Peak upper-torso turn 94–99° (±6°), peak pelvic turn 44–46° (±6°), peak X-factor 52–56° (±4°), shoulder tilt at impact 24–25° (±3°), pelvic tilt at impact 10–12° (±3°) across easy to hard swings. Peak X-factor, X-factor at impact and peak upper-torso turn were highly correlated with clubhead speed. The downswing started with the pelvis reversing, then the upper torso.

Myers J, Lephart S, Tsai YS, Sell T, Smoliga J, Jolly J (2008)

The role of upper torso and pelvis rotation in driving performance during the golf swing

Journal of Sports Sciences 26(2):181–188

Key finding: 100 recreational golfers with their own driver. More separation between upper torso and pelvis at the top and at the start of the downswing was moderately related to higher ball speed (r ≥ 0.50).

Cheetham PJ, Martin PE, Mottram RE, St Laurent BF (2001)

The importance of stretching the "X-Factor" in the downswing of golf: the "X-Factor Stretch"

Optimising Performance in Golf, Australian Academic Press

Key finding: 10 highly skilled vs 9 less skilled golfers. Shoulder–hip separation at the top was not different between groups, but skilled players increased it by 19% at the start of the downswing vs 13% for less skilled players. Sequence matters more than a bigger turn.

Okuda I, Gribble P, Armstrong C (2010)

Trunk rotation and weight transfer patterns between skilled and low skilled golfers

Journal of Sports Science and Medicine 9:127–133

Key finding: 13 skilled (avg handicap 0.8) vs 17 low-skilled golfers, 3D cameras and force plates. Skilled players turned the pelvis 28.4° + 19.1° and the upper trunk 51.7° + 54.7° from address to the top. Skilled players shifted weight to the trail foot earlier in the backswing and to the lead foot earlier in the downswing. Summarises earlier work: about 80% of weight on the trail foot at the end of the backswing and 81–142% of body weight on the lead foot at impact.

Ball KA, Best RJ (2007)

Different centre of pressure patterns within the golf stroke I: Cluster analysis

Journal of Sports Sciences 25(7):757–770

Key finding: 62 golfers from professional to high handicap. Two weight-transfer styles: "Front Foot" and "Reverse". Both start evenly between the feet, move to the back foot in the backswing and forward in early downswing. Both styles appeared at every skill level, so neither is an error.

Ball KA, Best RJ (2007)

Different centre of pressure patterns within the golf stroke II: Group-based analysis

Journal of Sports Sciences 25(7):771–779

Key finding: For Front Foot golfers, a larger and faster forward pressure shift in the downswing went with more clubhead speed. For Reverse golfers, different timing measures mattered.

Kim SE, Koh YC, Cho JH, Lee SY, Lee HD, Lee SC (2018)

Biomechanical effects of ball position on address position variables of elite golfers

Journal of Sports Science and Medicine 17:589–598

Key finding: 11 male tour professionals with a 5-iron, 3D motion capture. Trunk flexion at address was 41.5°. Moving the ball closer made the trunk, hips and knees more upright; moving it farther away bent them more. Cites Zheng et al. (2008): female pros used about 3° less trunk flexion than male pros.

Smith AC, Roberts JR, Wallace ES, Kong P, Forrester SE (2016)

Comparison of two- and three-dimensional methods for analysis of trunk kinematic variables in the golf swing

Journal of Applied Biomechanics 32(1):23–31

Key finding: 2D (flat video) and 3D trunk angles had similar shapes through the swing, but differed by about 6.5° in trunk flexion at the top, 8.7° in side bend at impact and about 16° in X-factor.

Nesbit SM, McGinnis R (2009)

Kinematic analyses of the golf swing hub path and its role in golfer/club kinetic transfers

Journal of Sports Science and Medicine 8:235–246

Key finding: Measured the path of the hands (the "hub" of the swing) in golfers of different skill. The hand path was not a circle: its radius changed through the swing, with a three-phase pattern in the downswing. A non-circular hand path was shown to transfer energy to the club better than a circular one.

Fedorcik GG, Queen RM, Abbey AN, Moorman CT III, Ruch DS (2012)

Differences in wrist mechanics during the golf swing based on golf handicap

Journal of Science and Medicine in Sport 15(3):250–254

Key finding: 28 male golfers, 3D motion capture with markers on the club and forearm. With the lead forearm parallel to the ground in the downswing, low handicappers held about 86° between club and forearm vs about 70° for high handicappers (more retained hinge). At impact, low handicappers had less lead-wrist radial deviation (about 8° vs 15°).

Cheetham P (Sportsbox AI) (2023)

Sportsbox AI accuracy

Sportsbox AI technical note

Key finding: Single-phone 3D swing analysis compared with the AMM3D sensor system on 30 golfers: about 2° average difference for turn and bend angles and 0.4 in for sway and lift, when camera setup and lighting follow best practice.

Sportsbox AI (2024)

What is early extension? (Pelvis Thrust)

Sportsbox AI help centre (vendor data, not peer-reviewed)

Key finding: Defines Pelvis Thrust as the down-the-line movement of the pelvis centre from address, in inches: 0 at address, + = toward the ball, − = away from it. Early extension = the pelvis thrusting toward the ball in the downswing, which takes away room for the arms. Impact is compared with a tour range built from over 100 PGA and LPGA professionals; for a male tour player of Bryson DeChambeau’s build the range is −0.9″ to +1.7″.

Titleist Performance Institute (2024)

The 90/90 test

TPI golf fitness screen (mytpi.com)

Key finding: Screens trail-shoulder external rotation in a golf posture: the trail upper arm raised to 90° with the elbow bent 90°, then the forearm rotated back. Passing means the forearm reaches at least parallel to the spine; not reaching it points to limited external rotation, which TPI links to swing faults such as a flying elbow and early extension.

Mitchell K, Banks S, Morgan D, Sugaya H (2003)

Shoulder motions during the golf swing in male amateur golfers

Journal of Orthopaedic & Sports Physical Therapy 33(4):196–203

Key finding: Video motion analysis of male amateur golfers, describing shoulder motion as elevation, horizontal adduction and external rotation through the swing. The trail shoulder moves into external rotation toward the top of the backswing, and senior golfers reached clearly less trail-shoulder external rotation at the top than younger (college) golfers.

Zheng N, Barrentine SW, Fleisig GS, Andrews JR (2008)

Kinematic analysis of swing in pro and amateur golfers

International Journal of Sports Medicine 29(6):487–493

Key finding: 3D motion analysis of professional and amateur golfers grouped by handicap. At the top of the backswing professionals showed more trail (right) shoulder external rotation, more lead shoulder horizontal adduction and more trunk rotation, with values generally rising with skill level.