How to Film Your Running Gait at Home for a Useful Screening

You do not need a motion-capture lab to learn something useful about how you run. A modern phone camera, a bit of space, and a few minutes of setup will get you footage that a trained eye, or a tool like the CritchPitch Run Lab, can actually screen. Two-dimensional (2D) video analysis is simple, accessible, and affordable, and a systematic review of the method found it can be a reliable and valid option for many gait parameters when it is captured well.[1] The phrase to hold onto is *captured well*, because the quality of your screening is decided before you ever hit play.

This guide walks through the whole setup: where to put the camera, which frame rate to use, treadmill versus overground, lighting and clothing, and, just as important, what 2D video honestly can and cannot measure. When your clip is ready you can screen your stride with our free tool and read your own footage with clearer eyes.

Camera placement: two views, one clean setup

Almost every evidence-based running analysis protocol is built on two camera views, a side view and a straight-on view, because each plane shows different things.[2] Clinical guidance on capturing quality 2D video stresses the same fundamentals whichever view you use: keep the camera level, keep it perpendicular to the runner, and keep the whole body in frame.[4]

The side (sagittal) view

This is your most valuable angle. Set the camera at roughly hip height, exactly side-on to your line of travel so the lens looks straight across the plane of motion, not angled up the track or down toward it. The optical axis should be perpendicular to your path. Frame the shot so your entire body stays visible through at least two or three full stride cycles. The side view is where 2D shines for measuring things like knee flexion, foot strike, trunk lean, and stride timing.[2]

The front and rear (frontal) view

Position the camera dead straight-on, centered on your path, at about hip or knee height, and film yourself running directly toward and away from it. This view is used to estimate frontal-plane motion such as hip adduction, pelvic drop, and knee position. Treat these numbers as useful estimates rather than precise angles, for reasons covered below.[7] If you want to compare against a full lower-body checklist, the running cadence guide pairs well with what the frontal view tends to reveal.

Frame rate: match it to what you want to measure

Frame rate is the setting people get wrong most often, in both directions. Higher is not automatically better for everything, but for the fastest events it matters a lot. The rule of thumb: the quicker the moment you want to catch, the more frames per second (fps) you need to catch it clearly.

Cadence, step timing, and joint angles change slowly enough that a standard 30 to 60 fps clip captures them well, and reliability studies of 2D running analysis have used footage in this range.[3] Foot strike and ground contact time are different animals. Ground contact in running lasts only a couple hundred milliseconds, so a 30 fps clip may only catch a handful of frames during the entire contact, which is not enough to pin down the exact instant the foot lands or leaves. Guidance on classifying foot-strike patterns recommends a high-speed camera, preferably capturing more than 120 fps, for that reason.[5] Most recent phones can shoot slow-motion at 120 or 240 fps, which is ideal for these fast events.

A practical approach is to film your side view twice: once at 60 fps for angles and cadence, and once at 120 or 240 fps focused on the feet for strike and contact time.

Treadmill or overground: both work, note which

A treadmill makes home filming far easier because the camera stays put and the runner stays in frame. The good news is that a large systematic review and meta-analysis found motorized treadmill running is largely comparable to overground running across spatiotemporal, kinematic, and kinetic measures, with many differences being trivial in size.[6] So a treadmill screening is genuinely useful.

The caveat is the foot strike. That same review found some real, if modest, differences at the moment of contact, including a reduced sagittal foot-to-ground angle and slightly longer contact time on the treadmill.[6] Practically, that means a treadmill can make a foot strike look a touch flatter than it would be outdoors. None of this disqualifies treadmill footage; it just means you should note which surface you filmed on and avoid over-reading tiny foot-strike differences.

Lighting and clothing: make the joints easy to see

Every analysis method, human or software, depends on seeing your joints clearly. High frame rates in particular need lots of light, because each frame is exposed for a shorter time and dim footage turns into a blurry mess. Film in the brightest even light you can find, avoid backlighting from a bright window behind you, and keep the background plain and uncluttered.

• Wear fitted clothing. Baggy shorts or a loose shirt hide the hip, knee, and ankle you are trying to see.
• Choose colors that contrast with your background so limb edges stand out.
• Keep shoes consistent between sessions if you plan to compare over time.
• Fill the frame with light. Bright, even, front-facing light beats a single harsh overhead source.

What 2D video can and cannot measure

This is the honest part, and it matters for reading your own results. 2D video is a flat, single-plane record of a three-dimensional movement, and that shapes what it can tell you.

What it does well

In the sagittal (side) plane, 2D analysis is on its strongest ground. It captures cadence and spatiotemporal timing well, and markerless systems compared against marker-based gold standards have shown moderate to excellent agreement for sagittal-plane hip and knee angles, with mean errors often only a few degrees.[8] Reliability studies of 2D running analysis also report good repeatability for many sagittal measures, which is why the method is used in clinics.[2][3]

Where it is only an estimate

Reliability is not uniform. Research consistently finds it depends on the specific variable, with some measures far more repeatable than others, so no single number from a video should be treated as gospel.[3] Frontal-plane values, such as hip adduction and knee valgus seen from the front, are especially prone to error, because a 3D motion projected onto a flat image gets distorted, and small camera-angle changes shift the reading. Treat these as directional estimates, not exact degrees.[1][7]

What it simply cannot do

Two things sit outside a single camera's reach entirely. First, force: video shows motion, not the ground reaction forces or joint loads that require a force plate or instrumented setup. Second, true 3D rotation, such as tibial internal rotation or the full twist of a segment, cannot be recovered from one flat view. A systematic review of 2D gait analysis is direct that the method has limits versus 3D systems and that additional validation is needed for specific parameters.[1] Knowing this keeps a screening in its lane.

A quick capture checklist

1. 1Clear a bright, uncluttered space, treadmill or a straight overground path.
2. 2Mount the phone on a tripod, level, at about hip height.
3. 3Set up the side view first: camera perpendicular to your path, whole body in frame.
4. 4Film 60 to 90 seconds once your pace has settled into a steady rhythm.
5. 5Add a 120 to 240 fps side clip focused on the feet if you want foot strike and contact time.
6. 6Reposition dead straight-on and film a front or rear view the same way.
7. 7Note the surface, your shoes, and your pace, then upload to screen your stride.

That is the whole job. Good footage does most of the work of a good screening, and the settings above are all it takes to get there. Once you can see your stride clearly, the next step is knowing what to look for, and a common first find is over-striding, which the overstriding guide breaks down in detail.