Filming Your Stroke: What Swim Video Analysis Can and Can't Fix

Ask a group of age-group triathletes what their stroke looks like and most will describe what it feels like. Long, relaxed, high elbow, legs up. Then show them video, and many are surprised. The legs are lower than they thought, the head lifts on every breath, the hand crosses in front of the face. That gap between perceived and actual technique is one of the main reasons swimmers plateau. Video is the obvious way to close it.

This article is upfront about the limits of the evidence. There's very little controlled research on video feedback for adult triathletes. What follows combines the small amount of direct evidence, the wider biomechanics literature and practical reasoning, which is labelled where it's an assumption rather than a finding. No specific products are recommended. A phone in a waterproof case is enough for most of what's described here.

Why seeing your stroke matters

The case for technique work in triathletes rests mostly on one striking finding. In 1990, Huub Toussaint compared six competitive swimmers with five triathletes at the same power output and found that the swimmers turned about 62% of their power into useful propulsion, against about 44% for the triathletes. At the same power, the swimmers covered 1.28 m per stroke and the triathletes 0.99 m. It's an old, small study, but the message is clear. A large share of a typical triathlete's effort in the water doesn't move them forward.

You can't fix what you can't see, and most stroke faults aren't visible from inside the water. That's the core reason to film. It's also inference, not a tested finding, that seeing the fault makes it easier to fix. The direct evidence is next.

What the research on video feedback actually shows

There's very little of it, and almost none in triathletes.

The most relevant adult study is by Giannousi, Mountaki and Kioumourtzoglou (2017). Sixty male novice swimmers, average age about 19, were randomly assigned to four groups for seven weeks: self-modelling (watching video of their own swimming), expert modelling (watching a skilled swimmer), verbal feedback, and a control group. Self-modelling improved technique the most. But there were no differences between groups in 25 m swimming speed.

Two points stand out. First, watching your own stroke appears to help technique more than watching an expert or being told what to change, at least in novices. Second, technique change didn't show up as speed within seven weeks. That fits with what many swimmers experience. A new stroke pattern often feels slower before it feels faster, and it may need to be trained at pace before it produces faster times.

The broader evidence is weak. A 2025 scoping review by Minkels and colleagues examined motor-learning methods for teaching swimming to 5- to 12-year-olds. It found only four video-based studies among 23 included. Across all the studies, 87% lacked retention or transfer tests, so they couldn't show whether improvements lasted. The authors described the literature as "scarce, generally atheoretical and of questionable methodological quality." That review is about children learning to swim, not adult triathletes refining a stroke, but it shows how little rigorous research exists on how swim technique is best taught.

What we don't know: whether video feedback makes adult triathletes faster, how much, how often you need it, or which type of feedback works best for experienced but inefficient swimmers. No study has tested this directly.

Assumptions this article makes

Since the direct evidence is thin, here are the reasonable assumptions behind the practical advice, stated explicitly:

1. Seeing a fault makes it easier to correct than relying on feel. This is consistent with the novice self-modelling finding, but not tested in triathletes.
2. Fixing obvious mechanical faults such as sinking legs, a crossing hand entry or a dropped elbow is likely to improve propulsive efficiency, given Toussaint's evidence that triathletes lose a lot of power.
3. One change at a time is easier to learn and easier to evaluate than several at once.
4. Technique changes need repeated practice at speed before they reliably improve times, consistent with the lack of speed change in the seven-week study.
5. There's no single correct stroke. In a small study of eight male swimmers, Deschodt and colleagues (1999) found considerable variation in movement patterns between individuals, with no universal "style." Video should be used to fix clear faults, not to copy someone else's stroke exactly.

What video can show you

From reasonable footage, you can usually see:

• Body position. Are hips and legs near the surface, or angled down? Sinking legs increase drag.
• Head position. Is the head neutral, or lifted, or does it rise to breathe? A lifting head often drops the hips.
• Hand entry. Do hands enter roughly in line with the shoulder, or cross the midline in front of the face?
• The catch. At the start of the pull, does the forearm get vertical with the elbow fairly high, or does the elbow drop so the hand slips through the water?
• Rotation. Is there some roll through the hips and shoulders, or is the body flat, or over-rotating?
• Kick. Is it compact and from the hip, or a large knee-driven kick, or a scissor kick when breathing?
• Timing. Is there a pause with both arms in front (catch-up), or a dead spot in propulsion?
• Stroke count and rate. Counting strokes per length and timing stroke cycles gives you numbers to compare over time.

The kick deserves a note. Deschodt and colleagues found that adding the legs increased 25 m sprint speed by about 10% in eight male swimmers. That seemed to come mainly from changing the path of the hand rather than from direct leg propulsion. If footage shows a messy kick, it may be affecting your arm stroke too.

What video can't show you

Video shows shapes and timing. It can't directly show:

• Forces and propulsion. You can see a hand moving backwards, not how much water it's holding.
• Drag. You can see body angle, not measure resistance.
• Whether a change helps you. A stroke that looks "better" might not be faster for your body, flexibility or fitness. Only testing tells you.
• What happens under fatigue or in open water, unless you film in those conditions.

That's why video works best paired with a simple pace test such as critical swim speed, covered in TriForward's companion article. Video tells you what to change, and testing tells you whether the change worked.

How to film usefully

Get permission and respect others

Before anything else, ask the pool. Many venues have rules about cameras, particularly where children are present. Film in a lane you've been allocated, at a quiet time if possible, and make sure no other swimmers are in frame. If a pool doesn't allow filming, a coached session where filming is part of the service is the alternative.

Angles, in order of usefulness

1. Side view, underwater, at mid-pool. Best single view: body position, catch, kick and timing.
2. Front view, underwater, from the end wall as you swim towards the camera. Shows hand entry, crossing and symmetry.
3. Above water, from the side. Head position, breathing and recovery. A friend on the deck is enough.

Practical tips

• Hold the camera at or just below water level for underwater shots, close to the wall, and keep it still.
• Film several lengths and use slow-motion playback. One length can mislead.
• Film at three speeds: easy pace, threshold or CSS pace, and at the end of a hard set. Many faults only appear when tired or when trying to go faster.
• Don't wear toys for the analysis footage. A 2023 study by De Matos and colleagues found paddles and fins "deeply" modify stroke structure. Film your normal stroke.
• Use consistent angles each time so you can compare like with like.

Turning footage into change

The step most people skip is deciding what to do with the footage. A simple process:

1. Watch without judging first. Just note what you see, in order: body position, head, entry, catch, rotation, kick, timing.
2. Pick the one fault most likely to be costing you. Body position and head position often come first, because they affect everything else. That's coaching logic, not a tested rule.
3. Choose one simple cue or drill that targets it, and practise it in short repeats with full attention, as described in "The 45-Minute Pool Session."
4. Follow each focused repeat with normal swimming that applies the change.
5. Re-film after 4-6 weeks from the same angle and compare side by side.
6. Retest your pace. If the stroke looks better and pace is stable or improving, keep going. If pace has dropped for more than a few weeks, reassess.

Expect your times to dip while a change beds in. The seven-week novice study saw technique change without a speed change, and that's a normal stage.

Common faults in age-group footage

These are the patterns coaches most often report seeing in adult triathletes' footage, with the kind of focus usually used to address them. This is coaching experience, not research. No controlled study has ranked these faults or tested these fixes in triathletes.

• Sinking hips and legs. Often linked to a lifted head or a weak, knee-driven kick. Common focus: head neutral with eyes looking down, and a smaller, steadier kick from the hip. Pull buoys hide this fault rather than fix it, which is one reason not to film with one.
• Lifting the head to breathe. Breathing by rotating the body and turning the head, with one goggle staying in the water, rather than lifting forward. Common focus: breathe "into the pocket" beside your head and return the face down early.
• Hands crossing the midline. Entering in front of the face causes a snaking body line. Common focus: enter in line with the shoulder, imagining the hands on two parallel tracks.
• Dropped elbow at the catch. The hand pushes down or slips back without holding water. Common focus: early in the pull, bend at the elbow so the forearm points down while the elbow stays relatively high. This takes time and some shoulder mobility, so build it gradually.
• Scissor kick on breathing. The legs split wide as you rotate to breathe, which increases drag. Often a balance problem that improves as rotation and head position improve.

If you spot several of these, start with head and body position, then entry, then the catch. The reasoning is that position affects everything downstream. That's a sensible order rather than a proven one.

Filming beyond the pool

Most technique faults are easiest to see in the pool. But triathletes race in open water, often in a wetsuit, with sighting and other swimmers around them. If you can get footage from a supervised open-water session or a training race, with permission and safety cover, look for how your stroke changes when you sight. Many swimmers lift the head high and drop the hips on every sighting stroke. TriForward's article on open-water sighting covers why sighting and race-start stress change the stroke. Footage is the simplest way to see whether it's happening to you.

DIY, coach analysis and wearables

There are three broad routes. None has been compared head to head in triathletes, so this is practical guidance rather than evidence.

• DIY filming is cheap and repeatable. Its weakness is interpretation. Without training, it's easy to miss the main fault or fixate on a minor one.
• Coach-led analysis, from the deck or from underwater footage, adds expert interpretation. A good coach can often identify one or two priority faults quickly and give a cue that works for you. It costs money, and quality varies, so ask what qualifications and experience the coach has with adult or triathlon swimmers.
• Wearable sensors and apps can count strokes and estimate stroke rate and distance per stroke, and some estimate other metrics. These give useful trend data across sessions, but they don't show what your body is doing, and accuracy varies between devices. Check a few lengths' stroke counts manually against what the device reports before trusting it.

A sensible combination for most age-groupers is DIY footage every month or two, plus occasional coach analysis at key points, with a wearable or manual counts used for trends.

When paying for expert eyes is worth it

These situations are where outside help is most likely to pay off. That's TriForward's judgement, not a tested threshold:

• You've done a structured 8-12 week block with regular testing and your pace hasn't moved.
• You've filmed yourself and can't tell what the main problem is.
• You're new to swimming, so habits are forming now and are easier to shape than to change later.
• You've changed your stroke and developed shoulder pain. In that case, see a physiotherapist as well as a coach.

References

• Toussaint HM. Differences in propelling efficiency between competitive and triathlon swimmers. Medicine & Science in Sports & Exercise, 1990;22(3):409–415. PubMed 2381311
• Giannousi M, Mountaki F, Kioumourtzoglou E. The effects of verbal and visual feedback on performance and learning freestyle swimming in novice swimmers. Kinesiology, 2017;49(1). Kinesiology journal record
• Minkels C, van der Kamp J, de Vries R, Beek PJ. Learning how to swim in 5- to 12-year-old children: a scoping review of evidence-based motor learning methods. Frontiers in Sports and Active Living, 2025. doi:10.3389/fspor.2025.1505301
• Deschodt VJ, et al. Relative contribution of arms and legs in humans to propulsion in 25-m sprint front-crawl swimming. European Journal of Applied Physiology, 1999;80(3):192–199. doi:10.1007/s004210050581
• De Matos CC, Guignard B, De Souza Castro F, Guimard A. Effects of paddles and fins on front crawl kinematics, arm stroke efficiency, coordination, and estimated energy cost. Frontiers in Physiology, 2023. doi:10.3389/fphys.2023.1174090