TL;DR. Filming your own squat is a solved problem. Running a floor of ten athletes across six racks is a different one: you are never standing where the bad rep happens, and you cannot operate a camera and coach at the same time. The answer is to stop being the person who presses record. Park a phone per rack or run one tripod on a rotation, set the buffer to the length of one set, and let the athlete tap the save after racking the bar. The buffer holds the seconds before the tap, so the set is already in the clip. Review lands in the rest interval, where it changes the next set instead of next week.
Coaching a floor is not filming a lifter
A solo lifter doing a form check has one camera, one bar, and full control of the timing. Everything about that setup breaks once you are the coach for a group. Twelve athletes work through a session in pairs or threes, five or six bars are moving at any moment, and your attention is a scarce resource you are already spending on cueing, spotting, and load changes. The rep that needed video is almost always the one you were turned away from.
Start-stop filming makes this worse rather than better. Every clip costs a walk to the phone, a press, a walk back, and a second walk to stop it, and the athlete waits through all of it. Do that four times in a session and you have spent a meaningful slice of the training block operating a camera. Coaches respond the way anyone would: they film the first two sets of the session, then stop, which is exactly backwards, because form degrades under accumulated fatigue and the sets worth watching are the last ones.
The second failure is where the review happens. Footage collected during a session and watched that evening arrives after the athlete has lost the felt sense of the rep. Work on augmented feedback in motor learning consistently finds that the value of a visual correction is tied to how close it sits to the attempt, because the athlete is matching what they saw against what they felt. A clip watched in the rest interval is coaching. The same clip watched at 9pm is documentation.
A rolling buffer removes both problems by inverting the recording model. The camera is always holding the last N seconds, so the decision to keep a rep happens after the rep, by whoever is standing closest to it, and the review happens where the natural pause already exists: between sets.
Three ways to cover a weight room
Pick one and commit to it for a training block. Mixing patterns week to week produces footage nobody can compare.
One mount per rack, for the lift of the block
Old phones are cheap and every athletics program has a drawer of them. A phone clamped to each rack upright, side-on at hip height, gives every station its own independent buffer. There is no pairing between devices and no shared session to join, so a fourth camera is exactly as much work to add as the first: install, start the buffer, walk away. This is the pattern for a block where the whole squad squats or pulls on the same day and you want every athlete's heavy triple on file.
The cost is collection. Each phone ends the session with its own clips, and someone has to walk round and export the keepers. Budget five minutes at the end of the session for that, or leave the clips where they are and review at the rack.
One rotating tripod, one station per block
A single tripod that spends one exercise block at each station, then moves. Over a four-block session every group gets filmed once, and over four weeks every athlete has a series on the same lift shot from the same angle. This is the cheapest pattern and the one that produces the most comparable footage, because there is one camera, one height, and one distance to keep consistent.
Set the tripod before the block starts, not during it. The camera placement planner gives the distance for the frame you want; in a weight room, 2 to 3 metres back at hip height covers a squat or a pull from the side, and 3 to 4 metres is needed if the athlete travels, as in a jerk or a loaded carry.
One fixed platform camera, always on
A permanent mount on the competition platform or the main lifting area, buffering for the whole session, saving only when someone triggers it. Whoever lifts there gets filmed. This suits weightlifting and any program where the technical work concentrates on one or two platforms while the rest of the floor does accessory work that nobody needs on video.
Storage stays flat because the buffer overwrites itself. A three-hour session on a fixed camera costs the same disk space as ten minutes; only saved clips accumulate. The storage and buffer calculator gives the numbers for your resolution and free space.
Who triggers the save
This is the decision that makes or breaks the setup, and most coaches get it wrong by keeping the trigger for themselves. If you hold the only trigger, video capture is bottlenecked on your attention, which is the resource that is already fully spent. Delegate it.
Athlete taps the phone
Simplest and free. The lifter racks the bar, reaches over, taps once. The buffer already holds the set, so the tap is a bookmark rather than a recording. Works on a phone clamped to the rack within arm's reach. Teach it once at the start of the block and it becomes part of racking the bar.
Coach taps a Wear OS watch
Free tier, and the right choice when you keep the trigger. A Tile or Complication saves in one tap from the wrist over Bluetooth, so you can be spotting on the next platform and still claim the rep you just watched go wrong. One watch pairs to one phone, so this fits the rotating tripod and the fixed platform camera.
Open palm at the lens (Pro)
Hold an open palm toward the camera and the clip saves with nothing touched. This is the one for chalked hands, straps, and a phone mounted out of reach on a rack upright. It is also the answer when hygiene rules discourage a shared touchscreen on a busy floor.
One practical rule: tell athletes to save the rep that felt wrong, not the one that looked good. A set they were happy with rarely needs a coach's eye. The grinder, the one where the hips shot up, the one where the bar drifted forward, those are the clips worth the thirty seconds of review.
Buffer length by what you are reviewing
Set the buffer to the natural unit of review, not to the maximum available. A longer buffer means more footage to scrub before the next set starts, and in a weight room the rest interval is the whole budget.
What you want on the clip
Real duration
Buffer to set
Heavy single, walkout to rack
12–15 s
45 s (free)
Set of five, squat or deadlift
30–45 s
45 s (free)
Set plus the unrack and the setup ritual
45–80 s
90 s (free)
Olympic complex, snatch or clean sequence
1–2 min
3 min (Pro)
Full EMOM block or a superset pair
4–8 min
6 min (Pro)
Whole group's rotation through one station
8–12 min
12 min (Pro)
Frame rate matters more than resolution for lifting. At 60 fps a clip played at 0.25x still shows 15 distinct frames per second, enough to read where a hip broke position. At 30 fps the same playback gives 7.5 frames and the answer is a guess. The frame rate and slow-motion calculator works out the slowest speed your capture setting supports cleanly. Gym lighting is the usual constraint on shooting at 60 or 120 fps, so check what your rooms allow before promising a squad frame-accurate feedback.
Solo form check vs coaching a floor
The app is the same. Almost every decision around it changes.
Decision
Solo lifter, one form check
Coach, whole floor
Cameras
One phone, one tripod
One per rack, or one tripod on a station rotation
Who saves the clip
The lifter, always
Delegated to the athlete, coach keeps a watch trigger
Buffer length
Sized to one set
Sized to the station's unit of work, 45 s to 12 min
When review happens
Immediately, at the rack
In the rest interval, one cue, no crowd around the phone
Bystanders in frame
Commercial gym members
Other athletes, often minors, so consent rules apply
Cost model
One unlock, or the free tier
One unlock per device that needs Pro, no per-athlete seat
A session rhythm that survives contact with a real floor
Decide the one thing you are filming this week. Bar path on the squat, or hip timing on the pull, not both. A block with one question produces a comparable series; a block with six questions produces a folder nobody opens.
Mount and start the buffers before the warm-up ends. Cameras that get set up mid-session get set up badly. Stealth mode blacks the screen while buffering, which saves battery and stops the phone becoming a mirror for whoever walks past.
Brief the trigger in ten seconds. "Rack the bar, tap the phone, only for the rep that felt off." Repeat it once at the second station and it sticks for the block.
Review in the rest interval, one cue. Watch at 0.25x, name a single correction, lift again. A second cue in the same rest interval cancels the first.
Bin the rest before you leave. Swipe right to keep, left to toss. The bin holds deleted clips for 24 hours, so an over-eager cull is recoverable. Ten minutes of triage beats a 300-clip archive nobody will ever open.
Export only what goes into a review. Saved clips are ordinary MP4 files, so they drop into whatever your program already uses for athlete reports or a shared drive. There is no upload step to configure and no proprietary format to escape.
Consent, minors, and a room full of other people's athletes
A weight room camera catches more than the lifter it is pointed at. In a college or high school setting most of those people are minors, and in a commercial or shared facility many of them never agreed to be filmed at all. On-device capture does not make the question go away, but it removes the sharpest edge of it: with ReplayR there is no account, no upload, and no vendor server, so footage of an athlete exists on one phone and nowhere else. The rolling buffer means anyone who walked behind the rack is overwritten within the buffer length unless someone deliberately saves that moment.
What still needs a written answer: who is permitted to film in your facility, whether parental consent is on file for athletes under 18, how the phones are secured when they are not in use, and what may be shared outside the program. Most governing bodies and school districts have a policy that already covers this. Our guide to filming ethics in sports facilities goes through the common requirements. If you share a floor with general members, put a sign at the entrance to the area you are filming and point the camera so the background is a wall rather than the rest of the room.
What this does not do
There is no athlete roster, no shared team library, and no tagging. Clips live on the phone that captured them until someone exports them, which is the direct consequence of having no cloud and no accounts. A program that needs a searchable, squad-wide video archive with permissions should buy a platform built for that; this is a capture tool.
There is no velocity measurement and no rep counting, so it does not replace a bar sensor or a VBT system. Load, sets and progression belong in your programming tool. If your athletes track their own gym work, Personal Trainer handles that side offline; the video answers a different question, which is whether the rep moved the way it was meant to.
Analysis stops at the frame. Pro adds frame-by-frame stepping down to 0.1x and drawing on an exported frame, which covers a vertical line down the bar or a horizontal across the shoulders. It does not do telestration on a moving clip, side-by-side comparison of two lifters, voice-over, or joint angle measurement. And there is no iOS build yet, so a staff split across platforms can only equip the Android side today.
FAQ
How do I film several lifters at once with one coach?
You do not film all of them, you film one rack at a time and move the camera on a schedule. The workable patterns are a rotating tripod that spends one exercise block at each station, a fixed camera on the main platform that catches whoever is lifting there, or one old phone per rack for the two or three lifts you care about that week. Each phone runs its own independent rolling buffer with no pairing between devices, so adding a second or third camera needs no coordination at all.
Who should trigger the save, the coach or the athlete?
The athlete, in most weight rooms. A coach walking a floor of eight lifters cannot be at the right rack at the right second, and the lift that needs review is often the one they did not see. Because the buffer keeps the seconds before the trigger, the athlete racks the bar and taps: the set is already captured. On the free tier the trigger is a phone tap or a Wear OS watch tap; Pro adds an open-palm gesture toward the lens, which is what chalked or gloved hands need.
What buffer length works in a weight room?
The free 45 second buffer covers a set of five on a squat or a deadlift, including the walkout, which is what most sessions need. Set 90 seconds if you want the set plus the setup and the rack. Olympic complexes, EMOM blocks and a full rotation through a station fit the Pro buffers of 3, 6 or 12 minutes. Longer is not automatically better: a longer buffer means more to scrub through before the next set starts.
Do I need a Pro unlock on every phone in the gym?
Only on the phones that need what Pro adds. Pro is a one-time in-app purchase per store account rather than a monthly seat, so there is no per-coach or per-athlete billing. The free tier covers the rolling buffer, the watch remote, review, slow motion and trimming, which is enough for a spare phone parked on a rack. Buy Pro for the phone you actually coach from, where the frame stepping, 0.1x playback and frame annotation get used.
Is it legal to record athletes in a college or high school weight room?
It depends on your institution's policy and, for minors, on consent from a parent or guardian. Technology does not settle that question, but on-device capture narrows it: with ReplayR the footage sits on the phone that filmed it, with no account, no upload and no vendor server, and the rolling buffer overwrites anyone who walked past within seconds unless someone saves. You still need written consent where your governing body requires it, a rule about who may film, and a rule about where clips may be shared.
Does ReplayR track sets, loads or velocity?
No. It captures and reviews video, and nothing else. There is no bar speed measurement, no rep counting, no athlete profile and no training log. Load and set data belong in whatever you already use to program, such as Personal Trainer for individual gym tracking, and the video answers a separate question: did that rep move the way it was supposed to.
Google Play
