Can a Smartwatch Count Your Reps? Apple Watch vs Garmin vs Fitbit vs Whoop (2026)
Can a smartwatch count your reps? Apple Watch needs an app, Garmin/Samsung do it natively (but shaky), Whoop & Oura don't. The honest 2026 comparison.
Riven · ProductSome smartwatches can count your reps automatically, but most can't — and the ones that try are inconsistent. In 2026, only Garmin and Samsung do motion-based auto rep counting natively, the Apple Watch needs a third-party app, and Whoop and Oura don't count reps at all. So can a smartwatch count reps? Yes, with big asterisks. Here's the honest device-by-device breakdown.
I've trained for fifteen years and tested most of these on a real gym floor — not a lab. Below is the verdict table first, then the why behind each one.
The verdict table: which watches actually count reps
| Device | Auto rep counting? | How it works | The catch |
|---|---|---|---|
| Apple Watch | No (not native) | Needs a third-party app (Riven, Motra, Gymatic) | Built-in Workout app logs HR/calories/time only |
| Garmin | Yes (native) | Wrist IMU detects arm return-to-start | Miscounts legs, machines, push-ups; you edit manually |
| Samsung Galaxy Watch | Yes (native) | Motion algorithm on free weights | No machines, leg press/extension/curl, push-ups, jump rope |
| Fitbit / Pixel Watch | Effectively no | Logs generic "Weights" workout | Rep/set entry is manual, not sensed |
| Whoop | No | Estimates muscular load/strain | Screenless; you set up the exercise, it doesn't count |
| Oura Ring | No | Auto-detects a strength session | Can't ID lifts or count reps from a finger |
That table is the whole answer if you're skimming. Now let's earn the detail.
Can a smartwatch count reps?
Yes — but "count reps" means two very different things. There's auto-detected reps, where the watch senses each rep from your wrist motion, and logged reps, where you type the number in yourself. Most fitness tracking dressed up as rep counting is the second kind. Only Garmin, Samsung, and a handful of Apple Watch apps (including Riven) attempt true motion-based detection from the watch's accelerometer and gyroscope.
Every one of these systems infers reps from a 6-axis IMU — a 3-axis accelerometer plus a 3-axis gyroscope on your wrist. That sensor reality is why accuracy is so uneven, and it's the thread running through this entire comparison.
Can the Apple Watch count reps?
No. The Apple Watch has never counted reps natively, and that's still true in 2026. If you pick "Traditional Strength Training" or "Functional Strength Training" in the built-in Workout app, it logs heart rate, duration, and active calories — not reps, not sets, not which exercise you did (MyHealthyApple).
To get auto rep counting on an Apple Watch, you install a third-party app. Riven, Motra (formerly Train Fitness), Gymatic, Rep Up, and Fitnexx all read the Watch's motion sensors to count reps for you (Rep Up). The difference between them is what they do beyond counting — more on that below.
Here's the practical upshot: if you already own an Apple Watch, the cheapest path to auto rep counting is an app, not a new device. Expect to confirm or correct a set occasionally. That's the honest state of the art.
How Riven counts reps on Apple Watch
Riven reads the Watch's motion signal to count your reps. Every rep produces a rhythmic motion as your wrist travels — the lifting push or pull — and Riven counts each one, handling both a slow squat and a fast curl. You tap once to start the set, then just lift. No tapping a screen between reps, no tallying in your head.
In our own testing, Riven's count lands within about one rep on the large majority of sets. When it's off, you fix it with a twist of the Digital Crown on the set-review screen. It also names the exercise (push/pull/legs, muscle group, best guess at the specific lift like "incline dumbbell press") and judges whether you actually hit failure via rep-speed loss. Count + identify + effort, one tap. I'll be straight: machine holds and low-wrist-movement isolation moves are the hard cases and may need manual entry. Nothing on the market auto-counts those cleanly.
Does Garmin count reps?
Yes — Garmin is the one major brand with native automatic rep counting, built into the "Strength" activity on watches like the Forerunner 165, vivoactive 5, Venu, and fenix. It uses the wrist IMU to detect when your arm returns to its start position (Garmin Support).
Native does not mean accurate. Garmin's counter works best on consistent-arm-motion lifts — bicep curls, overhead press — and frequently miscounts or fails on push-ups, leg press, leg extension, leg curl, low-range moves, and machine work. Garmin's own support docs tell you to manually edit reps after a set (Garmin Support). Plenty of lifters I know just turn auto-count off and set a target instead.
Can Fitbit, Samsung, or Whoop count reps?
Mixed bag. Samsung Galaxy Watch auto-counts sets and reps for dumbbell and barbell exercises — the closest native competitor to Garmin — but explicitly can't count machines, leg extensions, leg presses, leg curls, push-ups, or jump rope (MyHealthyApple). Fitbit and Pixel Watch don't do reliable motion-based counting; strength gets logged as a generic "Weights" workout and rep/set entry is manual. Whoop doesn't count reps at all — its Strength Trainer estimates muscular load from accelerometer and gyroscope data while you set up the exercises yourself, and the band is screenless anyway (Whoop).
And Oura? A ring can't count reps or identify lifts. With a membership it can auto-detect that a strength session happened (roughly 10+ minutes of movement) and log duration and HR, but a finger-worn sensor struggles with isometrics, slow eccentrics, and low-hand-movement sets (Vertu). Whoop and Oura are recovery and strain devices. That's a different job.
Which smartwatch is best for counting reps? (comparison table)
For pure native rep counting, Garmin and Samsung lead — Garmin across more exercise types, Samsung cleaner on free weights but useless on legs and machines. For Apple Watch owners, a dedicated app beats buying new hardware.
| Device / app | Counts reps | IDs exercise | Judges effort | Best for |
|---|---|---|---|---|
| Apple Watch (native) | No | No | No | Heart rate + calories only |
| Garmin (native) | Yes (shaky on legs) | No | No | Garmin owners doing curls/press |
| Samsung (native) | Yes (free weights only) | Partial | No | Samsung owners, dumbbell work |
| Whoop / Oura | No | No | Strain/recovery | Sleep + readiness |
| Motra / Gymatic (app) | Yes | Yes | No | Apple Watch, auto-logging |
| Riven (app) | Yes (~83% exact, ~99% ±1) | Yes | Yes (failure via velocity) | Apple Watch, reps + ID + effort |
Why the wrist is genuinely hard
This isn't a software bug you can patch away — it's signal physics. A wrist IMU has to infer a rep from acceleration and rotation that vary wildly by exercise, grip, and wrist angle. A systematic review of IMUs for barbell velocity found them generally less valid and less reliable than linear position transducers like GymAware and motion capture, with accuracy heavily dependent on the movement and hardware (PMC review). Your wrist rotates and translates differently from the bar, so raw wrist velocity isn't bar velocity. That's why machines, cables, and leg work are everyone's blind spot.
Counting reps is the floor — effort is the ceiling
Here's the thing the spec sheets miss: the rep tally isn't even the most useful number. What drives muscle growth is proximity to failure, and velocity loss within a set is a validated proxy for it. A 2023 Sports Medicine - Open study (Refalo et al., bench press at 75% 1RM) found velocity loss scaled cleanly with effort — about -22% first-to-last set when training to failure, versus -9% at 1 rep-in-reserve and -6% at 3-RIR (Refalo 2023). VBT coaches use ~40% velocity loss as the lower-body failure threshold and ~50% for upper body (VBT Coach).
A device that counts reps tells you how many. A device that reads velocity decline tells you how hard — which set was a grinder and which was a warm-up. That's the gap Riven aims at: it counts the reps, names the exercise, and estimates how close to failure you got, all from the watch you already wear.
FAQ
Can a smartwatch count reps automatically?
Some can. Garmin and Samsung count reps natively from wrist motion; the Apple Watch needs a third-party app like Riven, Motra, or Gymatic. Whoop and Oura don't count reps at all — they track strain and recovery.
Can the Apple Watch count reps without an app?
No. The built-in Workout app's strength modes only log heart rate, duration, and calories. Automatic rep counting on Apple Watch always requires a third-party app.
Is Garmin's rep counting accurate?
It's inconsistent. It works reasonably on curls and presses but miscounts leg exercises, push-ups, low-range moves, and machine work — Garmin's own docs tell you to edit reps manually after a set.
Do Whoop or Oura count reps?
No. Whoop estimates muscular load from motion while you set up the exercises; Oura can detect that a strength session happened but can't identify lifts or count reps from a finger.
What's the most accurate way to count reps from a wrist?
A dedicated, exercise-aware app beats a generic motion threshold. In internal testing, Riven hit ~83% exact and ~99% within ±1 rep, with one-twist Digital Crown correction for the misses. Machines and low-motion isolation moves remain the hard cases for every system.
---
Want reps counted, your exercise named, and your effort judged from the Apple Watch you already own? Try Riven.
Sources
- MyHealthyApple — Strength Training Using Apple Watch
- Garmin Support — How Rep Counting Works and Editing Reps Manually
- MyHealthyApple — Weightlifting with the Samsung Galaxy Watch
- Whoop — Introducing Strength Trainer
- Vertu — Testing the Oura Ring's Workout Tracking Accuracy
- Validity and Reliability of IMUs for Barbell Velocity (PMC systematic review)
- Refalo et al. 2023, Sports Medicine - Open — Proximity-to-Failure and Neuromuscular Fatigue
- VBT Coach — Velocity Loss Thresholds for Fatigue