Video summary
Samsung Galaxy Fit3 : Scientific Review (BEST deal?!)
Main summary
Key takeaways
Product reviewed
Samsung Galaxy Fit 3 (fitness smartwatch/health band). The review focuses on heart rate tracking, sleep stage tracking, and comparisons versus Galaxy Watch models and other wearable brands.
Key features (as discussed)
- Health & sports tracking
- Heart rate
- Sleep stages
- Activity including indoor cycling, outdoor running, outdoor cycling, and weightlifting
- No built-in GPS
- Uses phone GPS for GPS tracking, so performance depends on your phone’s GPS
- Battery life
- Rated 13 days
- Compared to Galaxy Watch 6 at roughly ~40 hours
- Samsung ecosystem fit
- Works within the Samsung ecosystem, but the reviewer implies it doesn’t feel as premium as Samsung’s higher-end watches
Main tests & findings
1) Heart rate tracking — indoor cycling (Rob)
Setup: Compared Galaxy Fit 3 heart rate to a Polar H10 ECG chest strap (treated as the accurate reference).
Performance:
- Correlation: 0.95 (good overall, not perfect)
- Most points stayed on/near the agreement line in both low and high HR ranges
- Occasional “weird artifacts” / exaggerated peaks (brief incorrect spikes)
Verdict for indoor cycling: Decently accurate; suitable if indoor cycling is your main goal.
2) Heart rate tracking — outdoor running (Rob)
- Correlation: 0.92 (decent, but better watches exist)
- Tendency: points often sat slightly above the reference line, especially at higher HR
- Issue type: cadence lock
- The watch tracked running cadence rather than true heart rate
- Example described: watch reading leaned toward cadence-derived values (e.g., 142 BPM vs cadence 163 SPM)
Verdict for running: Mixed bag—generally okay, but cadence lock can cause notable errors.
3) Heart rate tracking — outdoor cycling (Rob)
- Correlation: 0.69 (poor)
- Many deviations: points above and below the agreement line
- Sometimes reported about half the actual heart rate (reviewer notes the implied ~50% detection)
Verdict for outdoor cycling: Not reliable across rides; accuracy is inconsistent.
4) Heart rate tracking — weightlifting (Rob)
- Correlation: 0.69 (mediocre)
- Peaks during lifts/sets were often missed, usually detecting too-low HR
- Reviewer notes that for lifting, even generally good watches should reach ~0.9+ correlation to be acceptable; Fit 3 falls short
Verdict for weightlifting: “More or less useless” for accurate HR peaks. Best option: ECG chest strap—though the reviewer couldn’t connect the Polar H10 to the Fit 3 (unclear if possible; asks viewers for input).
Additional tester (“Rafael”)
A smaller set of trials by Rafael (friend/colleague) to check whether results generalize.
Rafael — spinning / running / cycling / weightlifting
- Spinning: looks okay (peaks mostly match; some tendency slightly higher HR)
- Running: correlation 0.93 (mostly good)
- No clear cadence lock signs
- Minor artifacts; one run showed deviations but not clearly cadence-driven
- Outdoor cycling: correlation 0.56 (worse overall)
- Some rides looked good, but at least one showed the “half heart rate” behavior
- Mentions big gaps, likely when the watch isn’t recording while not cycling (autopause-like behavior)
- Weightlifting: again struggled (low correlation; missed peaks)
Overall heart-rate scoring (combining both testers)
- Reviewer’s final heart rate assessment: ~3 out of 5 stars
- Okay for indoor cycling and running
- Major problems for cycling outside
- Poor/unusable for weightlifting
Sleep stage tracking
Method
- Tested across 13 nights
- Compared Fit 3 sleep stages vs zMAX EEG headband (brain-wave reference, with limitations around wake detection)
Results (Rob)
- Deep sleep agreement: about 20% (poor)
- Much of deep being predicted as light sleep (~55%)
- Light sleep agreement: about 75% (better)
- REM sleep agreement: about 30% (poor)
- Wake time:
- Reviewer mostly avoids heavily judging it
- Notes Fit 3 detected ~40% of awake time as awake
Why agreement was poor
- On some nights, Fit 3 marked the user awake when EEG indicated sleep
- Also missed sleep onset timing (example: plots suggesting “fell asleep too late”)
- Similar marginal agreement issues for REM segments
Context vs other watches
- Fit 3 compared to many devices using different reference setups (polysomnography vs Dream/EG vs zMAX)
- Fit 3 was placed near the bottom-left (poor overall), not among top performers
- Reviewer notes Galaxy watches generally aren’t top sleep trackers; Fit 3 may be even worse in this setup
Sleep score + verdict
- Reviewer gives 1.5 out of 5 stars for sleep stage tracking
- Overall impression: not great (deep/REM errors and too much “awake” detection)
GPS performance
- No built-in GPS
- Uses phone GPS, so tracking should match the quality of your phone’s GPS
Pricing / value claims and comparisons
- Reviewer positions Fit 3 as better value than Galaxy Watch series
- Fit 3 purchase price: about $65 equivalent
- Galaxy Watch 6 cited as 3–4× more expensive
- Despite value, the reviewer says Fit 3 doesn’t feel premium, and HR + sleep performance is limited versus many competitors
Comparison mentions (brands)
- Galaxy Watch 5 / 6 / 6 Classic / 5 Pro
- Similar-ish performance within Samsung’s line, but not top overall
- Other alternatives mentioned as better (especially for sports/HR accuracy):
- Huawei (e.g., Huawei Band 8, Huawei watch/fit models)
- Fitbit Charge 6
- Apple Watch and some high-end devices
- Described as top sleep stage performers
- Whoop and some selected devices
- Sometimes similar or better correlation depending on activity
Pros (from the video)
- Great battery life: ~13 days
- Indoor cycling heart rate tracking: good (correlation ~0.95) with only occasional artifacts
- Decent for running overall
- ~0.92 correlation for Rob
- ~0.93 correlation for Rafael
- Errors can still happen
Cons (from the video)
- Outdoor cycling heart rate tracking is poor
- Rob correlation 0.69
- Rafael correlation 0.56
- Sometimes shows “half HR” behavior
- Weightlifting heart rate is unreliable
- Misses most HR peaks
- Correlation ~0.69
- Judged “more or less useless”
- Sleep stage tracking is bad
- Deep ~20%
- REM ~30%
- Overall 1.5/5
- No built-in GPS
- Relies on phone GPS
- Possible cadence lock during running for at least one tester
- ECG strap connectivity limitation:
- Reviewer couldn’t connect Polar H10 ECG strap to Fit 3
Overall verdict / recommendation
Recommendation: Buy the Samsung Galaxy Fit 3 only if you mainly care about battery life and indoor cycling / general running, and you can tolerate major accuracy limitations for outdoor cycling, weightlifting, and sleep staging.
If you want reliable heart-rate tracking for hard workouts or accurate sleep stages, the video suggests looking at other brands (often Huawei/Apple/Fitbit-class devices) or using an ECG chest strap (if compatible).
Overall ratings given by the reviewer (stars)
- Heart rate tracking: ~3/5
- Sleep stage tracking: 1.5/5
Unique points mentioned (consolidated)
- Fit 3 is positioned as no-brainer value vs Galaxy Watch pricing (roughly $65 vs Watch 6 3–4× more)
- Samsung reportedly not globally released (US/UK/Canada excluded); availability varies (e.g., via Amazon)
- Indoor cycling HR: correlation 0.95, mostly accurate with brief artifacts/exaggerated peaks
- Outdoor running HR: correlation 0.92, sometimes cadence lock (tracks cadence rather than true HR)
- Outdoor cycling HR: correlation 0.69 (Rob) and 0.56 (Rafael); inconsistent and sometimes half HR detection
- Weightlifting HR: correlation ~0.69; missed peaks; judged not good/useless
- Sleep stages (13 nights vs zMAX EEG)
- Deep sleep agreement ~20%
- Light sleep ~75%
- REM agreement ~30%
- Fit 3 detects lots of awake time while EEG indicates sleep
- Sleep comparison vs other watches: Fit 3 among the worse, not top performers (Apple watches and others do better)
- No built-in GPS; uses phone GPS
- Battery life: 13 days vs Galaxy Watch 6 ~40 hours
- Reviewer suggests preferring an older Galaxy Watch (e.g., Watch 4/5) over Fit 3 in Samsung ecosystem for a more premium experience
Speaker-specific views
-
Rob (primary tester)
- Indoor cycling HR: strong (0.95) with artifacts
- Running: decent (0.92) but clear cadence lock
- Outdoor cycling: poor (0.69), including half-HR issues
- Weightlifting: poor (~0.69), missed peaks
- Sleep stages: bad (deep/REM poor; overall 1.5/5)
-
Rafael (secondary tester)
- Spinning: okay
- Running: mostly good (~0.93), no clear cadence lock
- Outdoor cycling: worse correlation (0.56) with at least one half-HR ride
- Weightlifting: again poor; missed peaks
- Notes about gaps due to autopause-like behavior when not cycling