Garmin Forerunner 165, Running Smartwatch, 43mm, Black

Table 1What it measures, and how

We read Garmin's Forerunner 165 product specification page and the Forerunner 165 Series owner's manual on Sep 30, 2026 [7][8]. Sleep stages, HRV status and blood-oxygen values are estimates computed from the optical signal and motion, and are labelled as such below.

Strip 1Accuracy strip

Each cell is one published result for a Garmin watch that shares the sensor family; results come from small groups of mostly young, healthy adults and may not transfer to you.

• Independent peer-reviewed study
• Manufacturer or regulatory filing data
• No published data found

• Heart rate, graded cycle testMAPE 1.16–1.39%vs. ECG, three target intensities, n = 30; tested Forerunner 945[1]
• Heart rate, treadmill, bike, ellipticalr_c = 0.81vs. ECG, all conditions pooled, 50 adults (two randomly assigned wrist devices each); tested Forerunner 235[2]
• Nightly HRV (RMSSD)CCC 0.87; MAPE 10.52%vs. Polar H10 electrode chest strap as reference, 13 adults, 150 Garmin nights; tested Fenix 6[4]
• Sleep vs. wake / sleep stagesκ = 0.35 / κ = 0.25vs. polysomnography, one lab night, n = 53; tested Forerunner 245 Music[5]
• GPS distance, track vs. cityMAPE 1.3% / 8.9%vs. measuring wheel, 1 participant, 12 trials per area; tested Forerunner 935; study part-funded by Polar[6]

None of these studies tested the Forerunner 165. They tested older or different Garmin models (Forerunner 945, 235, 245 Music and 935; Fenix 6), whose sensor hardware and software differ from this watch. A PubMed search for “Forerunner 165” on Sep 30, 2026 returned no validation study of it.

NotesReading the evidence

The cleanest heart-rate result comes from a hospital group in Taiwan that put a Forerunner 945 on the right wrist of 30 healthy adults during a maximal test on a stationary bike, with ECG as the reference [1]. At three target heart rates the watch's mean absolute percentage error stayed between 1.16% and 1.39%, a mean absolute error of 1.35 to 2.25 beats per minute. That is a good result, but it is a single bike protocol in healthy volunteers aged 20 to 40.

A 2017 comparison run at a US heart center shows the other side [2]. Across treadmill, bike and elliptical sessions, a Forerunner 235 agreed with ECG at a concordance of 0.81, below the chest strap tested alongside it (0.996). Agreement held up on the treadmill and the bike and fell apart on the elliptical trainer with arm levers, where none of the wrist devices reached 0.80. A 2025 study of a Forerunner 45 against a Polar H10 chest strap found no significant effect of self-reported skin tone on heart-rate error; the gap it did find appeared during the first ramp-up in intensity, when the strap read higher than the watch [3].

Night-time numbers are weaker. In a 2025 study funded by the US Air Force Research Laboratory, a Fenix 6 reported nightly RMSSD with a concordance of 0.87 and an average error of 10.52% against beat-to-beat intervals from a Polar H10 electrode chest strap; the authors rated this poor agreement [4]. In a sleep-lab study of 53 adults, a Forerunner 245 Music scored 98% of sleep epochs correctly but only 27% of wake epochs, so it overestimated total sleep time by 43.8 minutes on average. Its stage agreement with polysomnography was 50% (κ = 0.25), and its heart rate just before and during sleep read 5.0 bpm high on average [5]. That lab receives research support from WHOOP, a competing brand; the paper says WHOOP was not involved in the study.

For distance, a Swiss sport institute walked, ran and cycled measured courses wearing eight watches, with a trundle wheel as the reference [6]. The Forerunner 935 averaged 1.3% error on a track and 8.9% in a city center with narrow streets and partly high buildings, where it read short by about 189 m on routes averaging about 2 km. All measurements came from one person, and Polar, a competing brand, part-funded the work.

ProcedureGetting cleaner readings

These are setup steps from Garmin's owner's manual [8]. They reduce avoidable errors; they do not guarantee accuracy.

1. Wear it above the wrist bone

   The manual says the watch should sit above the wrist bone, snug but comfortable, and should not shift while you run.

2. Start with clean, dry skin

   Clean and dry the arm first, and keep sunscreen, lotion and insect repellent out from under the sensor.

3. Wait for a steady signal, then warm up

   Wait until the heart-rate icon is solid before starting an activity, and warm up for 5 to 10 minutes first; in the cold, warm up indoors.

4. Hold still for a Pulse Ox reading

   For a manual reading, sit or stay inactive, keep the arm at heart level and stay motionless until the reading finishes.

5. Pair a chest strap when it matters

   The watch accepts an ANT+ heart-rate strap (Bluetooth sensors only on the Music edition) and switches to it automatically, which is the practical fix for intervals.

LogPrice and where it is sold

On Sep 30, 2026 the Amazon listing “Garmin Forerunner 165, Running Smartwatch, 43mm, Black” (color Black, style Standard, not the Music edition or a bundle) showed $199.99 against a list price of $249.99, shipped and sold by Amazon.com, in stock, with a free 30-day refund or replacement. A renewed unit from a third-party seller was also offered at a lower price; we did not price it. See the listing on Amazon. Prices change; read this as a dated log entry.

Who should look elsewhere

• Anyone who needs a heart-rate or oxygen reading for a medical decision should use a device and a process their clinician chooses.
• Runners doing short intervals who want beat-by-beat accuracy: an electrode chest strap such as the Polar H10 is the usual reference in these studies.
• People who want music stored on the watch or Bluetooth sensor pairing: that is the separate Forerunner 165 Music edition, not the model priced here.
• Anyone who mainly runs among tall buildings or dense trees and wants the tightest distance: the manual's satellite settings for this model include no multi-band option.

Sources

1. Cathay General Hospital, Taipei, and collaborators. Apple Watch Series 6 and Garmin Forerunner 945 heart rate compared with ECG during a maximal cycle-ergometer test in 30 healthy adults. Digital Health, 2022. PubMed 36081752
2. Cleveland Clinic Heart and Vascular Institute. Five optical heart-rate monitors, including the Garmin Forerunner 235, compared with ECG on treadmill, bike and elliptical in 50 adults. Medicine & Science in Sports & Exercise, 2017. PubMed 28709155
3. Wake Forest University. Garmin Forerunner 45 heart rate compared with a Polar H10 chest strap across self-reported skin tones during walking and jogging. Frontiers in Digital Health, 2025. PubMed 40212900
4. The Ohio State University and the Air Force Research Laboratory. Nightly resting heart rate and HRV from five wearables, including the Garmin Fenix 6, compared with a Polar H10 chest-strap reference. Physiological Reports, 2025. PubMed 40834291 · full text
5. Central Queensland University, The Appleton Institute. Six wearables, including the Garmin Forerunner 245 Music, compared with polysomnography and ECG in 53 adults. Sensors, 2022. PubMed 36016077 · full text
6. Swiss Federal Institute of Sport Magglingen. Distance recorded by eight GNSS sport watches, including the Garmin Forerunner 935 and Fenix 5X Plus, compared with a trundle wheel in urban, forest and track settings. JMIR mHealth and uHealth, 2020. PubMed 32396865 · full text (Tables 2 and 4)
7. Garmin. Forerunner 165 product page and specifications. Accessed Sep 30, 2026. link
8. Garmin. Forerunner 165 Series Watch Owner's Manual (PDF). Accessed Sep 30, 2026. link
9. U.S. Food and Drug Administration. openFDA 510(k) dataset, search by applicant “Garmin”, Sep 30, 2026: no clearance listed for the Forerunner 165. link

This entry is educational and is not medical advice. Consumer wellness gadgets are not diagnostic devices; readings can be wrong. If a reading worries you, or you have symptoms, talk to a qualified clinician rather than relying on a gadget.