Techradar | The Technology Experts

Disclosure: This post contains affiliate links. If you click through and make a purchase, we may earn a small commission at no extra cost to you. Thank you for supporting this site!

I’ve burned through six chest straps in three years chasing accurate heart rate data from wrist-based optical sensors. The truth is, no wrist watch matches a Polar H10 for beat-by-beat precision during high-intensity intervals—but some come damn close. After logging 200+ miles with the Garmin Forerunner 265, Coros Pace 3, and Apple Watch Ultra 2, I can tell you which ones earn their spot on your wrist and which ones still lag behind. If you’re a runner who obsesses over lactate threshold or a lifter who just wants step count, this comparison will save you money and frustration. Let’s break down the real-world performance, firmware quirks, and where each device falls short.

Heart Rate Accuracy: Where the Optical Sensors Actually Fail

Optical heart rate (OHR) sensors use photoplethysmography (PPG) to measure blood volume changes, but they struggle with motion artifacts during running. In my tests, the Garmin Forerunner 265 (firmware 12.28) averaged ±4 BPM off the Polar H10 during steady-state runs at 8:00/mile pace. That’s solid, but during 400m repeats at 5:45/mile, the gap widened to ±8 BPM due to cadence locking—the watch mistaking footstrike for heartbeat. Coros Pace 3 (firmware 3.07) improved this with a new algorithm in late 2023, narrowing the interval drift to ±5 BPM. Apple Watch Ultra 2 (watchOS 10.2) was the worst here, showing ±10 BPM spikes during sprints, likely because its sensor array prioritizes resting accuracy over dynamic response.

For weightlifting, all three watches fail. During heavy squats or deadlifts, the OHR sensors lose contact due to wrist flexion, spitting out erratic readings like 180 BPM when you’re actually at 120. The Garmin and Coros both displayed false spikes above 170 BPM during a 5×5 deadlift set at 315 lbs. If you’re a lifter, skip wrist-based HR entirely—get a chest strap. The only watch that handles this marginally better is the Apple Watch Ultra 2, which uses a third green LED for improved contact, but it still lagged by 15+ seconds during a 30-second set of burpees. For runners, the Garmin and Coros are acceptable for steady-state work, but don’t trust them for interval pacing without a strap.

GPS Drift: The Real-World Numbers That Matter

GPS accuracy is where these watches separate from the pack. I ran the same 10K loop on a marked trail (verified with a measuring wheel) three times with each watch. The Garmin Forerunner 265 (with multi-band GNSS enabled) recorded 10.02 km, 10.01 km, and 10.03 km—drift under 30 meters total. That’s exceptional, thanks to its L1+L5 band support. Coros Pace 3 (also multi-band) showed 10.05 km, 10.07 km, and 10.04 km, with drift around 50 meters. The Apple Watch Ultra 2 (L1+L5) was close: 10.03 km, 10.06 km, and 10.02 km, but it lost signal for 12 seconds under dense tree cover on a wooded section, adding 80 meters of drift. In open fields, all three are within 1% of true distance—good enough for most runners.

Where GPS falls apart is urban canyons and trail runs with heavy canopy. I tested the Coros Pace 3 in downtown Chicago (firmware 3.07) and it cut corners by 15 meters on a 1-mile loop due to multipath errors. The Garmin handled the same route with only 5 meters of error, thanks to its SatIQ auto-select feature. For trail running, the Apple Watch Ultra 2’s precision start feature (which waits for a strong lock) helps, but it chews battery—dropping from 36 hours to 18 hours in a single 10-mile run. If you’re a trail runner who needs reliable distance, the Garmin Forerunner 265 is the clear winner. For road runners, the Coros Pace 3 offers 95% of the accuracy at $100 less.

Battery Life: The Trade-Off Between Features and Longevity

Battery life is a brutal trade-off. The Garmin Forerunner 265 lasts 13 days in smartwatch mode and 20 hours with GPS+music (per manufacturer specs). In my real-world test with daily runs (1 hour GPS, 30 minutes music streaming), it lasted 8 days before needing a charge. The Coros Pace 3 dominates here: 17 days smartwatch mode and 38 hours full GPS (no music). I got 14 days with daily runs and no music—impressive for a $229 watch. The Apple Watch Ultra 2 manages 36 hours in normal use, but with GPS and cellular, it drops to 18 hours. For a 100-mile ultra, the Coros can handle the distance without charging; the Garmin would need a top-up around 50 miles if using music.

Charging speeds matter too. The Garmin takes 90 minutes to full charge via its proprietary cable, while the Coros uses a USB-C adapter that hits 80% in 45 minutes. Apple’s fast charger gets the Ultra 2 to 80% in 60 minutes, but you need a separate puck. If you’re a multi-day backpacker or ultra runner, the Coros Pace 3 is the only sane choice. For daily training with short runs, the Garmin’s battery is fine, but don’t expect it to last a week if you use always-on display and music. The Apple Watch Ultra 2 is a hybrid—good for a weekend trip, but you’ll be hunting for a charger by Sunday night.

Training Features: For Runners, Not Lifters

These watches are built for runners, and it shows. The Garmin Forerunner 265 offers Training Readiness (based on HRV, sleep, and acute load) and a Daily Suggested Workout that adapts to your recovery. I followed its recommendations for 4 weeks and saw my 5K time drop from 22:15 to 21:40—partly due to smarter rest days. Coros Pace 3 has a similar feature called EvoLab, which estimates VO2 max and race times, but it lacks the depth of Garmin’s metrics. For example, Garmin’s Training Status (Productive, Maintaining, Detraining) is more actionable than Coros’s simple fitness score. Apple Watch Ultra 2’s Training Load is basic—it shows effort but doesn’t prescribe recovery. For runners, Garmin wins by a mile.

For lifters, these watches are nearly useless. None of them track specific exercises, rep counts, or rest timers effectively. The Apple Watch has a native Strength Training mode that guesses at sets, but it’s wildly inaccurate—I did 5 sets of bench press and it logged 3 sets with 8 reps instead of 10. Garmin’s Strength activity counts reps based on arm movement, but it missed every other rep during deadlifts. Coros doesn’t even have a strength mode. If you’re a lifter, buy a $20 notebook or a dedicated app like Strong. These watches are for endurance athletes, period. The only crossover feature is sleep tracking, which all three handle decently—but more on that next.

Sleep Tracking and Recovery: The Nighttime Reality Check

Sleep tracking is where these watches either earn your trust or frustrate you. I wore all three simultaneously for 14 nights, comparing them to a Withings Sleep Analyzer pad (which tracks movement and breathing). The Garmin Forerunner 265 was the most accurate for sleep stages: it matched the pad’s deep sleep detection within 15 minutes on average. Coros Pace 3 overestimated light sleep by 30 minutes per night, often mistaking late-night restlessness for deep sleep. Apple Watch Ultra 2 was the worst for latency—it took 20 minutes longer to detect sleep onset, probably because its algorithm requires consistent stillness. For recovery metrics, Garmin’s Body Battery and HRV status are more reliable than Coros’s Recovery Score, which I found too optimistic after hard workouts.

One specific bug: on Garmin firmware 12.28, the watch occasionally recorded a nap as nighttime sleep if I dozed off before 10 PM. Coros firmware 3.07 fixed a similar issue where it split sleep into two segments if you woke for a bathroom break. Apple’s watchOS 10.2 introduced a sleep schedule that forces you to set fixed times—unhelpful for shift workers. If you’re serious about recovery, the Garmin is the best of this trio, but none replace a dedicated sleep tracker like the Oura Ring Gen 3 for HRV accuracy. For most runners, the Garmin’s sleep data is good enough to guide training decisions—like skipping a tempo run after a poor night.

Price and Value: What You Actually Get for Your Money

Pricing is where the Coros Pace 3 shines and the Apple Watch Ultra 2 stumbles. The Coros Pace 3 retails for $229, and you get multi-band GPS, 38-hour battery life, and a lightweight design (30 grams without strap). The Garmin Forerunner 265 is $449, adding an AMOLED display, music storage, and deeper training metrics—but you pay $220 more for a brighter screen and 13 fewer hours of GPS battery. The Apple Watch Ultra 2 costs $799, offering cellular, a titanium case, and the Apple ecosystem, but its fitness features are shallow compared to Garmin’s. For a runner on a budget, the Coros Pace 3 is the best value: it covers 90% of what a $800 watch does for a third of the price.

But value depends on your priorities. If you want music without a phone, the Garmin Forerunner 265 can store 500 songs from Spotify (offline playlists), while the Coros Pace 3 has no music support. The Apple Watch Ultra 2 streams from your phone or cellular, but battery life tanks. For trail runners who need navigation, the Garmin has preloaded TopoActive maps and turn-by-turn directions; the Coros offers breadcrumb navigation only. The Apple Watch uses third-party apps like WorkOutDoors for maps, adding $10 per year. In short: Coros for budget endurance, Garmin for serious training features, Apple for ecosystem integration. Don’t buy the Apple Watch for running—buy it for iPhone compatibility.

Verdict: Buy This If You’re a Runner, Skip It If You’re a Lifter

After months of testing, here’s the bottom line. The Garmin Forerunner 265 is the best watch for runners who want detailed training insights, accurate GPS, and music on the go. It’s worth the $449 if you’re chasing PRs or training for a marathon. The Coros Pace 3 is the best value for endurance athletes—ultra runners, triathletes, or anyone who needs battery life over flashy features. At $229, it’s a no-brainer for budget-conscious athletes. The Apple Watch Ultra 2 is for iPhone users who want a smartwatch first and a fitness tracker second—its running features are decent but not class-leading. For lifters, skip all three: none track strength training well, and you’re better off with a $50 fitness band for steps and a notebook for sets. My recommendation? If you run more than 20 miles per week, get the Garmin. If you run longer than 10 hours per week, get the Coros. If you just want notifications on your wrist, get the Apple Watch—but don’t expect it to improve your 5K time.

Frequently Asked Questions

Can any of these watches replace a chest strap for heart rate accuracy?

No, not during high-intensity intervals or weightlifting. The Garmin Forerunner 265 and Coros Pace 3 are within ±4-5 BPM of a chest strap during steady-state running, but during sprints or heavy lifts, the error jumps to ±10 BPM or more. For zone-based training, a chest strap like the Polar H10 is still essential. If you want wrist-based accuracy for daily runs, the Garmin is your best bet, but don’t rely on it for threshold testing.

Which watch has the best GPS for trail running under tree cover?

The Garmin Forerunner 265 with multi-band GNSS is the leader, showing less than 30 meters of drift on a 10K wooded trail. The Coros Pace 3 is close with around 50 meters, but it struggles in dense forest. The Apple Watch Ultra 2 loses signal for 10-20 seconds under heavy canopy, adding 80 meters of error. For trail runners, the Garmin is worth the extra $220 for reliability.

How often do these watches need firmware updates for accuracy fixes?

Garmin pushes updates every 4-6 weeks, and they often address HR spikes and GPS glitches—firmware 12.28 fixed cadence locking issues. Coros updates every 2-3 months, with firmware 3.07 improving GPS drift in urban areas. Apple updates with watchOS releases every 2-3 months, but fitness fixes are bundled with broader updates. You should update immediately when available to get the best accuracy.