Heart rate data is one of the most useful training metrics if it is accurate and one of the most misleading if it is not. The gap between an accurate reading and a wrong reading often comes down to which device is on your body. A wrist watch and a chest strap measure the same number using completely different methods, and each method has predictable conditions where it succeeds and fails. Knowing the failure modes matters more than choosing the “better” device, because both have a place in a training setup.
The short version: chest straps are more accurate in almost every condition that matters for training quality, but watches are dramatically more convenient and good enough for many use cases.
How wrist optical sensors actually work
Watches use photoplethysmography, often shortened to PPG. The sensor on the back of the watch shines green LED light into the wrist, and a photodetector measures how much light bounces back. Blood flowing through the capillaries under the skin absorbs more green light than non-blood tissue does, so the absorption pulses with each heartbeat. The watch counts those pulses and reports beats per minute.
This works well when the watch sits firmly against still skin with good blood flow. It fails predictably in several conditions:
Cold skin reduces blood flow to the wrist, which weakens the signal. Cold-weather runs in the first 10 to 15 minutes often produce wildly inaccurate readings until the wrist warms up.
Loose watch fit allows ambient light in and the watch to move slightly relative to the skin, both of which corrupt the signal. The watch needs to be snug, often tighter than people normally wear it, for accurate readings.
Tattoos at the sensor location block the light. Dark inks especially. This is documented by Apple and Garmin in their support docs.
Cadence lock during repetitive arm motion. Bicep curls, kettlebell work, and any exercise that rhythmically contracts the wrist or forearm causes the sensor to lock onto the movement cadence instead of the heart rate.
Interval transitions show lag. The optical signal averages over a few seconds, so jumping from 120 bpm to 170 bpm during a sprint typically takes the watch 15 to 30 seconds to register, by which time the interval is half over.
Modern watches (Apple Watch Series 9 and Ultra, Garmin Fenix 7, Whoop 4) have improved optical algorithms, but the fundamental physics has not changed. The sensor sees blood flow at the wrist and cannot directly see the electrical activity of the heart.
How chest straps actually work
Chest straps use electrocardiography, often shortened to ECG. Electrodes against the chest skin pick up the electrical signal that fires each heartbeat. The signal is sharp and clear, the same kind of signal a hospital ECG uses (with fewer leads and lower medical-grade certification).
This is fundamentally more accurate because it measures the actual electrical event that triggers the heartbeat, not a downstream consequence of it.
Chest straps still fail occasionally:
Dry skin or new strap. Both lead to poor contact. Wetting the electrode pads with water (or sweat) before starting a workout solves this. A new strap often shows erratic readings in the first 5 to 10 minutes until the contact stabilizes.
Loose strap fit. The strap needs to be snug around the rib cage just below the chest muscles. Riding too low or too loose loses contact.
Worn-out electrode pads. After 12 to 24 months of regular use, the electrode pads degrade and accuracy drops. Replacing the pod or the entire strap is the fix.
Static electricity in dry winter conditions. The strap can pick up spurious signals from synthetic clothing rubbing against the chest. Wetting the electrodes more thoroughly helps.
Within those constraints, a working chest strap delivers data within 1 to 3 bpm of medical-grade ECG in almost any workout.
When the accuracy gap actually matters
For a casual moderate run at a steady pace for 30 minutes, a watch and a chest strap usually agree within 5 bpm. The difference is not training-relevant.
For interval training (Tabata, HIIT, 4x4 protocols), the watch lag during transitions can underreport the actual heart rate peak by 10 to 20 bpm. If you are trying to hit a specific intensity for a specific number of seconds, the watch’s lag corrupts the workout structure.
For weight training, the watch’s cadence-lock problem during isolation exercises makes the data essentially useless. Chest straps work fine.
For polarized training (mostly easy work, occasional hard work), watches are usually accurate enough in the easy zones and inaccurate enough in the hard zones to matter. Chest straps are the better tool.
For VO2max estimates and recovery scoring, chest strap data is meaningfully more accurate than wrist data. Algorithms that ingest poor data produce poor outputs.
Specific products by category
Chest straps. The Polar H10 is the long-standing reference. Dual Bluetooth and ANT+, replaceable strap, coin cell battery, accurate. Around $90. The Garmin HRM-Pro Plus adds running dynamics (cadence, vertical oscillation, ground contact) for Garmin watch users and runs around $130. Wahoo TICKR-X is a slightly cheaper option at around $80 with similar Bluetooth/ANT+ compatibility.
Watches with strong optical sensors. Apple Watch Ultra 2 is among the best wrist optical implementations. Garmin Fenix 7 and Forerunner 965 use Elevate Gen 4 sensors with reasonable accuracy. Whoop 4 has dedicated optical sensors and a tighter band fit that improves accuracy somewhat.
Armbands. Polar Verity Sense and Wahoo TICKR Fit are optical sensors worn on the forearm or upper arm. They are noticeably more accurate than wrist optical because the arm tissue is less prone to movement artifact. Verity Sense is a strong pick at around $90 for users who want optical accuracy without a chest strap.
When to use which
Daily life and general health tracking: watch. The convenience of always-on monitoring without putting on extra gear matters more than precise accuracy.
Steady-state cardio at conversational pace: watch is usually fine. Chest strap if you want to be sure.
Intervals, HIIT, threshold work: chest strap. The data quality on intervals is meaningfully different.
Weight training and CrossFit: chest strap or armband. Wrist optical is unreliable here.
Triathlon and structured training programs: chest strap. The investment in training time deserves accurate data.
Casual cycling: depends. If your hands are still on the bars, wrist optical works fine. Chest strap is more accurate but the watch is easier.
Cold-weather running: chest strap. Wrist optical struggles in the first part of cold-weather runs.
How to pair a strap with your watch
Most modern watches accept Bluetooth heart rate broadcasts from third-party straps. The pairing setup is in the watch’s sensors menu. Once paired, the watch uses the strap’s data instead of the wrist sensor. Some watches let you set the strap as the default heart rate source for specific activity types only (run with strap, daily life with wrist), which is the most practical configuration.
Apple Watch users need a Bluetooth strap. ANT+ does not pair with Apple Watch. The Polar H10, Wahoo TICKR, and Coros HRM all pair fine.
Garmin watches accept both Bluetooth and ANT+, but ANT+ is the more reliable connection for Garmin specifically.
The cost picture
A Polar H10 chest strap costs about $90 and lasts 3 to 5 years with periodic battery replacements ($3 every 12 to 18 months). Over five years, the total cost of ownership is around $110.
A high-end watch with reliable optical sensing costs $400 to $900. The optical sensor stops being accurate when the watch is replaced (typically 3 to 5 years).
For a serious training setup, the chest strap is the cheaper investment in data quality. Buying a watch for the convenience and a chest strap for the workouts is a common pattern and is what most experienced athletes end up doing.
For more on testing wearables, see our methodology.
Frequently asked questions
Are chest straps more accurate than watches for heart rate?+
Yes, in almost every condition. Published comparison studies consistently show chest straps within 1 to 3 beats per minute of medical-grade ECG, while wrist optical sensors can drift 5 to 15 beats per minute during intervals, weight training, and cold conditions. The gap closes during steady-state running but never disappears.
Why does my watch show 180 bpm during bicep curls?+
This is cadence lock. The watch's optical sensor sees rhythmic blood-flow disturbances from the contracting forearm muscle and locks onto the cadence of your reps instead of your actual heart rate. It is a known weakness of all wrist optical sensors during weight training.
Do I need a chest strap if I just do moderate cardio?+
For steady-state moderate cardio at a constant pace, a wrist monitor is usually accurate within 5 bpm, which is fine for most training purposes. For intervals, HIIT, or weight training, a chest strap pays off in real accuracy. Most serious athletes own both and use whichever fits the workout.
How long does a chest strap battery last?+
Coin cell batteries (CR2032 typical) in a chest strap last 12 to 18 months of regular use. Rechargeable straps last 200 to 400 hours of use per charge. Coin cell straps are cheaper to replace and last longer between battery changes, but they require occasional battery purchases.
Can I use a chest strap with any watch?+
If the strap supports Bluetooth (most modern straps do), it pairs with any Bluetooth-compatible watch. ANT+ chest straps pair with most Garmin watches but not with Apple or some other brands. Dual-protocol straps like the Polar H10 and Garmin HRM-Pro Plus pair with essentially everything.
Related product reviews
