Sleep tracking has gained immense popularity over the past few years. Biohackers, fitness enthusiasts and healthy living advocates who want to improve their well-being have all realized the importance of a good night’s sleep.
But to improve your sleep, you need a way to measure it objectively. That’s where sleep trackers can help — or so their manufacturers claim.
For this article, I tested and reviewed various wearable and non-wearable sleep trackers to find out which ones are accurate and reliable.
Spoiler alert: Most sleep trackers are gimmicky and utterly unreliable, with their inaccurate data paving the way for confusion and anxiety.
Continue reading to find out how sleep tracking works and what to look for when purchasing a sleep monitor. I’ll also recommend tracking devices that I’ve found to be reliable and worth your money.
Sleep isn’t the only factor that influences our health and wellbeing. Learn more about all five factors you need to pay attention to.
What Does a Sleep Monitor Do?
Sleep tracking can mean different things depending on who you ask and what type of sleep monitor you use.
In its most basic form, sleep tracking means measuring the time you’re asleep vs. the time you’re awake. This is often done by monitoring (wrist or full body) movement. The more reliable sleep tracking devices also keep tabs on your heart rate and other biometrics in an attempt to improve their accuracy.
The next stage of sleep monitoring includes detection and reporting on the four stages of sleep, which are:
- Awake time.
- Light sleep.
- Deep sleep (also known as slow-wave sleep).
- REM sleep.
It’s worth noting that detecting the different sleep phases is inherently more involved than simply differentiating between awake time and sleep time. As a result, you can expect a higher degree of unreliability among most sleep trackers when it comes to phase reporting.
How Can a Sleep Monitor Help You Sleep Better?
Many people use sleep tracking devices because they want to improve their sleep habits or sleep patterns by sleeping longer, falling asleep quicker, getting more restorative sleep (aka sleeping better), waking up less, or a combination of those goals.
Below are six common sleep parameters that you can directly or indirectly influence in order to improve your overall sleep:
- Total time in bed.
- Total sleep.
- Sleep latency (how long it takes you to fall asleep).
- Sleep consistency.
- Time spent in restorative stages of sleep (deep and REM).
- Wake time after sleep onset (WASO).
Before you try making adjustments to any of these six parameters, it’s important to establish a baseline using a reliable sleep tracker. Only then can you judge the extent to which certain lifestyle changes positively or negatively influence your sleep.
For example, I’ve noticed that I get more consistent deep and REM sleep if I go to bed at approximately the same time every day. That’s called sleep consistency. I’ve also noticed that drinking alcohol before bedtime makes me fall asleep quicker, but I don’t get as much deep sleep as when I don’t drink.
You can learn more about that and all the other factors that influence sleep in this article.
Side Effects of Inaccurate Tracking Devices
The problem with inaccurate sleep trackers is that they might induce stress and anxiety with misleading data and sleep scores.
A few years ago, I had a Jawbone UP wristband that I used to track my sleep. After a few weeks of monitoring my data, I got anxious because of the little deep sleep the device reported: according to the UP, I got only between 20 and 40 minutes of deep sleep every night.
After doing some research, I learned that that’s not nearly enough time for the body to recover and go through all its crucial restorative processes (such as releasing growth hormone). If those readings were correct, I would have felt terrible and exhausted during the day.
I didn’t, which led me to surmise that the device was wildly inaccurate. As a result, I stopped wearing the Jawbone and decided to look for alternative solutions, including apps that would work in combination with the Apple Watch.
While it’s difficult to determine the accuracy of a sleep tracker without performing validation studies, you can tell if a device is likely unreliable by understanding what sensors and technology it uses.
If you’re not interested in the details, you can skip the next section and jump right to my recommendations.
Types of Sleep Trackers
At a base level, there are two different types of sleep trackers: those you wear on your body and those that don’t require any contact with your skin. The latter are called non-wearable or contactless devices.
Wearable Sleep Trackers
As the name implies, wearable devices are worn on your body — typically on your wrist, arm or around your chest.
Popular devices in the wearable category include:
Of course, there are tons of other devices that I haven’t listed, but the ones above are among the most popular. Note that I have hands-on experience with all the sleep trackers in the list above, except for the Polar M430 and the Withings Steel.
Non-Wearable Sleep Trackers
Non-wearable or contactless sleep trackers are devices you place under your sheets, under your mattress, or on a nightstand next to your bed. Most of the devices in this category work by detecting movement via pressure sensors or via a radio frequency that monitors your respiratory rate (breathing).
Popular devices in the non-wearable category include:
- Beddit Sleep (acquired by Apple).
- Emfit QS.
- Pod Pro by Eight Sleep (check out my in-depth Eight Sleep review and YouTube video).
- ResMed S+.
- Withings Sleep (Nokia Sleep).
I was pretty excited to see what Apple was going to do around sleep tracking. I was hoping the company would introduce sleep tracking with the Apple Watch Series 5 and watchOS 6, but they didn’t — despite evidence in watchOS’s source code that they’re working on it.
Fast forward to the Apple Watch Series 7 and watchOS 8, and there is still no sign of sleep stage tracking (likely because of battery life limitations). But considering the fact that Apple acquired Beddit, it only makes sense for Apple to combine Beddit’s contactless technology with the wearable sensors of a future Apple Watch.
To be honest, I’m not convinced that most contactless sleep tracking solutions are accurate enough to be useful. There are just too many factors that can negatively influence the signal the sensors receive, including:
- Motion caused by a pet or the person you’re sharing a bed with.
- Difficulty acquiring a good heart rate measurement.
- Without good heart rate data, movement alone isn’t enough to reliably detect sleep.
As a result, I would generally not recommend non-wearable sleep trackers unless they work in conjunction with a wearable sensor (such as the Apple Watch). Based on my experience with some of the solutions in the list above, I’ve found only the Pod Pro by Eight Sleep to be relatively accurate.
How Do Sleep Trackers Work?
Sleep tracking can be accomplished using various technologies, some of which I briefly mentioned above.
In a nutshell, sleep trackers use one or a combination of the following biometrics in an attempt to accurately detect sleep:
- Heart rate (including heart rate variability).
- Skin conductance.
- Skin temperature.
- Respiratory rate.
- Blood oxygen saturation.
- Electroencephalography (EEG).
Let’s be clear: with the exception of an EEG that measures brain waves, none of these parameters measure sleep directly. Instead, all but one of the above biometrics can be used as a method for estimating sleep, as noted by Alan Schwartz, M.D., director of the Sleep Disorders Center at Johns Hopkins Bayview Medical Center.
So let’s explore how each of these technologies can impact the reliability and accuracy of sleep monitoring devices.
Basic sleep tracking devices, such as the Withings Steel, rely on wrist movement patterns to guess if you’re sleeping (and in what stage of sleep you’re in).
A lack of movement can certainly be an indication of sleep, but it’s an incredibly unreliable parameter. If you lie in bed watching TV, you might not move for an extended period, tricking this type of monitor into thinking that you’re asleep.
I experienced such false positives a couple of times with Jawbone UP and various Apple Watch sleep monitoring apps. I even experienced such a false positive with the first-generation Biostrap when I reviewed it a couple of years ago.
Monitor Heart Rate (HR)
An accurate heart rate reading, especially when combined with monitoring movement, can measurably improve the reliability of sleep trackers.
That’s because the heart rate changes throughout the various stages of sleep. For example, during deep sleep your resting heart rate is typically at its lowest.
Unfortunately, getting an accurate heart rate reading is more difficult than you might imagine. That’s a big problem because a discrepancy of 10-20 beats per minute can dramatically skew the sleep data.
Monitor Skin Conductance
Skin conductance, or electrodermal activity, is a term that describes the electrical characteristics of the skin. In other words, the amount of electricity that can pass through the skin varies, based on different factors.
The interesting thing is that skin conductance changes based on the stage of sleep you’re in. That’s why WHOOP has a built-in sensor that measures electrodermal activity to further improve the accuracy of sleep stage detection.
Monitor Ambient Skin Temperature
Your body’s core temperature fluctuates while you sleep. For example, “non-rapid eye movement (NREM) sleep episodes are accompanied by core and brain cooling.”
As a result, sleep monitoring devices can keep tabs on your skin temperature — much like a fever thermometer — to detect changes in your sleep cycle and, in particular, sleep onset. Both the WHOOP Strap 4.0 and the Oura Ring can monitor skin temperature!
Heart Rate Variability (HRV)
Heart Rate Variability or HRV is the difference in timing between heart beats. For example, if your heart beats 60 times per minute, the timing between each beat varies by a few milliseconds.
The amount of variability is determined by the two (competing) players of your nervous system: the sympathetic and the parasympathetic branch.
Studies have shown that when you sleep, your HRV is greater than it is during the day. That makes sense, because when you sleep the parasympathetic nervous system takes over to take care of digestion and recovery.
Sleep and fitness trackers, such as the WHOOP, have incredibly accurate HRV monitoring capabilities, which can help improve the accuracy of sleep tracking and many other things.
Rate of Breathing
A person’s respiratory rate doesn’t usually change much throughout the night. If you see sudden spikes, it’s usually a sign of illness or a respiratory problem.
Certain sleep trackers — including WHOOP, Biostrap and the Oura Ring — keep tabs on your respiratory rate to help you detect early signs of illness. Respiratory rate can also be used as an indicator of when you fell asleep, because your breathing is likely to be slower while sleeping than when you’re awake.
Some contactless tracking technologies rely on radio frequency to monitor your respiratory rate.
I’ve never tested such devices, but I have gathered from other reviewers* that they aren’t incredibly reliable.
Force and Pressure Sensors
Another technique that non-wearable trackers use is force and pressure sensors that detect body movement and even heart rate.
I was highly skeptical of using a sensor under my mattress to detect my heart rate and sleep stages. For example, my experiments with Emfit QS yielded highly inaccurate results. Plus, feedback from others suggests that such sensors often detect sleep when there is a lack of movement caused by inactivity*, such as reading a book.
However, I’ve been using the Pod Pro by Eight Sleep (a temperature-controlled smart mattress) and it’s got built-in pressure sensors to track sleep. Based on the data I’ve collected, Eight Sleep’s sleep tracking is fairly accurate and in line with what WHOOP, Biostrap and Muse S have reported.
I’m still looking forward to what Apple will come up with since they acquired Beddit (one of the devices that uses under-the-sheets sensors). I can imagine that paired with a wrist-worn device — such as an Apple Watch — you can get fairly accurate data, even with low HR sampling from the wrist-worn device.
Arguably the gold standard in sleep tracking is EEG, which monitors brain waves and detects the various stages of sleep.
Studies have shown that EEG-based sleep tracking is 91% accurate and that it can be further improved by being paired with wrist-worn devices that use the technologies noted above.
If you’re suffering from a sleep disorder and are considering going to a sleep lab to do a sleep study, just know that you probably have to spend more than one night there to get accurate data.
Imagine you have to sleep in a medical facility, hooked up to a bunch of sensors. That’s anxiety-inducing and, chances are, your sleep might not look anything like it does in your own bed.
I’ve been told that many sleep labs throw out the results of the first night and ask patients to come back for a second night. The assumption is that you’ll be more comfortable the second time around, resulting in much better and more usable data.
The good news is that there are now EEG devices available that consumers can use from the comfort of their home. One example is Muse S, a brain-sensing headband that I’ve been using for meditation and sleep tracking. Check out my hands-on Muse S review to learn more about the device!
How HR is Measured
Most wrist-worn HR monitors use a technology called photoplethysmography to measure heart rate.
That basically means that they use LEDs to shine light through your skin.
As blood pumps through your blood vessels, some of that light gets reflected. By measuring how much light gets reflected each time, an optical sensor can measure how often the heart beats.
Signal Analysis and Algorithm
Besides the physical challenges of shining light through blood vessels and accurately capturing the reflection, it’s up to complex algorithms in the device to convert the signal into a more or less accurate heart rate reading.
One of the big issues at play is “noise.” For example, if the sensor doesn’t have sufficiently tight contact with the skin, ambient light can interfere with the reflecting light from the LEDs, thus reducing the overall signal quality.
On the other end of the spectrum, a sensor that fits too tightly might restrict blood flow and thus also reduce signal quality. Either way, the resulting heart rate reading might be inaccurate.
Consequently, if the HR reading is inaccurate, the resulting sleep monitoring will be as well. In my experience, most standard straps that look like regular watch straps don’t provide a good enough fit for the sensor to remain in constant contact with the skin.
That’s why I use the Sport Loop *with my Apple Watch and the ProKnit strap with my WHOOP*. Regardless of what brand or type of device you choose, make sure it comes with a good strap that can be easily adjusted for a perfect fit. Trust me, that makes all the difference!
The second influencing factor is the quality of the signal processing algorithm. Unfortunately, most brands don’t share much information about their algorithm and the expertise of the engineers who designed it.
That’s why I’d recommend sticking with brands that either have a track record of developing advanced technology in the field, or that can back up their work with validation studies involving well-regarded data scientists.
Another factor that can influence the reliability and accuracy of any sensor is the frequency of the biometric data acquisition.
Most trackers measure your heart rate only once every few minutes. That’s also true for the Apple Watch, unless the Workout app is active. During workouts, the Apple Watch measures your heart rate every few seconds.
The most accurate sleep trackers I’ve seen, such as the WHOOP*, measure your heart rate (and other biometrics) 100 times per second or more. That’s important, because it allows them to immediately detect small changes in heart rate patterns.
How Accurate Are Sleep Trackers?
Most sleep trackers that I’ve tested or researched are inherently inaccurate for one, or a combination of, the following reasons:
- Reliance on only one or two biometric sources of information (i.e., movement or heart rate).
- Low sampling data.
- Inadequate algorithm,
- Issues with capturing biometric data due to lack of contact with the skin (i.e., strap too lose).
Confirming the accuracy of any sleep tracker is difficult, unless you compare it head-to-head with an EEG (which is what I’ve done for my Muse S review).
Fortunately, the company behind my favorite sleep tracker — WHOOP — has done exactly that. That’s one of the reasons why I trust WHOOP’s data more than that of any other wrist-worn device.
If your sleep tracker of choice doesn’t have any validation studies to show for, here are some factors that can help you determine how accurate (or not) the device might be.
Sensors and Biometric Input
The fewer sources of biometric data your sleep tracker captures, the less accurate it will be.
I would categorically avoid any tracking devices that rely only on an accelerometer to detect movement. That’s because movement alone is not a reliable predictor of sleep.
An example of a device that falls into this category would be the Withings Steel.
Devices that combine an accelerometer with a heart rate sensor are much better.
However, there’s a significant difference between sensors that capture the heart rate once every five minutes vs. 100 times per second.
The higher the sampling frequency, the easier it is to detect changes in heart rate that might indicate a change of sleep stage.
My advice is to use a sleep tracker that leverages as many different input sources as possible.
The signal processing algorithm plays a major role in the quality of the results. That’s true in particular when the sampling conditions are less than perfect, such as when the sensor doesn’t have sufficient contact with the skin due to a loose strap.
It’s also important that the algorithm doesn’t fill gaps in the captured data with “guesses” just so that it can “complete the picture.”
That means if the sensor can’t capture good enough data, it should discard it instead of guessing what the data (such as heart rate) might have been, based on readings before and after the gap.
Companies like WHOOP have entire data science teams that leverage machine learning combined with huge datasets from professional athletes and fitness enthusiasts to fine-tune their algorithms.
That makes a huge difference in the quality of the reported sleep tracking data.
Type of Strap
As I previously mentioned, the type of strap directly impacts how well in-contact the sensor is with the skin.
The better the contact of the sensor with the skin, the better signal it will receive. Poor signal quality means more reliance on the sleep tracker’s software algorithm, and often results in a less reliable heart rate measurement and sleep tracking.
The strap around your wrist should be tight enough to prevent both the sensor from moving and ambient light from reaching the sensor. However, it shouldn’t be so tight that the pressure limits blood flow in your skin.
As a general rule, it should be difficult to get your pinky finger between the strap and your wrist.
Most straps that I’ve tried in the past are difficult to correctly adjust for a perfect fit — especially traditional watch-like straps, which often produce either a much too loose or much too tight fit.
That’s why I like endlessly-adjustable straps like the Apple Sport Loop or the WHOOP’s ProKnit.
Best Wearable Sleep Trackers
Now that you know everything you need to know about sleep tracking and the pros and cons of the technology that makes it possible, let’s run down the best and most accurate wearable devices.
Before we get into the pros and cons of each of the recommended sleep trackers, here’s a comparison table from one of the nights that I used all but one:
|Biostrap||Eight Sleep||Muse S||WHOOP|
As you can tell from the comparison table above, all four sleep trackers reported relatively similar results, with a few exceptions. For example, Muse S and Eight Sleep reported approximately the same amount of REM sleep, while WHOOP didn’t.
Biostrap doesn’t report REM sleep at all, instead lumping REM sleep together with light sleep. That explains why Biostrap reported a much longer light sleep phase than the other devices.
1. Muse S Brain-Sensing Headband
- Uses EEG technology to monitor brain waves.
- 10-hour battery life.
- Detailed sleep stage and sleep intensity reports.
- Validated accuracy based on several studies.
- Can also be used for meditation and biofeedback.
- Requires an active Bluetooth connection.
- Not as comfortable as a wrist-worn device.
Muse S is a brain-sensing headband that I’ve been using to get real-time feedback on my brain activity during meditation sessions. However, it’s also an incredibly accurate sleep tracker because it uses EEG technology in combination with an optical heart rate sensor and a gyroscope to understand what stage of sleep I’m in.
What I like about Muse S is that it’s the only device I’ve ever tested that monitors actual brainwaves to identify sleep stages. That’s the gold standard of sleep tracking and it is, by definition, more accurate than any other method.
I also like that Muse S gives me insights into how intense my deep sleep is based on delta-wave activity. In addition to reporting my total time asleep and the time I spent in the various stages of sleep, the Muse app also records my sleep position, heart rate and stillness. Additionally, I get a score every morning indicating the quality of my sleep.
What I’m not a fan of is the fact that Muse S streams my sleep data via Bluetooth to my phone instead of storing everything locally and then synchronizing it with the phone. I usually enable Airplane mode on my iPhone at night to reduce my exposure to electromagnetic fields but when I use Muse S, I have to keep Bluetooth enabled.
The other downside to Muse S is that it’s not quite as comfortable as a wrist-worn device. Don’t get me wrong: Muse S is as comfortable as a headband can be. But I generally prefer having a device strapped to my wrist rather than my head.
If you want to give Muse S a try, you can use the link below. Make sure you use code MKUMMER to get 10% off your purchase.
To learn more about Muse S and how I use it for both meditation and sleep tracking, check out my in-depth Muse S review.
2. WHOOP Sleep Tracking
- Multiple sensors, paired with an advanced machine learning algorithm
- Developed by data scientists
- Validated with pro athletes
- Slick design and comfortable to wear 24/7
- Actionable data that improves your health and well-being
- Five-day battery life
The WHOOP strap is arguably the most advanced and most reliable wrist-worn sleep tracker on the market.
Funnily enough, sleep stage tracking wasn’t even the primary goal when WHOOP set out to develop the most sophisticated fitness tracker for both consumers and pro athletes.
So how did WHOOP became such a reliable sleep monitor?
The team at WHOOP wanted to get the most accurate heart rate variability (HRV) readings possible to determine the function and state of the wearer’s nervous system. The problem is that HRV is a constantly-changing parameter that can fluctuate widely, especially when you’re awake and receive sensory input from your surroundings.
The best time to take an HRV reading is during deep sleep, because during that phase of sleep the body doesn’t receive and process any external input. So WHOOP had to develop technology that could reliably detect deep sleep to take an HRV reading.
The byproduct of that effort is the advanced sleep tracking capability that’s built into WHOOP.
To accomplish that, the team at WHOOP not only employs data scientists and machine learning, the company also equipped the wrist-worn device with four different sensors that sample data 100 times per second.
Onboard sensors include:
- Optical heart rate sensor, including pulse oximetry
- 3D gyroscope to detect movement and respiratory rate
- Skin conductance to detect changes in electrodermal activity
- Ambient temperature to detect changes in skin temperature
All of those sensors, combined with a machine learning algorithm that constantly learns from data from thousands of elite athletes and fitness enthusiasts, makes WHOOP my favorite sleep tracker.
What makes WHOOP stand out from the competition is the validation studies the company has performed. In other words, there’s scientific evidence that backs up the accuracy and reliability of WHOOP’s data.
Of course, as I mentioned above, a perfect fit is paramount for WHOOP to do its job. If your WHOOP is too loose or too tight, you’ll get bad data.
See here for a full WHOOP review that goes into much more detail on all of its features.
If you’d like to give WHOOP a try, you can use the link below to get $30 off your membership fee.
3. Oura Ring
- Combination of sensors
- High sampling data
- Long battery life
- Comfortable to wear
- Doesn’t track skin conductance
- Lack of validation studies
The Oura Ring is — as the name implies — a ring that’s packed with advanced sensors that measure heart rate, body temperature, heart rate variability and movement 250 times per second.
What’s interesting about this form factor is that if you get the proper ring size, you can practically ensure a perfect fit. The problem is that some people (including me) have thicker knuckles than fingers. That makes it difficult to find a ring size that slides over your knuckle while providing a tight fit around your finger.
Besides that, I think that a ring is an appealing method for wearing a device 24/7. Some people might not like having a tracker strapped on their wrist literally all the time, but most are OK with wearing a ring.
What’s missing from the Oura Ring is validation and studies involving scientists and real-world data from a combination of regular users and elite athletes. That’s where I think WHOOP has a leg up. I’ve found only one study that compared the accuracy of Oura as a sleep tracker and it confirmed what I noticed during my tests with Oura: Sleep tracking isn’t 100% accurate.
The other issue with Oura Ring isn’t related to sleep tracking at all; it’s that you shouldn’t wear the ring when working out because you might scratch or damage it when your hands come in contact with exercise equipment.
You can learn more about Oura in my in-depth review and comparison with WHOOP.
4. Biostrap Evo
- Monitors heart rate, HRV, O2 saturation and respiratory rate
- Tracks your snoring levels
- Some features are backed up by validation studies
- Was fairly accurate in my tests
- Samples data only every two minutes
- Without membership, the sampling rate is even lower
Biostrap is another interesting technology that relies on various biometric parameters to detect sleep, including heart rate, heart rate variability, oxygen saturation, and respiratory rate.
It doesn’t do skin conductance and temperature, and it only samples data every two minutes if you sign up for a monthly membership. Without a membership, the sampling frequency is even lower.
What I like about Biostrap is that some of its features — especially around the heart rate sensor — are backed up by validation studies. However, I didn’t see any specific studies that confirm the reliability of sleep and sleep stage detection.
During my tests, the Biostrap Evo was fairly spot on with WHOOP most of the time but not always.
The other thing that might be an inconvenience for some users is that you have to take the strap off for charging every morning because the battery doesn’t last more than 10 hours if you turn on “Max Biometrics.”
In comparison, the battery of my WHOOP straps lasts for five days and I can charge it by simply slapping on the battery pack (without having to take off the strap).
Unlike most other sleep trackers I’ve tested, Biostrap can also monitor your leg movements (when you purchase the optional shoe pod) and it can tell you if you’re snoring. I knew before getting Biostrap that I had a habit of snoring after consuming alcohol, but it was Biostrap that made me realize that I also tend to snore more often after consuming dairy.
(That’s likely because dairy increases mucus production, which can tighten your nasal passages, thus increasing the likelihood of snoring.)
It’s also worth noting that Biostrap doesn’t report REM sleep, because the company believes that there isn’t a way to accurately do that without the use of an EEG. As a result, Biostrap lumps REM and light sleep together into a single metric.
Wearable Sleep Trackers I Wouldn’t Recommend
The list below features sleep trackers that I wouldn’t recommend, based on the technology they use.
These devices might accurately track your heart rate, but I doubt — given their technological limitations — that they can reliably track your sleep (and your stages of sleep in particular).
If you’re reading this and you represent one of these brands, I’d be more than happy to wear your device for a few weeks and report back here — just give me a buzz.
Fitbit offers a range of smartwatches and fitness straps that the company claims can accurately track sleep.
Fitbit’s flagship model is the Sense, and it has implemented sleep tracking via a combination of an optical heart rate tracker, a MEMS three-axis accelerometer, and a skin conductance sensor.
Considering that the Fitbit Sense has all the necessary sensors to offer relatively accurate sleep tracking capabilities, I was disappointed to discover that it fails to effectively leverage that technology.
In my tests, neither the total time asleep nor the sleep stages data was accurate and comparable to WHOOP and Eight Sleep (as I discussed in my Fitbit vs. WHOOP comparison article).
I’ve even heard some reports of the Fitbit Versa incorrectly recording HR data while it wasn’t even being worn.
If the software algorithm in the Fitbit can’t even tell the difference between a real pulse and light reflected by a countertop, I have no trust in its ability to distinguish between the various stages of sleep.
Considering that the Apple Watch’s heart rate sensor and ECG feature are both FDA approved and have shown to be incredibly accurate in third-party testing (and my own validation trials), you would expect that sleep tracking would be fairly accurate as well.
Unfortunately, that’s not the case (yet).
Note that while the Apple Watch by itself doesn’t offer any serious sleep tracking features, I expect Apple to introduce those capabilities at some point.
When I first tried NapBot, the app detected hours of light sleep while I wasn’t even wearing the watch because it was charging. After I manually edited the incorrectly recorded sleep, I tried it again the next night.
Based on NapBot’s data, I spent only 2% of my sleep in deep sleep. If that was the case, I would have felt like trash that day.
The bottom line is that the Apple Watch in its current state isn’t a good sleep tracker — even though I wish it was.
The primary reason for that is the Apple Watch’s slow HR sampling frequency (every few minutes) and the lack of additional sensors.
You can learn more about how the Apple Watch compares to the WHOOP Strap in this side-by-side comparison.
Polar is known for making excellent heart rate sensors. I used to have a Polar chest strap in high school when I was a cross-country runner and I still have one today that I occasionally use during CrossFit workouts.
Based on their technological expertise, experience and feedback from users who have done side-by-side comparisons, I don’t doubt that their heart rate sensor is fairly accurate.
However, considering that the Polar M430 uses only HR and motion to detect sleep, I seriously doubt that sleep stage detection is accurate and reliable.
The latter is the reason why I didn’t even bother purchasing this wearable for testing.
Withings Steel (Nokia Steel)
Sleep tracking in the Withings Steel is a gimmick, plain and simple. The watch relies only on its high precision MEMS three-axis accelerometer. That means that if you don’t move your wrist for 15 minutes, it assumes you’re sleeping.
That’s all I have to say about that.
What Are the Best Non-Wearable Sleep Trackers?
As I mentioned throughout this article, accurate sleep tracking without relying on sensors that are in direct contact with your skin is incredibly difficult. That’s why I generally don’t recommend non-wearable sleep trackers, with one exception.
Pod Pro by Eight Sleep
The only device that I’ve found to be relatively accurate and reliable is the Pod Pro by Eight Sleep. It’s a temperature-controlled smart mattress that has sensors built into its mattress cover. The company calls it the ActiveGrid, and I’ve been using it to track my sleep for the past few years.
Based on the comparison data I’ve collected, Eight Sleep’s tracking capability is more or less in line with WHOOP.
You can learn more about the Pod Pro by Eight Sleep in this article.
If you want to give the Pod Pro a try, you can use the link below and get $150 off the purchase price.
As indicated above, there are several other non-wearable sleep tracking devices on the market but none that I’d consider to be accurate.
Still, despite my skepticism and all the negative reviews I’ve read of those devices, I ordered the Emfit QS to give it a try.
The Emfit QS is an under-the-mattress sleep tracker that can supposedly also capture your HRV. What’s convenient about the Emfit QS is that it goes between the box spring and the mattress, instead of directly underneath the sheets. That makes the sensor much less likely to move around or be uncomfortable.
The device relies on a technology called ballistocardiography. That means it can detect the force of your beating heart through the mattress via breathing movements. Think of it as a super-sensitive stethoscope.
The science behind ballistocardiography is sound, and there are numerous validation studies available to back it up. So I was interested to see how well this works in a shared bed, and how Emfit can translate a good HR reading into accurate sleep analysis.
Update After Testing Emfit QS
I had a chance to test the Emfit QS sleep tracker for a couple of days and, at first, the recorded data was similar to what WHOOP reported.
One night, my wife and I went out for a date night and I over-indulged in Persian kabobs. Needless to say, I didn’t sleep well and woke up at 4:23 a.m. A few minutes later, I got out of bed and started my day.
I looked at my WHOOP data and it correctly recorded that I slept from 8:46 p.m. to 4:23 a.m. Emfit, on the other hand, reported that I fell asleep at 7:50 p.m., had my first deep sleep episode at 8:30 p.m., and was in REM sleep between 4:50 and 5:10 a.m.
That’s completely inaccurate, and it confirmed my suspicion that contactless sleep trackers only work under perfect conditions, such as when you’re alone in bed and trying to sleep instead of having sex, or having someone next to you who might influence the sensor’s readings.
Beautyrest is a brand of the Simmons Bedding Company, which is a mattress manufacturer. While you could argue that they should know a lot about sleep, I frankly doubt they know much about the physiology of it or the technology required to accurately track it.
My opinion aside, Beautyrest uses passive piezoelectric sensors (pressure sensors) under the mattress to detect even the smallest amount of pressure from movement.
Using those sensors, Beautyrest claims it can accurately monitor respiration and heart rate, light sleep, deep sleep and REM cycles, in addition to body movement and interruptions.
Sleeptracker® by Serta Simmons is clinically proven to monitor sleep patterns with an accuracy of over 90% for 90% of the population compared to medically operated professional polysomnography.
The problem is that I couldn’t find any proof that the underlying technology is as accurate as Fullpower (the tech provider behind Sleeptracker) claims.
Beddit Sleep Monitor
Beddit also uses Piezo force sensors to detect movement. But instead of going under the mattress, you have to place the 2mm-thick sensor strip under your sheets.
What I like about Beddit is that the company doesn’t claim to accurately detect the stages of sleep. Instead, Beddit simply differentiates between time in bed, total sleep time, time to fall asleep and time awake.
The accuracy of Beddit’s data ultimately boils down to your sleeping patterns. If you go to bed without reading a book or watching TV, you can probably get reasonably accurate data.
However, if you often lie in bed doing something other than sleeping that doesn’t involve movement, you’ll probably get a lot of false positives.
Withings Sleep (Nokia Sleep)
Withings Sleep is another under-the-mattress device that can supposedly track your overall sleep, sleep stages, respiratory rate, heart rate and snoring.
The idea behind this product is that your beating heart, breathing or snoring all cause different movement patterns that the sensor can pick up.
As I mentioned before, I’m skeptical of whether a $99 device can accurately pick up and correctly identify those patterns, especially if several of them happen concurrently.
Unlike all the other contactless monitors I mentioned above, the ResMed S+ doesn’t have to be placed underneath the mattress. Instead, you have to position it on a nightstand between 1.3 and 3.9 feet away from your bed.
Additionally, you should angle the unit so it faces your chest, so that the ultra-low-power radio waves can detect your breathing and movement.
By relying so much on the correct positioning of the device relative to your body, I think there’s a lot of potential for stuff to go wrong, resulting in inaccurate data.
Additionally, it seems unclear if ResMed still supports this device. The official homepage has a 2015 copyright notice and a lot of reviewers on Amazon* have complained that the backend service that stores all your data has shut down.
Frequently Asked Questions
How accurate are personal sleep trackers?
Most of the personal sleep trackers on the market are gimmicks because they don’t have the required sensors to accurately detect sleep and, in particular, stages of sleep. To get accurate data, make sure to choose a device that combines the input from multiple sensors and that has been validated in independent studies.
What’s the difference between tracking sleep vs. tracking stages of sleep?
In its simplest form, sleep tracking means being able to detect when you’re not awake. While sleeping, you go through four stages of sleep, and the change from one stage to the other is often gradual (and thus difficult to detect).
Can sleep trackers really measure sleep stages?
Some devices, such as Muse S and the WHOOP Strap, have been shown in lab tests to reliably measure how much time you spend in each of the four sleep stages. However, most devices fail miserably in that regard, leaving their wearers with incorrect and misleading data.
How do you get more deep sleep?
You can increase the amount of time you spend in restorative phases of sleep — including deep and REM sleep — with a few simple lifestyle changes, including maintaining a consistent sleep schedule, avoiding stimulants and alcohol before going to bed, and other tricks that I discuss in this article.
How long should deep sleep last?
On average, deep and REM sleep combined should make up 40% of your total sleep. Your body’s requirements usually determine how much deep vs. REM sleep you get.
What happens when you don’t get enough sleep?
Your physical and mental performance deteriorates, you become irritable, and you might get sick. For more information on the negative consequences of sleep deprivation, check out this article.
Can sleep trackers diagnose sleep disorders?
Yes, some sleep trackers (e.g., Biostrap) can assist in detecting sleep apnea and restless leg syndrome by monitoring your respiratory rate and leg movement.
What is a good sleep pattern or sleep cycle?
During the course of a night, your body consistently cycles through four stages of sleep in approximately 90 to 120-minute intervals. In general, most deep sleep occurs in the first half of the night, and most REM sleep occurs in the second half of the night.
Which Fitbits track sleep?
Fitbit offers different levels of sleep tracking among its smartwatches (e.g., Fitbit Versa 2) and fitness bands (e.g., Fitbit Charge 3). You can see support for the three primary sleep tracking features below.
Sleep tracking: Sense, Versa 2, Versa LITE EDITION, Ionic, Charge 3, Inspire HR, Inspire, Ace 2.
Sleep stages: Sense, Versa 2, Versa LITE EDITION, Ionic, Charge 3, Inspire HR.
Sleep score: Sense, Versa 2, Versa LITE EDITION, Ionic, Charge 3, Inspire HR.
Does the Fitbit Charge track sleep?
The Fitbit Charge 3 can track sleep and give you a sleep score via the Fitbit app. However, I consider the sleep tracking of Fitbit mostly a gimmick, so don’t expect too much.
How good is sleep monitoring in the Fitbit Ionic?
It’s about as gimmicky as in the other Fitbit models.
What are the four stages of sleep?
The four stages of sleep include awake time, light sleep, deep sleep and REM sleep. Sometimes you also hear the term NREM (or non-REM) sleep, which includes light and deep sleep.
What is the best sleep app?
I haven’t found a smartphone app yet that accurately and reliably tracks sleep. The exceptions, of course, are apps that work together with reliable sleep trackers, such as the WHOOP band.
What’s a smart alarm?
A smart or silent alarm is supposed to prevent waking you up during deep or REM sleep. If you wake up during those restorative stages of sleep, you might feel groggy and not rested.
As you can imagine, a smart alarm is only as reliable as the ability of the tracker to accurately detect the stage of sleep you’re in.
However, if you maintain a consistent sleep pattern, you can wake up every day at the same time without requiring an alarm clock. That’s what I’ve been doing for the past year or so.
Is the WHOOP app available for Android?
Yes, you can download the WHOOP app from the Google Play Store. WHOOP used to be available only on iOS, but that’s not true anymore.
Does a GPS improve sleep monitoring accuracy?
Some devices, such as the Apple Watch and the Polar M430, have a built-in GPS. However, that’s only used for activity tracking, such as recording your running route. It’s got nothing to do with sleep monitoring and won’t improve its accuracy or reliability.
Best Sleep Trackers: Final Thoughts
Accurately tracking sleep is incredibly difficult because most of the consumer-grade devices we have access to can’t measure brain waves (with the exception of Muse S) — the only direct indication of sleep.
Instead, many wearable and contactless sleep monitors have to use proxies like movement, heart rate, skin conductance, ambient temperature and respiratory rate to guess — more or less reliably — if a person is truly sleeping or not.
Besides getting quality data from as many different biometric sensors as possible, it’s up to the underlying software algorithm to make sense of the data and report back to the user.
The algorithm is where I think most sleep trackers fall short. It takes validation studies and large amounts of data for algorithms to reliably predict if a subject is sleeping and in what stage the person is in.
Based on everything I’ve seen, there are only a handful of sleep trackers out there that have both the underlying technology and algorithm to accurately track and help you improve your sleep.
All the other trackers are gimmicks that might over-report or under-report your sleep, and are thus virtually useless if your goal is to introduce meaningful lifestyle changes that result in better health and performance.
What sleep trackers have you used, and what was your experience regarding their reliability and accuracy? Let me know by leaving a comment below!
I’m a healthy living and technology enthusiast.
On this blog, I share in-depth product reviews, actionable information and solutions to complex problems in plain and easy-to-understand language.
Best Sleep Trackers I’ve Tested [And Why Most Don’t Work] is written by Michael Kummer for michaelkummer.com