We’ve spent more than 120 hours walking, running, sleeping, grocery shopping, kettlebell swinging, and cycling (indoors and out) to learn everything there is to know about fitness trackers. After considering new options and testing 19 top-rated trackers over the past two years, we think the Garmin Vívosmart HR+ is the best fitness tracker for people who want to track their activity levels and progressively monitor their workouts.
The Garmin Vívosmart HR+ nails all the core competencies of a good tracker by combining an always-visible display, five-day battery life, and the most accurate heart-rate readings we’ve seen to date in a wrist-mounted tracker that’s waterproof to 50 meters. It also has basic-but-effective sleep tracking. But what really sets the HR+ apart is its advanced fitness tracking features like built-in GPS (for more accurate run-tracking and pacing feedback) and Move IQ (which automatically recognizes and logs other types of exercises like swimming or cycling). These advanced features make it worth the premium over your typical wrist trackers such as the original Vívosmart HR (which is still good at a lower price point). It also receives smartphone notifications and can control your phone’s music playback remotely via Bluetooth. The Garmin Connect app is compatible with iOS, Android, and desktop computers, but it doesn’t have as many social networking features or users as Fitbit’s app. We think Garmin’s more capable hardware and improved accuracy make it the better pick for most people.
The full-color Gear Fit2 is a sleek-looking gadget with many of the same features as our overall winner. It’s noticeably thinner and a bit cheaper, but it also has GPS, a heart-rate sensor, and automatic activity detection. It’s the only tracker that lets you receive notifications from any phone app and even interact with some of them: you can actually reply to texts and email with canned responses, a functionality that disappointingly doesn’t work on the iPhone app. But all these smart features come at the expense of a measly two-day battery life—which limits its efficacy for sleep tracking and means you have to charge it almost daily. It’s also not super reliable: More than once during testing, the tracker began the workout before locking on to GPS, which means you could end up halfway through a workout without distance and route recording. Overall, the Gear Fit2 is a slick and smart device, but the more reliable Vívosmart HR+ is better for tracking fitness.
If you just want a simple way to monitor and track your daily activity (including workouts), nightly sleep habits, and get reminders to be more active, the Flex 2 is a great choice—especially if all your friends are on Fitbit. Unlike other Fitbits, it’s water resistant to 50 meters so you can track swimming and shower with it. However, there’s no screen—just five status LEDs to track progress towards your daily step count goal. It also doesn’t track heart rate, but Fitbits in general continue to struggle with heart-rate accuracy, so we don’t see this as a major issue; it helps the Flex 2 maintain its slim profile and lower price. The Flex 2 syncs wirelessly to the Fitbit app on a smartphone or the Fitbit website on a computer to keep a record of your activity and link you to other Fitbit users—a real highlight, as research shows that friendly competition can be very motivating.
With no battery to charge and no pressing need to sync to a phone, the waterproof Garmin Vívofit 3 is the best option for basic tracking without the need to use a separate app. While pairing it with a phone lets you access more information, such as sleep tracking, the small screen is enough to get most of what you need at a glance: time of day, steps taken, steps needed to reach your daily goal (in case you don’t want to do math), distance traveled, calories burned, and “active minutes” (Garmin’s measure of sustained walks and other workouts). It also requires no charging, instead lasting up to a year on a replaceable watch battery. Despite its generally lower-tech approach to fitness tracking, it does still have some smarter features, like automatic workout tracking via Move IQ and inactivity alerts to get you going after a too-long sedentary stint.
For anyone who doesn’t want to wear something on their wrist, the Fitbit Zip is our recommendation. The tiny Zip is truly clip and go—it runs on a replaceable watch battery, so it requires no charging—and it has an easy-to-read display for you to check your progress at a glance. Costing at least $20 less than the other clip models, it’s the best value in the field. However, it lacks sleep tracking; if that’s important to you, upgrading to (and paying more for) the Fitbit One is your only solid option. Either gives you access to Fitbit’s very active social community to keep you going.
I’m a certified personal trainer (NASM-CPT), a running coach (USATF Level 1 and RRCA), and a regionally competitive runner. I also served as a staff writer for the Good Housekeeping Institute for nearly five years, working closely with the engineers and other scientists to interpret product test results.
Our survey of more than 800 Wirecutter readers told us that when they shop for a fitness tracker, they prioritize a slim design with a display, accurate step counting, waterproofing, and an integrated heart rate monitor. We also interviewed industry experts Jill Duffy of PCMag and Ray Maker of DC Rainmaker, and discussed heart-rate monitoring with Clinton Brawner, PhD, a clinical exercise physiologist at Henry Ford Medical Center in Detroit.
The lines that separate a dedicated fitness tracker from a GPS running watch, a smartwatch, or even a smartphone have never been blurrier. Now you can find fitness trackers with GPS and push notifications, running watches with fitness tracking, and smartwatches with both. But as our testing reveals, a device that’s designed to do all of those tasks isn’t necessarily good at any of them.
Generally speaking, if getting information or advice on your overall activity level is your primary goal, a fitness tracker is your best bet. These devices provide at-a-glance feedback regarding steps taken (as well as reminders to get up and move), measure the intensity of your exercise, and report on the quality of your sleep. Taken together, these measures can help you make positive changes to your health. Because their batteries typically last four to seven days between charges as opposed to the couple of days you get from a smartwatch, they give you a more complete picture of your fitness level. Such devices also let you set up friendly competitions among your relatives and friends who have trackers of the same brand, and their ability to pass data to other health apps on your phone enhances the value.
If you’re more curious about your daily movements, or if you want a quick way to track a run, you may be satisfied with the tracking capabilities of a smartphone and apps. Unless you turn it off, your iPhone (5s or newer) tracks your step count automatically in the Health app, and Google Fit on Android can track your steps, too. However, a smartphone is less practical for providing the whole picture of your wellness efforts unless you carry it on your body constantly. And it can’t track your heart rate or sleep quality without additional hardware or software.
No matter which of these options you choose, Brawner told us, “The most important thing is for people to be more active. If technology helps to motivate someone, then I am all for it. One trick doesn’t work for all.”
We want to be very clear that, while some of our picks and many of the models we tested measure heart rate, these trackers are not a replacement for a medical device. In fact, Fitbit is named in a class action lawsuit for the inaccuracy of its heart rate monitors. Nothing named in this guide is a substitute for a chest-band heart rate monitor. If high heart rates are a health concern for you, don’t rely on a fitness tracker to help you toe that line.
Two new models from Garmin—the Vívosmart HR+ and the Vívoactive HR—plus two from Fitbit—the Charge 2 and the Blaze—Samsung’s Gear Fit2, the UA Band from Under Armour, and the TomTom Touch fulfilled the criteria we followed on our last go-round: all of the expected activity tracking (steps, stairs, sleep) plus onboard heart-rate monitoring.
I set out to test these products both as a group and individually. To gauge step-count accuracy, I wore five of the new bands plus the winners of the last round (Vívosmart HR and Charge HR) simultaneously for two full days, paying attention to which devices’ instructions recommended wearing on the dominant hand versus the nondominant hand, if specified (I received the Charge 2 and the TomTom Touch later and tested them separately). I compared their readings with an older-model Omron pedometer, which was the winner of a Good Housekeeping test I once helped run (that is unfortunately no longer available online). Step counts were generally consistent among models from the same brand (meaning the Fitbits counted consistently high and the Garmins consistently low). So they all work fairly well if absolute accuracy is less important than just getting a general figure you can aim to improve.
The major low-end outliers: The TomTom Touch and the UA Band deviated from the control pedometer on average a mind-boggling 48 percent and 31 percent, respectively—terrible enough that we opted to disqualify them from further testing. One of the best performances came from the Samsung Gear Fit2, which on one day was 1 step different from the control, though that most likely is a coincidence; still, it was only 3.6 percent different the second day, so as far as step accuracy goes, it was most impressive.
To home in on what might be affecting the accuracy, we tried several ways to isolate arm movement from leg and hip movement. First, I clapped along to Queen’s “We Will Rock You” (resisting the urge to also slap my thighs) to determine if arm movement minus leg movement registered as steps. This would be particularly important for anyone who sits at a desk and doesn’t want their arm movement to register as steps while they’re sitting. The claps foiled all but one wrist tracker—the consistently low-counting Vívoactive HR.
I also took the trackers for a stroll pushing a grocery cart around my local supermarket to see if they could capture my strides when my arms weren’t moving. This is important because you want to get credit for stepping even if your arms aren’t moving, like if you were carrying something. Only two trackers registered any of the steps I took—the Fitbit Charge 2 and the Blaze—and even then they registered just a percentage of those recorded by the control.
In the gym, I did a marathon set of 90 kettlebell swings to determine whether arm-plus-hip movement logged any steps. The purpose of this is to see if a rhythmic, non-stepping hip movement can foil the trackers. The kettlebell swings were counted as steps by all of the trackers, including the control hip pedometer.
Next, I ran for 1 mile at a steady pace on a treadmill to see how well each device captured distance traveled. My movements were even and consistent in this test, as would be expected in a steady-state run. All of the trackers contain technology that determines stride length based on factors such as your height and movement intensity. On a few of them, you can set your particular stride length manually to improve accuracy. Instead of calculating and entering mine, I used the defaults in this treadmill test, assuming that most people might not fuss with something the trackers are supposed to automatically figure out.
Because three of these trackers have onboard GPS, I also did a few tests to gauge its accuracy. I ran four laps (1 mile) on a standard track, and I ran 1 mile in my neighborhood, as measured by MapMyRun.com.
The Garmins came the closest to matching the actual distance on both the indoor and outdoor tests, and the Samsung was not far behind. The Fitbits, which do permit customized stride length, both came up short, clocking in at 0.79 mile (Charge 2) and 0.81 mile (Blaze) on the 1-mile treadmill test (similar to the results I got last time with the Charge HR, Surge, and Fitbit One).
The bottom line when it comes to step-count accuracy: All of the manufacturers concede that accuracy can vary. After all, wrist movements are consistent with leg movements only when someone is striding forward in a more or less straight line, which is not always how people move. But given that these devices’ ultimate intention is to capture the user’s activity and encourage more of it, all models that made it past our initial round delivered adequate performance. To determine your personal baseline, wear your new device for a few days without making any concerted effort to change your usual habits, and then use those numbers to build on.
Having an idea of your resting heart rate—the slowest rate, measured in bpm (beats per minute), at which your heart pumps blood, a figure best taken before you get out of bed in the morning—helps you gauge improvements in your overall fitness. A resting heart rate that lowers over time is one indication that your heart muscle is getting stronger and more efficient at pumping blood.1 The other benefit to heart-rate tracking is the ability to quantify your training intensity.2
I performed two tests pitting each band against a Garmin chest-strap monitor: a five-minute steady-state treadmill run and a sequence of 30-second intervals of jumping jacks and recovery. For each, I recorded heart-rate readings every 30 seconds during the exercise and for 2 minutes of recovery following.
When we last tested for this guide, trackers were generally slow to notice an increase in heart rate over short intervals when compared with the strap monitor, and in far too many cases, they overshot or undershot the control heart rate by more than 5 bpm (on the very outside of the tolerance that exercise physiologist Clinton Brawner told us is acceptable for training use). We’re happy to report some good news: The Garmins and the Samsung performed remarkably well on the steady-state runs, with readings for the most part within a beat or two of the chest strap.
The jumping jack test wasn’t as successful; after my short interval of effort (30 seconds); every wrist tracker struggled to catch my heart-rate peak before it fell during the recovery period. The Samsung was most successful, often catching my elevated heart rate within a split second after I took my readings.
The bottom line is, if heart-rate training is your primary aim, you’re better served with a chest strap monitor. Otherwise, relying on the talk test (checking that you can say a few sentences without excessive breathiness) or gauging effort based on how you feel on a scale of 0 to 10 may be a better intensity indicator for your workouts, regardless of what your wrist monitor is telling you.
Finally, I spent some quality time with each band individually, assessing comfort, user friendliness, and overall impressions. I even wore four of the contenders to bed to get an idea of how they compare as sleep trackers. One of the most interesting takeaways: I used standard sleep tracking on the Fitbit Blaze for the first night of sleep, and it wasn’t very detailed, so I used the “sensitive” sleep tracking setting on the second night (which is turned on via the website, not the app). It’s apparently really sensitive, recording many of my movements as my being restless/awake—I’d be making an appointment with a sleep lab post-haste if I actually snooze this poorly! (I got similar results with the Charge 2 when I tested it separately in both regular and sensitive mode.)
The Vívosmart HR+ has everything you’d expect from a good tracker but adds onboard GPS tracking and activity auto-detection, both of which performed above average in our tests. It still includes all the standard fitness tracker functions we’ve come to expect: step and stair counting, distance traveled, calories burned, and sleep statistics. But its long battery life (4 to 5 days without GPS) and full waterproofing (down to 50 meters) give it an edge. It will stay charged almost four days longer than the Gear Fit2 if you don’t use the GPS, and unlike most Fitbits and the Samsung, you can wear it in the shower after your workout.
The Vívosmart HR+’s most significant leg up against the competition is its integrated GPS. This is not a huge surprise from Garmin, a company best known for its running watches, cycling computers, and car navigation units. What is surprising is that the developers managed to fit a GPS receiver and an HR monitor into such a small package—it weighs just 1.4 grams heavier than the first model, sticks up only a hair more, and is just as comfortable as long as it’s not strapped on too tightly (the raised lip of the heart rate monitor can leave a dent). It’s slower to connect to a signal compared to larger running watches or your smartphone, but it’s impressive nonetheless.
With the added GPS, you’re able to get more detailed information about your walks and runs such as distance and pace. It also informs some of the workout modes, which include run-walk intervals and a virtual pacer that alerts you if you’re moving slower or faster than your preset pace. The GPS proved accurate, though it took longer to connect (minutes rather than seconds) than the near-instant but much larger Vívoactive HR—that’s the cost of going small. If you want to use it on a bike ride, you’ll have to select the outdoor “cardio” setting and change the activity classification later in the app or within your account at connect.garmin.com. Even without GPS, we found the distance-measuring capabilities (based on internal algorithms that take height and cadence into account) of all three Garmins to be excellent on our 1-mile treadmill test.
The other interesting addition is Move IQ, Garmin’s activity-detection software. If you head out for a workout—running, walking, biking, swimming, elliptical training—and don’t remember or don’t bother to turn on the correct exercise mode (or any exercise mode at all), this technology auto-detects that you’re being more active as long as you do it for 10 minutes or longer and even classifies what you’re doing. You get credit on the app for your hard work, though the details aren’t as robust as when you use the timed workout modes. Fitbit and Samsung have similar software, so while this is a good add, it’s somewhat de rigueur now in high-end trackers.
Though Garmin tells us that the HR+ uses the same Elevate heart-rate measuring technology as the first Vívosmart HR, our tests showed that its measurements were much more accurate than its predecessor’s. This was particularly true during the steady-state run, during which the HR+ produced near-perfect data when compared with a dedicated heart-rate strap. However, like most wrist-based trackers, it struggled to keep up with short intervals, not registering a peak in my heart rate until I’d already entered my rest period. Overall, if your heart-rate training predominantly requires you to hover around the same bpm in steady-state exercise, you’ll be happy with the HR+, but if you want to use peaks and recoveries or you need to measure your heart rate more accurately for medical reasons, you’re still better off with a strap.
Naturally, the HR+ tracks more standard information as well: steps walked, floors climbed, calories burned, distance traveled. In terms of step-count accuracy, it consistently undercounted the control pedometer on our tests (as did the previous iteration), yet it counted both hand claps and kettlebell swings as steps—more or less a break-even in the end. (Overall, we’d prefer a tracker that tells us we haven’t done enough than one that consistently overcounts.) Like the others, it auto-detects sleep and provides a log of hours spent awake and in light or deep sleep, based on your movement.
The touch-sensitive passive OLED screen does the job and, unlike the Fitbits and Samsung Gear Fit2, its backlight comes on with a touch—no twisting of the wrist to get it to pop up. Garmin claims the battery will last for up to 5 days of activity tracking without GPS, which seems about right in our testing. That said, it’s not nearly as nice to look at as those full-color active LED screens, or even the passive full-color Vívoactive HR.
The screen views are customizable, with displays showing step count, flights of stairs climbed, intensity minutes (a measure of workout intensity over the course of a week), calories burned, distance traveled, and heart rate (current and average resting). You can receive all manner of smartphone notifications—phone, text, calendar, email, and social media. It has optional music controls, at-a-glance weather forecasts, and a handy “find my phone” feature that lets you play Marco Polo with your misplaced phone if it’s in Bluetooth range. If you happen to own a Garmin VIRB camera, you can use the Vívosmart HR+ as a remote control for that as well, but we didn’t test this function. Like its predecessor, it’s completely waterproof to 50 meters (as opposed to just “splash-proof” like Fitbit and “water-resistant” like Samsung) for worry-free use in the shower or pool.
The Garmin Connect app and website have seen no major updates since our last test (when it had just received a huge overhaul), and remains utilitarian, customizable, and detailed, though not as easy to navigate as Fitbit’s app. However, Garmin’s Apple Health integration gives you more flexibility in how you use your data than the Fitbit, which still has no plans to integrate with Health.
The app that Garmin offers with the Vívosmart HR+ can’t compete with Fitbit’s, and that’s its biggest flaw. Garmin Connect has steadily improved from its early days, but it still feels overly complicated next to the Fitbit’s streamlined simplicity. This is largely due to the fact that Garmin uses this app to sync data from a large variety of devices that range from fitness trackers to cycling sensors to top-of-the-line triathlete and mountaineering watches. Meanwhile, Fitbit only needs to focus on the casual fitness tracking audience. As a result, Fitbit’s app shows everything a person casually tracking his or her fitness could want in one scrollable screen, whereas Garmin Connect requires multiple screens to show the same thing. It’s not bad; it just sometimes feels like a convoluted way to access the info you’re after.
The other area in which Garmin lags behind Fitbit is in social integration. When it comes to fitness, friendly competition is a great way to inspire improvement, and although Garmin isn’t hurting for users, the company doesn’t have the audience of step-count-focused users that Fitbit has. Garmin is making strides to change that by installing activity tracking in a number of its newer GPS watches, but its core audience still skews much more heavily toward enthusiast runners and cyclists.
We found the heart-rate functionality to be very good for a wrist tracker. It nailed the steady-state test, including heart-rate recovery once stopped. On the interval test, it was the only tracker to capture both the peaks and recoveries; every single time I was about to record what seemed like an erroneous reading, it would snap to an accurate value in the split second it took me to read the chest-strap watch—as far as we’re concerned, that’s a win.
The auto-activity tracking is a real highlight. After about 10 minutes of a sustained activity, the unit vibrates and the screen switches to a timer that indicates approximately how long you’ve been doing said activity, with a nice “Keep it up!” message to boot. It registers that you’ve stopped within a minute or so, and rewards you with the words, “Well done! That was a great [walk/run/ride].” It sounds cheesy, but we won’t lie: We missed it when testing other trackers that aren’t so cheerleader-y (Garmin’s “Move Bar Cleared” is a nice effort but kind of underwhelming). It has a pleasant way of goading you to get off your duff after a long period of sitting still; after 50 minutes of “inactive time,” a screen reminds you it’s “Time to get on your feet.” In comparison, the Garmins just say: “Move!”
The GPS is another area where we had some issues. Most GPS-enabled watches let you wait until the GPS signal is acquired before you start the workout timer. When you select an outdoor activity with location, Samsung starts a countdown from 3, then tells you to start your workout, with no immediate acknowledgement that it has found a signal (later, possibly some miles into your run, it will tell you if it hasn’t). I did one 7-mile run in which the Gear Fit2 didn’t find a GPS signal for at least the first 3 miles. During the time it’s not connected to the satellites, it makes an estimate of distance traveled based on internal algorithms and my step cadence. For this measurement, it did an acceptable job on my 1-mile treadmill test, so the estimating is probably okay, as long as you’re not a stickler on record-keeping. When the GPS does record, though, it is very accurate and even shows a graphic map of exactly where you went on the unit’s screen (it’s very cool).
The Gear Fit2 was also the most accurate step counter compared to the pedometer I wore. On the first all-day test, it was one step different, which has to be mostly a coincidence, as on that day I rode my bike to and from work, taught a running class, trained clients, and worked out, as well as walked a bunch. Indeed, I found that it counted kettlebell swings and claps as “steps.” Day 2 wasn’t quite so dead-on, but overcounting by only 3.6 percent isn’t too shabby.
Its full-color touchscreen offers a lot of viewing options. It’s dark unless the wrist is raised (to save power) but reacts quickly and reliably. You can completely customize which scrollable screens are shown, and screens for steps walked, floors climbed, and heart rate can be expanded for history. Two weeks of data can be viewed at a glance. There are even manual counter screens where you may keep track of your water and caffeine consumed, respectively, as well as a “together” screen that can show you if you’re ahead or behind other users in a step-count challenge; it’s a nice idea, if you happen to know anyone else who uses the S Health app (of the nearly 400 contacts in my phone, only two came up).
The Gear Fit2 isn’t quite as capable as a smartwatch, but it comes pretty close. It can be set up to play music from your phone within Bluetooth distance, including Spotify, or from the 4 gigabytes of internal memory on the unit itself, which can be loaded from tracks saved to your phone. It can display notifications from any phone it’s paired with, and even respond to texts with canned replies on Android. Finally, there’s also the option to download various apps from the Samsung Gear store.
The Gear Fit2 has 15 workout modes for more detailed exercise tracking, but a few of those are for specific exercises like squats and lunges, which show a graphic of how to properly perform them; these are cute, but largely extraneous. Annoyingly, there’s no Cardio option, and the “other workout” doesn’t track heart rate, so when I did my jumping jack interval test, I ended up using the indoor treadmill mode to capture my pulse.
As with Garmin, the Samsung app is also a bit of a weak point. Perhaps we should say “apps,” as you need two: one to control the Gear Fit2’s settings and the other, S Health, to interpret and log your activity data (plus two more background “helper” apps if you have a non-Samsung Android). In a way, we didn’t mind that there were two—it made it easier to figure out how to change settings on the device rather than digging around in one app for that info. Another disappointment is that while Samsung finally began offering an iOS app, you can’t reply to texts or emails from an iPhone.
S Health suffers from the same problem as Garmin Connect: With so much customizability, it can make finding the data you want overly complicated. Still, it’s all there, for your data-drilling pleasure.
Another quibble is that the screen does turn off even during workouts, and though it becomes visible again quickly with the flick of your wrist, there’s a slight delay that can be bothersome during a fast-paced workout. Another complaint: I wasn’t always getting regular HR readings in the passive mode; one day, I ran a five-mile race with it on and deliberately didn’t set the workout mode—it didn’t take a single reading during that time, when quite clearly my pulse would have been significantly elevated. Finally, it’s lovely to look at, but only when you can see it. It gets a terrible glare in sunlight, making it very tough to read during outdoor activities.
Another minus is the battery life. With moderate-to-heavy use including smartphone notifications (but not always GPS), we had to charge the Gear Fit2 about once a day, which is well below the three to four days claimed by Samsung for “typical” use. That said, it became an easy habit to plop it on the charger while I showered and it charges quite quickly. I also found that all that Bluetoothing for notifications killed my Google Nexus 5X battery even faster, so I now turn the Bluetooth on only when my phone is in my bag or similarly less accessible.
In long-term testing the Gear Fit2, two of our testers experienced an issue in which the band refused to accept a charge from its charging cradle. Mentions of this problem happening both after months of use and right out of the box have also cropped up on the Samsung community message boards. In some cases, flipping the watch over on the cradle (button-side down) or shutting down the watch before charging allows it to connect. If neither of those methods work, the band is effectively bricked. A representative from Samsung told us that the company is “looking into each reported case to identify the specific cause. If it is identified as a warranty defect, Samsung will replace the unit.” We will be mindful of this issue when we test the new Gear Fit2 Pro, due out in September 2017.
It’s also not fully waterproof, though with an IP68 certification, it would be able to handle an accidental dunking (up to 30 minutes in 5 feet of water).
Flex 2 also knows when you’re not being active, sending out a soft vibration if it detects you haven’t moved in the last hour. Unlike other Fitbits, the Flex 2 is waterproof to 50 meters, which means you can track swimming workouts but also not worry about damage if you want to wear it in the shower or accidentally drop it into the tub.
Without a screen, you lose out on instant feedback and even things like being able to see the time, and while those lights also can be set to go off in different colors when you’re receiving a call or text — Fitbit calls these “smartphone notifications” — if you really want more detailed info on your wrist, you should at the least choose our non-HR runner-up, the Garmin Vívofit 3.
The Flex 2 lacks heart-rate monitoring, but Fitbit’s heart-rate technology continues to lag behind Garmin’s and Samsung’s in terms of accuracy anyway, so there’s no reason to pay the premium for that functionality in the Charge 2 or Blaze, unless you want a better display or connected GPS for accurate distance measuring of walks, runs, and bike rides. It should be noted that all of the non-HR models tested have similar capabilities, but there’s one thing that the Flex 2 (and the Alta) provide with which other companies can’t compete: access to the vast social community that Fitbit has cultivated, letting you join step or other exercise challenges against other users — a powerful way to keep you motivated to up your fitness game.
However, both the Flex 2 and the Alta way overcounted on steps, at least in the traditional sense: up to 30 percent over the recordings of the control pedometer. Unsurprisingly, they both counted claps and kettlebell swings as steps, and missed the lion’s share of actual steps made during the grocery cart test. But at least the recordings were consistently high, so if you’re serious about hitting 10,000 steps, you may just want to aim for 13,000. Also typical to Fitbits, at least in our tests, was a tendency to underestimate distance. It recorded only 0.73 miles of a 1-mile run.
Flex 2’s battery life is an estimated five days, on par with many of the models tested, but if you want a tracker you really don’t have to think about, our runner-up, Garmin Vívofit 3, operates on a battery that’s good for a year and doesn’t need to be charged. Finally, the band that comes with the Flex 2 can be a total PITA to fasten, often leaving me with sore thumbs as I pressed and pressed to get those little pegs into the holes.
The Vívofit 3 runs on a battery that doesn’t need charging and should last for a year before it needs replacing, removing all need for daily or weekly maintenance. The always-on display provides at-a-glance time of day, steps taken, steps needed to goal (in case you don’t want to do math), distance traveled, calories burned, and “active minutes” (Garmin’s measure of sustained walks and other workouts) — all of which can be edited out if desired via the Garmin app. Like most new trackers, the Vívofit 3 also notices if you’ve been sedentary for too long and beeps to remind you to get on your feet, and it includes automatic workout detection, which loads activities to the Garmin Connect app when in a paired smartphone’s Bluetooth range, should you choose to pair it.
As for accuracy, the Vívofit 3 fared a bit better than the Fitbits but still measured high on our tests: about 13 and 14 percent over the pedometer on each day tested. It, too, counted claps and kettlebell swings as steps and missed all the steps when pushing a grocery cart. The bottom line, really: If pure step count is your aim, go for a waist-worn tracker or pedometer. Curiously, the distance measuring was subpar for a Garmin, clocking just 0.86 of the mile run; we found the other Garmins tested all measured quite accurately, even without GPS.
For a basic clip-on tracker that doesn’t have to be worn on your wrist, look to the Fitbit Zip. It couldn’t be simpler to use. Its passive LED display is always visible, unlike those of the Fitbit One and the discontinued Withings Pulse Ox, which both require a tap or a button press to light up. And with a tap, you can scroll through your stats, unlike on the Nokia Go, which shows only the time and an estimate of your daily progress toward your goal on its e-ink screen.
The Zip also requires no charging, running instead on a replaceable watch battery that should be good for six months. It syncs up wirelessly to the Fitbit app on your iPhone or Android handset, or to the Fitbit website on your computer, our favorite companion software for any fitness tracker. There, you can keep a record of what your tracker records, as well as join competitions against other Fitbit users, which can keep you going as you aim to improve your fitness.
With regards to accuracy, we found that that both the Zip and One registered step counts that were about 8-9 percent high compared to our control pedometer. Neither tripped up on the biking test, but they both registered kettlebell swings as steps. However, one might argue that the Zip is getting a better idea of your activity by including some of those irregular non-stepping movements in the mix—and as the device is clipped to your torso or waist, those movements are more likely to be valid indicators of your true activity level, as opposed to what you get with wrist trackers that can capture just arm movement. If pure step-count accuracy is essential to you, both Nokia devices we tested were more on a par with the control pedometer, at just 2 percent over.
The Zip wasn’t great at estimating distance on our treadmill test, but then again, none of the clip trackers were. The two Fitbits were short, measuring only 0.79 mile (Zip) and 0.87 mile (One) during a 1-mile run. The Nokia devices were long at 1.16 miles (Go) and 1.23 miles (Pulse). The Fitbit trackers allow you to set your stride length manually in their apps, though, so you may have better results if you do that.
Note that the Zip doesn’t offer sleep tracking. The other three tested clip models do, but you must wear them on your wrist for them to do so (bands are included), and only the Nokia Go does it automatically—you have to remember to start the sleep functions on the Fitbit One. For these reasons, we think a wrist-based tracker that records sleep automatically is a better all-around choice for evaluating your z’s.
Finally, the Zip is not waterproof—none of the tested clip models are—so don’t leave it attached to your jeans when you toss them into the laundry.
Samsung announced the Gear Fit2 Pro, an upgrade to our runner-up pick, the Gear Fit2. It has a lot of the same features as the original, like Spotify support, but the Pro has continuous heart-rate tracking and is also water-resistant, so you’ll be able to track your swims. You can preorder the Gear Fit2 Pro now.
Garmin announced its new line of wearables, including the Vívosport. The Vívosport combines the features we love about our current top pick—including GPS and HR monitoring—with the sleek look of the Vívosmart 3 and a full-color screen. It’s available to preorder now for $200, which makes it more expensive than any of the fitness trackers we currently recommend. We won’t be sure whether the Vívosport is worth that higher price until we test it, which we plan to do once it’s available.
The Garmin Vívoactive HR has 15 sport-related workout modes, which may appeal to the sport dabbler. But it’s a bit big for a simple fitness tracker, and many of those who would gravitate towards it would be better served by a dedicated biking or running device.
Our former top pick, the Garmin Vívosmart HR, is still a worthy choice, if not as advanced as the new HR+ and not quite as on-the-ball with its heart-rate tracking. Otherwise, it looks, feels, and performs the same as our top pick.
The Garmin Vívosmart 3 appears to be Garmin’s answer to Fitbit’s design aesthetic, with its dark-unless-tapped-or-jostled screen. This felt like a miss, though, because the screen is finicky and slow to react. Otherwise, its on-board fitness-tracking features are nearly identical to the Vívosmart HR, but without GPS. But if you want a solid basic activity tracker that isn’t a Fitbit, it fits the bill, with a couple of fun bonuses in all-day stress tracking (based on heart-rate variability) and estimated VO₂ max (a gauge of aerobic fitness level).
The user-friendly Fitbit Charge 2, with its exercise auto-detection and motivating movement alerts, is a fine pick, especially if you want workout modes to log your exercise sessions manually or would like connected GPS to track runs or bike rides using your phone. But with the erratic heart-rate tracking technology, Fitbit isn’t for you if your exercise program follows strict heart-rate zones. The updated sleep-tracking software (also available on the Alta HR and the Blaze) utilizes resting pulse rates to better inform how your night went.
The Fitbit Alta HR, an update to the Alta (below) has a trim appearance with many of the same features of its predecessor, plus on-board heart-rate monitoring like the Charge 2 and the Blaze (also below). The Alta HR lacks a workout mode, relying instead on exercise-recognition software. This makes tracking your heart rate during a workout nearly impossible (I did my HR tests by watching the HR rate via the smartphone app), but that’s just as well, as we found the Fitbit tracking software to be rather erratic and inaccurate during exercise. It seems to work just fine, however, when informing the sleep-tracking data.
The Fitbit Blaze offers robust smartphone notifications and Fitstar-guided workouts, but we don’t believe these differences from the Charge 2 justify its higher list price. It’s also a lot larger on the wrist, which not everyone likes in a fitness tracker.
Mio Slice adds an interesting new element to the market, measuring all-around activity with its proprietary PAI (“personal activity intelligence”) rather than highlighting step count as the primary stat. Even after testing, it still seemed a mystery as to exactly what counts and what doesn’t. Still, the challenge of hitting activity targets over a week’s time rather than a day may be more motivating to some users. The onboard heart-rate sensor appears to be quite accurate, though without an always-on screen, it can be annoying to view during a workout.
We tested the UA Band, part of Under Armour’s heavily promoted HealthBox. For a variety of reasons, it didn’t measure up: In my two-day-long step-count tests, its accuracy was abysmal, though consistently so: 30 percent and 32 percent too low, respectively. It’s stiff and the fit is odd: too oblong for my wrist, so it creates pressure on the top and bottom yet is a little loose on the sides. Finally, the underwhelming PMOLED display looks fuzzy.
The TomTom Touch purports to measure body fat percentage and muscle mass along with standard fitness tracker metrics such as heart rate. However, it performed so poorly in our basic all-day step-counting tests (registering almost twice as many steps as our control on both days!) that we’re setting it aside for now.
The Misfit Ray abstracts activity as “active points” that you earn by moving in a variety of ways, which it auto-detects. The smartphone notifications are about as helpful as Flex 2’s (e.g., not very) and Fitbit’s social aspect and Garmin’s super simplicity edged it out as a pick. But if you like its look or you’re an avid swimmer, it’s worth your consideration.
Fitbit Alta’s narrow screen is so small that the info it displays is truncated (step count is rounded up, for example), which somehow feels like a tease. Its tracking results were similar to the Flex 2’s (not very accurate) and it’s not swim-ready, though you can wear it in the shower (unlike other Fitbits, which are merely splash-proof).
The Withings Pulse Ox’s somewhat blurry-looking display and lack of any sort of smartphone notifications make it feel behind the times in comparison with the other units. Features like fingertip pulse tracking and blood-oxygen reader couldn’t make up for these shortfalls. The Pulse Ox has been discontinued in favor of the upcoming Nokia Steel HR. We’ll see how its new features hold up when it comes out in the fall.
Fitbit’s Charge HR was a former top pick, and it’s an acceptable option if you want heart-rate info and don’t mind the tiny screen, but you’ll probably have to find it on close-out as it’s no longer available on Fitbit’s site.
While initially intriguing, we dismissed the Microsoft Band 2 during our last test due to its consistent overpromising and under-delivering, questionable accuracy, and awkward landscape-only screen view. It has now been discontinued.
The Fitbit Surge didn’t make our cut, as its GPS features were substandard compared with what was available in fitness-tracking running watches when we tested. For its sticker price, you can do much better.
The Polar A360, reviewed in our last roundup, simply didn’t measure up in either its tracking or other functions, and was particularly disappointing in its heart-rate measurements, especially since that technology, in chest-strap form, is what Polar is best known for.
The Fitbit One is a higher-end option for anyone who wants a clip-on fitness tracker that also records sleep. Unlike the Fitbit Zip, it has an altimeter for measuring stairs climbed. The One registered step counts a bit high on our tests and came up short when measuring a 1-mile run (0.87 mile recorded).
The Nokia Go has a passive e-ink display that only shows a stylized watch screen and a chart that estimates progress against your daily step goal (both of which aren’t the easiest to understand at a glance). To get more detail, you must sync it to the app. It also costs $20 more than our clip-on pick.
The Huawei Fit’s phone app doesn’t connect to a social network, and the touchscreen often misses the difference between vertical or horizontal swipes. Despite its larger, watch-face-shaped screen, notifications are still hard to read due to the strangely huge font. In spite of its much lower cost, we think if your main goal is to track your activity, one of our top picks is a better bet.
Jawbone recently ceased selling its products on its own website, and the units are going at very steeply discounted prices on Amazon, leaving us to assume they’re on the way out. Business Insider recently reported that the company is out of inventory with no current plans to restock and has even stopped paying customer service contractors due to a lack of funding. We’ve reached out to Jawbone for comment, but have yet to hear back. In the meantime, we definitely can’t recommend buying them. When we tested the Up2 and Up3 we found they were generally more accurate than the other trackers tested on step count and distance measuring, and the Jawbone app does a nice job of presenting the data and providing advice on ways to improve (“Smart Coaching”).
(Photos by Michael Hession.)