Smartwatches: What can they actually tell us?
You’ve just read the TRIAX blog on building your tank and have been inspired to hit the track again. Whether you are running for the first time in 10 years or a sub-elite athlete, one of the most important steps in building your tank outside of actually running is tracking the intensity and volume of training you are completing.
20 years ago, this might have involved walking laps of your oval with a trundle wheel or driving around the block and use your cars odometer to figure out how far you have run.
However, with recent advancements in technology we can live track and monitor our running speed, distance covered, energy expenditure, heart rate and location by simply wearing a smartwatch.
It’s no surprise then that millions of people have joined the craze and got themselves a smartwatch to track their daily physical activity. If you are reading this, you too probably have a smartwatch or are considering buying one in the near future but how much do you actually know about what your smartwatch can tell you?
What ARE YOU USING YOUR
Before you read any further you need to ask yourself what are you planning to use your watch for? To help I have listed some stereotypical smartwatch user types below to help you figure out what you want to use your smartwatch for.
Business User – “Fitness functions? What are those?” You have/want this watch so you can see your messages and emails without having your phone on you all the time. I think you might be in the wrong place reading this article if you have a smartwatch for the above reasons.
Casual User – You are interested in collecting some basic health and exercise data, such as your 10,000 daily steps and F45 classes. However, you mainly bought the watch because they are all the rave at the moment, and you can’t miss the opportunity to post about your “fitness journey” on your Instagram story to make your mates jealous.
Sub-Elite Athlete/Weekend Warrior
You are a serious about your exercise and want to use this watch to collect as much data as possible. In particular, you want to be able to accurately monitor your training to ensure you keep progressing your fitness and training. So much so, you are even considering buying add-ons such as a heart rate belt to increase the accuracy of your exercise data.
If you are a business or casual user you probably don't care about the accuracy of your watch, in fact you are probably hoping it overestimates your activity so you can gloat a bit more. If thats the case, this might be the finish line for you. However, if you are serious about your exercise I recommend you keep reading to find out what your smartwatch can actually tell you about your exercise and how accurate this information is.
Once you’ve figured out what type of smartwatch user you are, it’s time to understand what all that information your watch spits out at you actually means. The specific metrics your smartwatch will provide you with will vary depending on the brand and model you own. However, the primary metrics provided by the major smartwatch companies (Apple, FitBit, Garmin, Polar, Samsung) are generally the same. To help you comprehend these fitness metrics and to get the most out of your smartwatch I have put together a list of the most common smartwatch metrics and accuracy/limitations of each.
This one is a no brainer, but like all good watches your smartwatch also tells the time. Thankfully, you don’t have to sit there watching the clock in your kitchen just to set your watch. To get the correct time you just simply need to connect via Bluetooth with your smart phone and your watch will automatically sync with the local time zone.
You’ll no longer have to wear that pedometer you got out of a box of coco pops 10 years ago to track your daily steps. Unlike, your bright yellow coco pops pedometer which used a mechanical switch to track steps and could be easily cheated by simply shaking it, smartwatches use more accurate and modern technology to count your steps. Using in-built accelerometers, smartwatches record the number of steps you take when walking/running by capturing your arm movement, with the watch recording each complete swing of the arm as two steps having been taken.
For the most part the technology used in wrist based smartwatches has been shown to be quite accurate and reliable for tracking your daily steps, However, there are some instances in which their accuracy has been found to decrease (1):
Walking movements that lack arm movement (ie. pushing a pram, holding shopping bags, mowing the lawn) resulting in underestimating your steps.
Non-walking movements that involve arm movement (ie. stirring a pot on the stove, brushing your teeth, going over a speed bump while driving) resulting in overestimating your steps.
Research has also shown that the accuracy of smartwatch step counting can be limited by step frequency and gait type. (2,3) Studies investigating users who walk slowly or have an unusual walking gait (ie. waddle, drag foot or limp gait) indicate that smartwatches struggle to detect steps in those with inconsistent gaits and tend to underestimate the number of steps taken. (2,3)
Overall, wrist based smartwatch step counts are a mostly accurate measure and provide a great indication of your daily physical activity levels. However, if you have an unusual gait pattern or engage in high levels of walking with limited arm movement I would recommend that you take caution when interpreting your daily step count data as it may be inaccurate. (2,3)
TRIAX Bonus Bit
Evidence suggests that hip based pedometers provide the most accurate step count readings due to the limited interference of non-walking movements. (2) Why is this important you might be asking? Do you carry your smartphone in your pocket? Well, smartphones also have an in-built accelerometer that records the number of steps you take. If you are serious about your step count accuracy I recommend that you compare your smartwatch and smartphone data to see if there is any significant difference (Note - you need to make sure your phone is in your pocket anytime you are walking/running).
Distance & PACE
For those of you who have/want a smartwatch to track how far and fast you have run, these metrics and their accuracy will be very important to you. There are two primary ways in which distance covered are measured by smartwatches:
1. GPS Tracking
The most common method for tracking the distance and pace of an activity is through the use of in-built GPS receivers. I won't get into the nitty gritty of how GPS technology works but if you want more information on this check out Rob's blog on GPS Tracking in Team Sports. Of the two options, GPS technology is the most valid and reliable for tracking the distance and pace of outdoor exercise, with the error margin for distance covered and pace on most devices found to be 0.6% or less. (4)
Note: You cannot use GPS technology to track exercise completed indoors. So, if you do most of your exercise inside you will need to rely on the step count x stride length method.
2. Stride Length, Step Count & Duration
Don't fear, for those smartwatches without in-built GPS technology or for those times when your GPS is offline, there is an alternative method to measure distance and pace. Without access to GPS your watch will automatically calculate distance by using the following equation:
Distance = stride length x steps taken during activity
Once the distance value has been calculated the watch can then determine your movement speed/pace using the following equation:
Pace = distance/duration to cover distance
Note: These values are recalculated regularly by your watch to provide the live distance and pace data that you see on your watch while exercising.
If you are using this method to track your distance I recommend that you customise your stride length (this can be done in the settings on your watch or smartphone app) to ensure that you obtain the most accurate data.
Heart rate (HR) is a standard metric available on just about every smartwatch these days and is most commonly measured using in-built optical sensors. To explain the way in which these optical sensors measure your HR will require me to quickly take you back to the classroom for a science lesson. I promise I will break this down as quickly and simply as I can so we can start talking about the accuracy and validity of measuring HR on your smartwatch.
If you own a smartwatch, have you ever wondered what the green LED lights underneath your watch were for?
Well, they are actually part of the in-built optic sensor system used to measure HR. But how do these lights actually measure HR?
The optic sensor system works by emitting increasing levels of green LED light onto the skin underneath your watch and measures the amount of light that is reflected back off your blood, a process called pulse oximetry. (5) With each heartbeat, the optical sensor detects the level of oxygenated blood at the wrist by the change in the skin’s absorbance or reflectance of LED light. (5) Each time the volume of light reflected back to the watch is then processed using a complex algorithm to provide a HR value. (5)
Put simply, the more light absorbed = greater blood flow = higher HR
So, back to the important stuff, how accurate is this technology? Smartwatches containing optical sensor technology have been found to be accurate at measuring HR during exercise with a consistent HR and while resting. However, during intermittent exercise (ie. team sports or interval running) where HR constantly changes the accuracy of devices significantly decreases, with errors of +/- 20 beats per minute often seen. (6,7) Despite instances of inaccuracy, over extended periods of time (ie. a day or week) the average HR detected by these devices has been shown to be valid and reliable. (5-7)
Some further potential limitations of using smartwatches with optical sensors that you need to consider are: (7)
Skin Tone/Tattoos - darker toned or tattooed skin has the potential to impede the light from being absorbed/reflected by the blood, potentially resulting in inaccurate or skewed HR measurements.
Wrist circumference/High BMI - the larger the circumference/density of fat at the wrist, the greater amount of light that is required to penetrate the skin to be absorbed/reflected by the blood, potentially resulting in inaccurate or skewed HR measurements.
VO2max is lauded as one of the must have and gold standard smartwatch metrics but what actually is it? The VO2max metric is an estimation of your current cardiorespiratory (aerobic) fitness. To calculate this value, manufacturers have devised algorithms that analyse the relationship between internal (HR) and external (movement speed) workload during aerobic exercise. In addition to HR and movement speed/pace, these algorithms also rely on the user providing accurate; age, height and weight data to provide an accurate and individualised VO2max value.
If you want a more comprehensive overview of what VO2max is, how to interpret your score and how to develop it check out our blog on Aerobic Training Methods.
How accurate is the VO2max metric? In general, the VO2max metric provided by smartwatches is quite inaccurate and tends to overestimate your aerobic fitness. Although, the level of inaccuracy varies based on the accuracy of your individual characteristics (age, height & weight) and HR data. Apple and Garmin both outline that the accuracy of their algorithms are limited by the accuracy of their HR data, with Garmin publicly stating their margin for error is 5-10% when predicting VO2max.
Note: If you are serious about using your smartwatch for its VO2max and HR metrics I would recommend that you invest in a compatible HR chest strap to ensure the most accurate results.
Energy expenditure (EE) or calories burned as it sometimes known is another popular smartwatch metric, in particular for those looking to lose or manage their weight.
The EE metric represents the total number of calories burned, combining both resting and active calories. What are resting and active calories you are asking?
Resting Calories - Are number of calories expended while sedentary in order to sustain life (ie. to keep you breathing and your heart beating). This value is calculated based on your HR and VO2max during these resting/sedentary periods.
Active Calories - Are the number of calories expended while active/exercising. This value is calculated based on your activity level (ie. low/mod/high intensity), the type of activity performed and personal characteristics such as age, gender, height, weight and HR.
While, EE is a standard feature for most smartwatches currently on the market but is also one of the most inaccurate. So why is EE inaccurate? Smartwatch EE values are obtained using at least 7 different variables; age, gender, height, weight, movement type/intensity, HR and VO2max. There is also a high level of estimation, in particular, the age and gender data is normalised and does not discriminate for individual differences. (6) Therefore, when all of these variables are combined together there is high margin for error and inaccuracy. (6)
Training load (TL) is a metric that gives you an overview of the physical strain of your training. (8) The algorithms used to calculate TL vary slightly between manufacturers, however, each rely on the principles of the following equation:
TL = Duration (ie. minutes) x Intensity (ie. HR, pace)
This arbitrary value is typically presented on a per activity (in newer devices) or 7-day basis to provide an overview of your acute TL. Once you have completed a minimum of 2 outdoor cycling/running sessions for at least 4 weeks your watch will also provide you with a measure of your chronic (4-week) TL. After this period, your acute and chronic TL can be used to calculate other metrics such as training status and training load focus. (8) Similar to the above metrics, TL accuracy is dependent on the accuracy of your devices HR and VO2max data.
Acute 7-day Training Load
Low - Your acute TL is significantly lower than your previous 7-days and may not be sufficient for maintaining or improving your fitness levels.
Optimal - Your acute TL is ideal for maintaining and/or improving your fitness levels.
High - Your acute TL is significantly higher than your previous 7-days. You may need to reduce your TL in following weeks to appropriately recover to ensure you can continue to maintain/improve your fitness.
Chronic 4-week Training Load Focus
Overreaching - Your chronic TL is very high and might be counterproductive to your performance, think about taking a rest or reducing your TL over the next few weeks.
Productive - Your current chronic TL is ideal for continuing to improve your fitness levels.
Maintaining - Your current chronic TL is suitable to maintain your current fitness levels.
Unproductive - Your current chronic TL is too low or has not allowed for sufficient recovery, resulting in your fitness levels to decrease.
* training status & training load focus information is specific to Garmin devices *
Recovery time is a metric that provides a recommendation on how long you should recover for following your most recent bout of exercise based on your current fitness level (VO2max) and TL. (9)
While, in more modern devices this metric also incorporates health and wellbeing data when determining your overall recovery time. (9)
This is a more holistic approach and allows for both your physical performance and factors such as a stressful day or week (ie. work deadline or exam period) or poor nights sleep to be accounted for in your recovery time. (9)
While recovery time is a standard feature in most smartwatches on the market its validity is largely unproven and it is more of an estimation than an exact figure. Additionally, like the other predictive smartwatch metrics (VO2max and EE) the accuracy of recovery time is dependent on the accuracy of your devices HR, VO2max, sleep and stress data.
WHAT CAN SMARTWATCHES BE USED FOR?
So, we have all of these metrics but during what types of exercise or activities can your smartwatch provide you with data? Well, your smartwatch will collect data during just about any type of exercise or activity that causes a change in HR. However, more specifically devices now come with over 40 pre-installed workout modes including the following:
Walking/Running: outdoor, indoor, treadmill and trail
Gym/Fitness: strength, cardio, elliptical, stairs, rowing, pilates and yoga
Outdoor Recreation: hiking, surfing, skiing, rowing and snowboarding
Cycling: track, mountain, road, and indoor
Swimming: indoor, outdoor and open water
There you have it. That's TRIAX's Smartwatches: What can they actually tell us?
Want to know more about Smartwatches? Tune into our Youtube channel or episode 9 of the Deakheads Podcast where the TRIAX boys discuss all things smartwatches and whether they are worth purchasing.
Anything you think we missed? Let us know via our social links below.
To keep up to date with TRIAX Performance, please subscribe to our mailing list and follow our social media accounts.
If we can be of further help to you and/or your team in any way, please reach out and contact us!
about the author
Strength and Conditioning Coach
M.App.Sp.Sci, ESSA L1 Sports Scientist, ASCA L1 Coach
Available for individualised online coaching
2. Genovese V, Mannini A, Sabatini AM. A smartwatch step counter for slow and intermittent ambulation. IEEE Access. 2017 May 6;5:13028-37.
3. Duncan SJ, Schofield G, Duncan EK, Hinckson EA. Effects of age, walking speed, and body composition on pedometer accuracy in children. Research quarterly for exercise and sport. 2007 Dec 1;78(5):420-8.
4. Johansson RE, Adolph ST, Swart J, Lambert MI. Accuracy of GPS sport watches in measuring distance in an ultramarathon running race. International Journal of Sports Science & Coaching. 2020 Apr;15(2):212-9.
5. Støve MP, Haucke E, Nymann ML, Sigurdsson T, Larsen BT. Accuracy of the wearable activity tracker Garmin Forerunner 235 for the assessment of heart rate during rest and activity. Journal of sports sciences. 2019 Apr 18;37(8):895-901.
6. Xie J, Wen D, Liang L, Jia Y, Gao L, Lei J. Evaluating the validity of current mainstream wearable devices in fitness tracking under various physical activities: comparative study. JMIR mHealth and uHealth. 2018;6(4):e94.
7. Shcherbina A, Mattsson CM, Waggott D, Salisbury H, Christle JW, Hastie T, Wheeler MT, Ashley EA. Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. Journal of personalized medicine. 2017 Jun;7(2):3.