I’ve been wearing the Apple Watch since Launch Day to work, to play, to exercise, and to travel. This is Part 2 of a three part series. Part 1 is here.
It was hard to break new ground in Part 1. I’m just a guy who managed to be awake at 3AM EST on the right day. Whereas tech press had both dedicated staff and a head start. Before my watch arrived, The Verge even trumpeted theirs with the title, Apple Watch: the definitive review.
Yet every evaluation of the health and fitness features I’ve read to date has no depth. They jump on an elliptical, describe the features, and call it a day. Little to no thought given to the practical or even strategic impact of these features. It’s just not their wheelhouse. But it is kind of close to mine. So, to quote Hansjörg “Giorgio” Moroder,
Well, now I may have a little bit of a chance.
The Great Watch Hope
As a huge Healthcare IT nerd, I often say: Compliance is the great challenge of modern healthcare. We know healthy choices will help us live longer. But adopting those habits is difficult even for very sick people. We just naturally seek maximum enjoyment and we’re completely willing to round our choices up or down to get there. It’s super easy to come up with an excuse not to exercise if you want to do so. As reported by Planet Money, your mind will even round up your memory of how often you’ve gone to the gym. A huge chunk of the fitness industry is founded on this idea: People will sign up for a gym membership and then convince themselves not to go.
So I’ve long believed we could all benefit from an automatic, objective dashboard of our health choices. Earlier fitness trackers like the Fitbit and the Fuelband tried to do this, too. But highly detailed trackers were overwhelming to novices, especially if they required hooking up to a PC. On the flip side, ultra-simplified stats like Nike’s Fuel Points seemed out of touch with reality.
Apple’s made this leap before. They turned the MP3 player from a niche product into the category-defining iPod. iPads aren’t too shabby when you think about it. And the story of the iPhone is practically a modern legend. But this time they’re aiming for more than just defining and/or dominating a category. This time, they hope to save you from yourself, starting with how much you sit around all day. To quote Tim Cook,
“If I sit for too long, it will actually tap me on the wrist to remind me to get up and move, because a lot of doctors think sitting is the new [smoking],”
Is that right? Is sitting the new smoking? Only when you consider total damage and ignore other unhealthy habits. On an individual basis, smoking is still ridiculously more dangerous and sugary drinks might be doing something terrible all their own. But, if you don’t smoke and eat a healthy diet, reducing your hours of sitting is a good next step.
In fact, that’s why I switched to a standing desk years ago and expected to sail past this particular health metric — My standing isn’t minutes per hour, it’s hours per day. Alas, the Watch has a very high error rate with my standing desk. There were so many false positives in the first week that I began to develop a theory:
There’s reason to believe this isn’t strictly true. But my brother and I saw it over and over with many standing activities. And we do know that the famously secretive company tests products mainly with its employees. That’s a pretty constrained data set. What else might Apple have overlooked? If only we could get a look inside their top-secret fitness testing facility. But we’d never get a peek, let alone a guided tour from some Apple executives, right?
Ah. Well, then.
A few things jump out at me immediately:
Indirect Calorimetry (the masks) is a very accurate calorie measure.
Test subjects are in a tight band of body types and demographics.
Testing was focused on steady-state cardio activities.
The only non-cardio activity mentioned was Yoga.
Apple plans to continue testing.
So, for some people, Apple’s doing rigorous research where it matters. If you sit at a computer all day (and most early adopters probably do) then you might look a lot like their testing cohort. This really goes to a point made by my brother about the stand alerts in particular and the Watch’s health measures in general:
The algorithms are bad, but their heart is in the right place.
It’s a good point. Even if the Watch fails to credit you with all of your effort, standing and exercising a little more is more likely to help than hurt most people. So if you look like an Apple employee who does mainly cardio, or if you just want to stand up and exercise more, then the Watch will help. I am confident its basic design can do that.
However, if you’re a little more intense, or just looking for some deeper analysis, then follow me now as we leave Apple’s designated fitness area for the real world.
Chasing The Megacalorie
The Watch tracks three health goals (Stand, Exercise, Move) but only one can be set by the user. Stand is locked at 12 times/day and Exercise is locked at 30 minutes/day. Only Move is a personal goal. On the first day, I aimed high and picked 1,000 ‘Active Calories’ per day. And then I proceeded to fail at reaching my goal for almost a week.
Whether wishful thinking or just plain hubris, I stubbornly stuck to my goal as the Watch suggested (every Monday morning…) that I give up and aim lower based on the previous week’s average. Sorry, that’s just not in my own particular idiom, Apple Watch.
Obstinate pursuit of The Megacalorie turned out to be informative. The drive to find ever more physical activity made me test the Watch in a broad array of situations. I was constantly on the lookout for extra calories. Anything and everything could be a workout.
I put the watch through 42 workouts in four weeks and then took it on hiking trips through Sequoia, Yosemite, and Muir Woods.
In the real world my activities included walking, rucking, hiking, backpacking, running, biking, skateboarding, and the range of Crossfit movements: Olympic Weightlifting, Powerlifting, Atlas Stones, Kettlebells, Chippers, Burners, etc. In the world of the watch these were narrowed to five kinds of activity: Running Outside, Walking Outside, Cycling Outside, Other, and not categorized.
Apple Watch Calories (‘Active Calories’) proved to be a little odd. Here are some per minute averages from my results:
Backpacking, 30+ lbs (not categorized): ~2.5 Calories
Rucking, ~33 lbs(‘Outdoor Walk’): 4 Calories
Walking, no weight (‘Outdoor Walk’): 4 Calories
Cycling, 5 lbs (‘Outdoor Cycle’): 7 Calories
Skateboarding, 5 lbs (‘Other’): 7.6 Calories
Crossfit, various (‘Other’): ~10 Calories
Running (‘Outdoor Run’): 11.2 Calories
Compare the workouts within broad domains (on foot, on wheels, etc.). What do you see? The Watch’s cardio-centric worldview rewards speed over difficulty. Do some pause back squats with a couple hundred pounds and the Watch is like, “That’s nice…why don’t you go for a run now?”
This is because Fitness Trackers don’t actually know what you’re doing. All of them (including Apple Watch) use tiny sensors called accelerometers to make an educated guess at what you did based on the movement of your arm. Then they make a second guess (using height, weight, etc.) about how many calories you burned. There’s plenty of potential for error, especially if you‘re the sort of person who does any weight lifting.
Getting a truly accurate picture of exercise and its benefits requires measuring oxygen consumption during exercise and for up to 38 hours afterwards. You might also want to track heart rate, electromyography, and body composition. However, not much of that can fit in a watch. The Apple Watch can track heart rate but it weirdly omits the most basic of alerts — It can’t tell you when you’re outside your aerobic range during a workout.
Paging Dr. Tufte
In 1869 Charles Joseph Minard produced what we still recognize as the gold standard of information graphics, Carte figurative des pertes successives en hommes de l’Armée Française dans la campagne de Russie 1812–1813:
It explains everything about Napoleon’s disastrous failed invasion of Russia. With a glance, you know the broad strokes. Look at the retreat temperatures table and you can see the crushing losses of soldiers after defeat. Drill into the smart, incisive annotations and you can quickly answer the rest of your questions. Very talented people aspire just to equal Minard’s work today. These people do not work at Apple.
Sure, the Watch’s iconic ‘rings’ are good at displaying your progress on the watch at a glance for today. That’s arguably table stakes. You couldn’t bring a fitness-tracking smart watch to market without something like that. And, in fairness, most of what is displayed on the Watch itself is pretty good. Limited, but pretty good.
Where Apple’s graphs fall apart is in the historical fitness data on your iPhone. The user interface is pretty disappointing. As an example, let’s take a simple question and try to answer it: Was my last workout more aerobic or anaerobic? I know where my personal thresholds are and Apple Watch logs my Heart Rate, so a simple line graph will easily answer this question.
First, however, we have to figure out where to look. Apple spreads fitness data across two different apps: Health and Activity. The latter seems like a natural fit for this — it’s the Lord of (the Watch’s) Rings! But it just has averages and totals for specific days and workouts. My graph of Active Calories shown above was produced by manually entering totals into a spreadsheet. The Activity app itself has no trending and few graphs:
But maybe that’s not its job? Maybe it’s more of a historical scoreboard, storing snapshots in time. Let’s head over to the Health app and try that app instead. This is supposed to be your generalized health data repository. It can receive from multiple sources and even export into clinical medical records. Unfortunately, its UI doesn’t measure up to its API: Your only options are the dashboard, a single-graph page, or a list of raw data:
You can’t even pick which date it graphs, only the time ranges of Day, Week, Month, or Year. Tapping should zoom to the time range selected, but it just takes you into the raw data for today. And the smallest resolution available is 1 hour increments. It’s all basically useless for answering our question. Technically, I could manually enter 600+ raw data points into a spreadsheet such as Numbers (iOS or Mac) and produce my own graph but let’s be honest: No one, including me, is going to do that.
Some might note that Copy/Paste didn’t exist in iOS 1.0 and counsel patience. Others might suggest that I learn Swift and write my own app to answer the question. But these rejoinders miss my point — It’s not about any one question, it’s about visualizing the user’s data in a useful way. Apple’s designers are famous for working more than skin deep, for producing things that are useful as well. But to my great disappointment, they waved that off in the Health app.
Good Enough for Now
One thing the Watch does measure accurately is ‘Minutes of Exercise’. Defined as time spent at or above the effort of a brisk walk, I haven’t been able to trick the sensors (yet). Not coincidentally, Apple’s built-in goal of 30 minutes/day is the recommended amount.
Recommended, that is, for keeping you alive. Heart disease is the obvious target — It’s the #1 killer in the world. But this gets more interesting when you examine the top 10 causes of death from the CDC. Daily exercise can reduce your chances of dying from Heart Disease, Stroke, Chronic Lower Respiratory Disease, Diabetes, Suicide, and even Cancer.
This last stat, all by itself, is the great hope of public health we talked about in the beginning. For reasons we’ll get into next time, Apple Watch is unlikely to catch the iPhone. But wearables and smartwatches are coming. No matter what wins the market, if Apple can make ‘30 minutes of exercise per day’ as well adopted as tablet computing or MP3 players, then they’ll exceed ‘denting the universe’ — They’ll save millions of lives.
That potential is super exciting and chasing the Megacalorie has even helped me to lose a few pounds. But the Watch still leaves something to be desired as a general fitness tracker.
There are the measurement quirks discussed above. There’s the awful data visualization. And then there’s this physical design issue: I have to wear mine upside down (Crown: Left) because simple exercises (e.g. pushups) press the buttons on every rep. Apple has some good ideas here, but clearly there are zero fitness nerds giving Jony Ive design feedback. So you might say I’m a bit conflicted.
Luckily, one of our greatest national treasures summed up this feeling almost 30 years ago:
You’re sort of everything I’ve ever wanted
You’re not perfect, but I [like] you anyhow
You’re the [smartwatch] that I’ve always dreamed of
Well, not really… but you’re good enough for now
Next Time: Your intrepid reviewer recalls the G4 Cube, and tries his hand at prognostication. What does the future hold for the Apple Watch and for wearables in general?