Running in the Era of 'Big Data'; What Information Do You Need?
Back in the day (as a high school and university athlete), I met my coach down at the track, or out at the golf course or trailhead, and we did workouts (often with the coach holding the stopwatch, calling out split times, just like in the movies), talked about them afterwards, and made plans for the next session. Later, as a club athlete, I was a “runner/coach” and decided the sessions for my training group.
This was a time before Strava, so when I recorded workout data (which was very rarely), I did so with a pen, in a notebook. Also, this was before GPS watches, so I used my trusty Casio G-Shock or Timex Ironman. Those watches were indestructible, except that you had to replace the band every couple of years when it broke.
I can hear you thinking, "Okay, boomer". Well, yeah, but believe it or not, the World Wide Web has been around for only around 30 years. Some of us have been running longer than that. And I am on Strava today and for years I have worn a GPS watch.
What data can you get?
Last September, Apple introduced its Watch Series 6 with the slogan “The future of health is on your wrist.”
But although the Apple Watch is certainly attractive, it's not the GPS watch for most serious endurance athletes, because it doesn't have the battery life. Apple claims the Watch Series 6 can deliver seven hours of battery life in GPS mode, but unless you're Jim Walmsley, that's not going to be enough to get you to the finish line of a 100km ultramarathon.
The big names in the GPS-for-endurance-sport game are Polar, Suunto, Garmin and Coros, all of which produce a high-quality watch. In fact, I have owned watches from all those companies, albeit over a 25-year period, starting with Polar, and moving to Suunto, then Garmin, and finally Coros.
The watches themselves (the hardware) provide pretty much the same functionality, depending on how much you're willing to pay (i.e. all the manufacturers have product lines that offer functionality – and battery life – that increases with price). What's interesting and useful to athletes is the data the watches can collect and present to you either on your wrist, in an off-watch app, or exported to a third-party app such as Strava, RunKeeper, MapMyRun or one of the shoe company apps such as Nike+ Run Club or adidas Runtastic.
So what data do these watches deliver? Well, as soon as you press "stop" to finish your workout you can know your time (and most watches incorporate an "auto-pause" feature that means you no longer have to press "stop" when you hit a traffic light or need to stop to take a selfie, LOL), pace, distance, heart rate, and elevation gain.
Once you've exported your data to an app, you can know much, much more. You can review your pace in kilometer (or mile) increments, which can show you if you perhaps started too quickly and paid a price toward the end. You can see your stride cadence, which also can show you if you may have started too quickly, but in addition can show if you may need to improve your leg strength (e.g. most runners slow down at the end of a marathon not because they're winded, but because their muscles are not working as efficiently as at the start).
Other data features include power and intensity measurements (estimates, really) and an assessment of how much rest you (may) need before your next workout. These features are nowhere near as accurate as laboratory tests, but they're certainly easier.
Finally, a feature that is not so useful in Singapore, but might actually save your life in wilder environs, is routefinding. Many GPS watches are able to download GPX files that show offroad routes plotted either by yourself or others. This is a great way to explore new places (e.g. if you’re in Rauma, Norway for the first time, you might check Kilian Jornet’s Strava, see what routes he favours, download one to your watch, and grab yourself some vert!) without getting lost; your watch can guide you along the route.
The gamification of fitness data
Everyone's physiology is different, of course, and the real value of fitness data comes after you've recorded enough of it within your chosen fitness application to establish benchmarks that are relevant to you, rather than a generic "male, aged 42, weighing 72 kilograms and 1.83 centimeters tall".
The fitness apps (and watches) know this, and they're happy to help you "gamify" your training. Strava famously lets users demarcate "segments" (e.g. Rifle Range Road, MacRitchie Reservoir Loop, Bedok Reservoir, etc.), which allows you to measure your progress against yourself (and your friends and enemies!), but also can turn every workout into a competition (if only against yourself) and lead to training overload and injury.
It can be fun to see how you rank on various segments (especially if you rank first), but a big problem in Singapore for Strava runners is data pollution. It's no fun when you record a fast run and see that on the list ahead of you are half a dozen "runners" who were obviously riding bicycles or scooters, or were even in cars. Strava understands the issue, and there is a reporting and take down mechanism, but who wants to spend their life reporting inaccurate Strava segment results? Not me.
At the other end of the spectrum, a few years ago I uploaded the data from a trail run I completed in Northern California and Strava informed me that I was a "Local Legend" and had recorded the "fastest time this year" on the trail. Feeling pretty satisfied with myself, I clicked through to the list of results and I found that I had been the only Strava user to run that trail so far that year (it was August). Not that impressive, then.
Health and wellness
The final piece of the running data puzzle brings us back to Apple and its health and wellness focus. Many running watches and apps monitor your heart rate, and the Apple Watch has sensors that can detect atrial fibrillation, an irregular heartbeat that can be deadly if not properly managed. Many watches now incorporate pulse oximeters that can monitor your blood oxygen levels. As with power and intensity measurements, the accuracy of this wrist-collected data is pretty good.
We're moving closer, it seems, to Dr. McCoy's tricorder ("Star Trek" reference, for those who had sheltered childhoods), except that we won't need Dr. McCoy; we'll be monitoring our health ourselves, from our wrists. The problem, of course, is that few of us have completed medical school and can confidently separate the signal from the noise. A study by the Mayo Clinic published in September 2020 found that only 11.4 percent of people who went to the hospital after their Apple Watch detected an irregular pulse ended up with a “clinically actionable” diagnosis of a heart issue.
The same is true for running data. Do you need a pulse oximeter in your watch? If you're climbing Mount Everest, I'd say it's a good idea. If you're running 10Ks in East Coast Park and are otherwise healthy, you probably don't.
It's easy to go down the rabbit hole of evermore data (everyone needs a hobby!), but at a certain point you probably have everything you need. Thinking back 25 years, I had everything I needed with the stopwatch function on my Casio G-Shock and Timex Ironman. And so did (pick your distance hero) Sebastian Coe, Haile Gebrselassie, Greta Waitz, Joan Benoit and Yiannis Kouros.
If you're not currently a GPS watch user you may be wondering about the accuracy of GPS tracking. There is a lot of comparative data on the Internet (I like the DC Rainmaker website for really deep dives into product capabilities and comparisons), but in my view the big four GPS watch makers – Polar, Suunto, Garmin and Coros, and let's throw Apple in there as well – do a fine job of showing you where you ran and how fast.
Some manufacturers go farther – for example, my Coros Apex includes a pretty cool "Track Mode" that significantly increases the accuracy of recorded track workouts by ‘snapping’ to the track and even a specific lane – but I'd say this is "nice to have" rather than "have to have".
One area in which not all watches are equal is battery life. Earlier I noted that the Apple Watch Series 6 can deliver seven hours of battery life in GPS mode. If you're running 10Ks, that's fine. If you're running marathons, and remember to charge your watch before the start, that's fine.
But if you're a member of the fast-growing ultramarathon community, battery life is probably the single most important feature you want to consider in buying a GPS watch. There are a number of factors that affect battery life (most manufacturers have a long-life mode that records fewer GPS data points), and with most manufacturers, the more you pay, the more battery life you get.
Garmin's top-of-the-line Fenix 6 (I had the previous generation Fenix 5X Plus) claims up to 25 hours in full GPS mode. Garmin's Forerunner 945, more of a pure running watch, claims up to 36 hours.
My mid-range 46mm Coros Apex claims up to 35 hours in GPS mode, and the top-of-the-line Coros Vertix claims an eye-popping 60 hours.
Both the Garmin and Coros watches offer much longer battery life if you go into long-life mode. My Coros Apex will last up to 100 hours, and in use as my daily wristwatch, I recharge it only around once a month. That's a significant (and welcome) difference to my old Garmin, but of course nowhere near as good as my much older non-GPS Casio, which would need a new lithium battery only once a year (or two years!).
Data is only a tool
Back in the day before GPS watches, heart rate monitors and power meters, we had stopwatches and surveyors’ wheels (for measuring courses). After each workout we could look at our times and ask ourselves how we felt. Today it’s easy to immerse yourself in metrics (and obsess about them), but it’s important to keep the data in perspective, to ask yourself how you felt afterwards (and during).
If you struggled through a workout that yielded indifferent performance data, it’s probably time to take a recovery day. Or two. And yes, your watch may suggest this to you as well, but remember, your watch doesn’t know how you actually feel. Not yet anyway.