Two Ways That Tracking Your Training Data Can Boost Your Performance

As a young runner in the 1990s, I spent a ton of effort collecting and interpreting fitness metrics data. I would check my pulse every morning as soon as I woke up, then again ten seconds after getting out of bed. The difference between those two numbers, I had read, was a sensitive early warning sign of overtraining. I plotted my weekly mileage in a Lotus 1-2-3 spreadsheet, and scrutinized the peaks and troughs and trailing averages. I faithfully memorized and recorded every split of every interval workout.

These days, of course, we’ve shifted from scarce and laboriously collected data to an abundance that I could hardly have imagined—which, for an obsessive enthusiast like me, is a mixed blessing. There’s a seemingly endless array of devices that offer you real-time feedback to optimize your training and performance.

Over the years, I’ve written lots about heart-rate variability and continuous glucose monitors and muscle oxygen sensors and so on. I’ve also pondered the relative merits of subjective versus objective metrics, and descriptive versus prescriptive use of data. Overall, I’m still conflicted about it. Data, like Homer Simpson’s paean to alcohol, seems like both the cause of and solution to a lot of problems for endurance athletes. As a result, I’m always interested in new ways of thinking about our relationship with data.

How We Think About Fitness Metrics Data

A study in the European Journal of Sport Science considers two different ways that we might benefit from what it calls “augmented feedback,” defined as information from an external source that we wouldn’t be able to obtain with our sensory system alone. The authors, led by Matteo Bugnon of the University of Fribourg in Switzerland, consider this question in the context of motor control and learning: the example they provide is being shown the precise speed of a tennis serve. Most of the data that endurance athletes use—even timing your laps on a track—falls into this category.

Augmented Feedback Improves Performance, but How?

There’s plenty of evidence from the past four decades showing that augmented feedback enhances performance. And it can work pretty much instantly.

Put up a screen showing how much explosive power you produced in a leg press, and you’ll immediately produce more power in the rep, and in some cases will show greater gains over the course of weeks of training.

The unanswered question is: How much of this improvement is a result of the information contained in the feedback? And how much is simply the result of trying harder due to some version of the Hawthorne Effect—the idea that people behave differently when they know they’re being observed?

The new study comes up with a clever way of distinguishing between the effects of motivation and information. If you give subjects the wrong information, any motivational benefits should still show up but any informational benefits should be erased. To test this idea, they used two different tests of quadriceps function: maximum strength (just produce as much force as you can while attempting to straighten your leg) and maximum power (produce as much force as possible as quickly as possible).

Following each contraction, the subjects were shown their torque (for the strength exercise) or rate of torque development (for the power exercise) for the current and previous rep, along with the percentage difference between them. But in the “incorrect information” group, the percentage was reversed: if you produced 5 percent more force, you’d be told you had produced 5 percent less force.

In the context of motor control, producing maximum strength is fairly simple, while producing maximum power requires a much more complex coordination of muscle fiber contractions. As a result, the researchers figured that the difference between correct and incorrect information would have greater importance for power than strength.

Sure enough, that’s exactly what they found. In the strength tests, subjects performed better when they received augmented feedback, regardless of whether the feedback was correct or incorrect. EMG measurements showed greater muscle activity when feedback was provided, supporting the idea that they were trying harder when they had a number to beat. In the power tests, correct feedback improved performance, but incorrect feedback made it worse, suggesting that the subjects were using the information provided by the feedback to modify how they executed the task.

This is, of course, one small study (32 subjects) of one specific pair of motor tasks. It’s hard to know how well, if at all, the results generalize to other tasks, or where any individual action like running with a GPS watch falls on the motivation-information spectrum. Still, I think this framework is an interesting way of thinking about what you expect to gain from any given data stream.

For example, I’ve always found that I run faster when someone is timing my efforts, even if that someone is myself. The effect of that time or pace feedback, at least for me, is primarily motivational. If I want to ensure that I don’t push too hard during a workout, I’ll run a time-based fartlek on an unmeasured loop to ensure that I receive no augmented feedback and have to rely instead on my internal sensations. Conversely, if I want to get the most out of myself, I’ll head to a track or familiar measured loop so that I’ll get accurate feedback.

But even if I don’t have access to a measured loop, I can still get some of the same benefits by making up my own arbitrary loop and then running repeats of it. Simply comparing my effort to the previous rep helps me to push harder. The information isn’t incorrect in the sense of Bugnon’s study, but it’s irrelevant: my time for an arbitrary loop that I’ll never revisit has no meaning other than its motivational power.

Perhaps even incorrect info can help in the real world. I vividly remember my horror while watching some of my training partners do a track workout—I must have been injured or tapering—and seeing my coach miss the start of an interval. As the runners came through the first lap, he nonetheless stared at his watch and barked out splits: “66, 67, 68….” How many times had he done the same thing to me without me realizing it? And did it actually matter?

What Kinds of Fitness Metrics Data Actually Help People Train Better?

The big question these days is: Which sources of data have valuable informational content? Proponents of the Norwegian training model that’s currently in vogue might argue that a lactate meter fits the bill.

When I wore a continuous glucose monitor for a few months, there were perhaps some useful nuggets of information to be gleaned. But on a day-to-day basis, the biggest effect I noticed was a greater awareness of my food intake. I already know that a cookie isn’t a great mid-afternoon snack, but seeing the associated spike in blood sugar definitely provided a motivational spark to consider other options. I’ve made similar arguments about using heart-rate variability to guide your training: when it tells you that you need a break, it’s usually telling you something that—if you’re honest with yourself—you already knew.

Personally, I think the “truly useful information” bar is a very tough one to clear, at least in the context of endurance sports and general wellness as opposed to complex motor skills like serving a tennis ball. But Bugnon’s study is a reminder that motivation has value too, which can translate into better performance. That’s a good reason not to be too dismissive when someone insists their newest wearable gadget really helps them.

It’s also a good reason not to worry about what you’re missing, because you might already be getting the same benefits from your old gadget.

For more Sweat Science, join me on Twitter and Facebook, sign up for the email newsletter, and check out my book Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance.