Like most runners, Lucy (who works in our marketing team) uses a sports wearable to help her track her pace and record her data to examine after the race.

And like most runners, she knows that her wearable is far from accurate.

As runner Ian Williams puts it:

“Learning not to overly rely on your GPS when you are hoping to sneak a PB is an uncomfortable rite of passage for runners. I have done it myself and seen my goal time come and go, while the finish is still an agonising hundred yards away.”

In fact the open secret in fitness-tracker world is the big trade-off that athletes have to make between an accurate device, and one with an adequate battery life. We’re out to break that trade-off. We firmly believe that you should be able to have your cake and eat it (especially if you’re a runner)

So when Lucy tackled the Cambridge Half Marathon with hopes of running a sub 2hour time, we saw the perfect opportunity to pit our D-Tail technology against her regular wearables, to show just how much runners are forced to compromise with today’s wearables.

We strapped a standard Google Pixel phone running Android Fused Location, alongside our D-Tail tracker to Lucy’s arm, then waved her off.

Here’s how the different apps saw her run:

Why does it matter?

Apart from making Lucy look like she was slightly drunk, swam across the river Cam a number of times and charged through several walls and gardens, the inaccurate timing and positioning from the regular tracker hampered her ability to run her optimum race.

We spoke to marathon runner, Jordan Foster or ‘Project Marathon Girl’ as she’s affectionately known as on the UK running scene,. about how this affects athletes

Better data = better training plans

In order to train using pace and heart rate, you need a true representation of your speed and distance. Variances that current wearables experience can cause huge differences in end performance and a poor picture of how your fitness translates into speed. For athletes like Jordan, this data can mean the difference between winning and losing, and it’s especially bad in cities:

“If you’re striving for a PB, it can really throw you if your pace is all over the place. This ALWAYS happens with events in London, especially if you’re running through Canary Wharf”

Truth is everything

Lucy’s Garmin Forerunner recorded 20.91km, whereas fellow runner Crispin’s Fossil Sports Watch recorded 19km via Google FIT. That’s a difference of 27 seconds per kilometre, for the exact same route! Not only that, but at an elite level, being sure about who has even won a race can be questionable.

As Jordan says, this is especially irksome if you’re even slightly competitive on Strava

“It’s a bit of a first world problem, but it’s really annoying when you get a PB at a certain distance but when you upload your run it comes up short due to GPS.”

Value for money

If you’re spending £150 or more on a sports watch, it should at least deliver on its promise of accurate tracking and corresponding ability to train. Unfortunately that’s just not the case right now.

How do they work, and what’s the D-Tail difference?

A lot of GPS wearables will be checking satellites every 5 seconds or so, taking a reading and then joining the dots between them. This can cause two problems – firstly no-one runs in a perfectly straight line in those moments between readings, and more worryingly, those readings can be highly inaccurate, especially in built up areas where satellite signals get bounced off buildings.

Some wearables and mobile apps will use a combination of GPS and sensor fusion from accelerometers and gyroscopes to make the best determination of location and thus pace, but even then, the movement of the runner can confuse the compass in the device, creating a false reading. With traditional methods of combining GPS and sensor data there can be a chicken-and-egg problem – the compass heading and the GPS data don’t agree – so which one is wrong and which one is right? With only 2 systems in conflict you cannot answer this question.

Our D-Tail technology uses a machine learning-based human motion modelling to build a centimetre-accurate view of movement through space, enabling precise measurement of location, pace and even gait, and it is the human motion modelling technology that can provide that critical third piece of information to help determine whether the compass data or the GPS data are in error.

Critically, it does this at an extremely low energy-requirement, removing the trade-off between accuracy and battery life.

We’re constantly trialling and improving our products, if you’re interested in how you can integrate it into your device or app – or what a difference it can make to your data, get in touch.