Canadian Cycling Magazine
FOR TORONTO’S XERT, THE FUTURE IS IN AN OLDER TECHNOLOGY
You’re having a good workout. The intervals actually seem easier than they usually do. Traditionally, you’d either be the dilgent athlete and stick to the target power the workout or coach has prescribed. Or, you might amp up the watts on those intervals. Either way, you wouldn’t be able to get a picture of your actual fitness until later, once you did a fresh FTP test.
Armando Mastracci wanted to build training software that could tell you your fitness without the need for an FTP test. The Toronto-based computer science engineer wondered if he could figure out a rider’s fitness from regular ride data in an automated way. His experiments in early summer of 2015 revealed it could be done. Roughly a year later, he released Xert. Once you upload your ride data with power metrics (power data is essential), the software’s algorithm can tell you your FTP. In fact, with the Xert Garmin Connect IQ app, you can see just how long you can hold a specific effort, in real time. Have you cranked it up to 300 watts? Well, the computer says you can only do that for three minutes, so make it count. Wait. Did you just hold that for four minutes? Great. The software can then adjust your fitness signature accordingly. The software can also recommend workouts based on your fitness, your goals and your level of fatigue. It can even predict your future capabilities.
So, with Xert, will coaching by algorithm replace human coaches? “It will put bad coaches out of business,” says Jamie Sprules, Xert’s senior vice-president of operations. He stresses that Xert doesn’t replace the art of coaching: knowing an athlete from his or her work schedule and family life to past injuries.
As for the future of Xert, it’s going for the heart. Currently, power is the tool for calculating your fitness signature. Heart-rate data, however, was behind a lot of the development of the software. Then it was left behind because heart rate is more complicated to work with than power. “Heart-rate data integrates everything that’s happening to you,” says, Armando Mastracci. “It’s all encoded there. But you have to find a way to decode it. What’s going on? Why is my heart rate increasing? Because I’m going too hard? Because my threshold is too low? Or because I’m overheating? Or I just ate something? All of these things leave a footprint on the heart rate. We have ways to pull that information out. It’s not easy.”—