Building A Better Runner With AI And Smart Shoes

Building A Better Runner With AI And Smart Shoes

Recently, I ran the second annual Disney Star Wars Dark Side Half Marathon. It was 13.1 miles of unparalleled contentment, as I got to satisfy my running nerd and Star Wars geek at the same time over the course of several hours. The only problem was motivating myself to train for the run in the first place. In Florida, April is very much “end of season” and, having run several races already since things got rolling back in October, I wasn’t all that interested in training hard.

So I did what any good tech head would do. I threw technology at the problem. Altra shoes, newcomer Vi, and performance gear giant Garmin all provided me with gear to see if I could hack my way to a better half marathon PR. So did it work? Well I definitely was sweatier at the end of it all.

Altra Torin IQ Smart Shoe

For a while, when I got ready to retire a pair of running shoes, I would take a picture of the soles. It was initially more of a philosophical thing “See how much effort has gone into destroying these shoes” and less practical. But as I’ve run more, I’ve found that there is a lot of utility in those old pictures. They don’t just show how much I’ve run; but how I’ve run. Is one part of the sole wearing out faster than the rest? Is there one area in particular that wears quicker than the sole around it? Am I striking evenly between my feet or do I favor one foot over the other?

Useful questions; but impossible to answer until you get home, take off your shoes, and flip them over. And even then, you can’t really see a pattern emerge until you’re weeks or even months into a pair of shoes.

The Altra Torin IQ Smart Shoe changes that by embedding impact sensors into the footbed of a zero drop, wide toe box running shoe. The sensors send feedback in real time to a companion app, giving you immediate feedback on your cadence, contact time, and “landing zone.”

The Landing Zone metric was what really interested me – I’m a toe striker rather than a mid-foot runner. The Altra IQ app backed me up on that assessment. Those runs at the end of the week, before my long weekend training run, were perfect Platonic examples of my ideal stride – nearly 100 toe and forefoot. But after a few runs with the Altras, I could see that, the more tired I was on a run, the more likely I was to shift my foot strike to a more mid-foot stride. It’s something I would never see just looking at the soles of my shoes and helped provide deeper analysis of my training runs.

The companion app is supposed to offer real-time feedback; but I was never able to get the audio configured properly. It’s just as well, because the GPS was rather unreliable. The app would still log the proper amount of time for my run; but wouldn’t log any mileage. It’s something the team will need to look hard at for future updates. After all, not everyone runs with multiple logging devices on runs (which, incidentally, may have been part of my problem).

App issues aside, the shoe itself was a dream to run in. That wide toe box meant that my toes could spread naturally and that any swelling that happened as a result of long runs was easily accommodated. I haven’t been this comfortable in a shoe since I ran in five-finger Vibrams. The footbed is very cushiony, with a lot of spring under the forefoot. It makes it a bit weird to walk in; but once you’re running, you’ll really appreciate the extra support under the ball of your foot. Plus the Torin IQ doesn’t look bad at all – not like the brightly colored clown shoes I usually throw on.

The Altra Torin IQ did what I didn’t think was possible, it revealed a whole new subset of data that I hadn’t been able to reliably access previously. The data gathered by the shoe itself is invaluable in what it can show me about how previous injuries affect my running today and can keep me tuned in to how successful I am in shifting my mechanics going forward. You can pre-order your own pair for $220 at the Altra IQ website.

Vi by Lifebeam

Vi is a new digital fitness assistant from newcomer Lifebeam. She’s packed into a neckband with some high quality harmon/kardon headphones and equipped with an in-ear heart rate monitor. I say “she” because once you load up the companion app and get going, there’s no way of thinking of Vi as anything but a gendered entity. She even jokes on start up about being a robot, then laughs it off and continues on with her introduction. The interaction is immediately engaging as Vi walks you through a series of questions to determine what your goals are and how often you’d like to run.

The program starts you off slowly – as with all the running tech I tested out, you need to put in about an hour’s worth of run data before you’ll get personalized results. As you get more comfortable running with Vi, she’ll add new twists – like improving your cadence and helping you improve awareness of your “running zones” so that you know when you’re under- or over-exerting yourself.

One of my favorite features of running with Vi was one of the most subtle. Every mile, about a tenth of a mile out, you’ll start hearing a beacon, this is your signal that a milestone is coming up. It’s surprisingly motivating, especially on those harder miles where you want to pack it in early. Once you chase down the beacon, you’ll be rewarded with an audible flourish and Vi will tell you how your doing and what your stats are for the previous mile (if you’ve run with Runkeeper, it’s like that, but less jarring).

Speaking of Runkeeper, one of the questions I had throughout my time with Vi was what I was getting from her that I couldn’t get from my phone, an app, and a pair of Bluetooth headphones (like the Razer Hammerhead). The answer is mixed. In-ear HRM is a valuable metric and I greatly appreciate not having to run with a chest strap in the Florida heat. Updates on cadence and pace are very useful for consistent performance, when they work, but occasional GPS issues kept it from being reliable. And when the pace indicator did work, Vi didn’t do much more than let me know that my pace was all over the place and that I should “try to be more consistent.” Which is a frustrating thing to hear when you feel like you already are being consistent. If Vi is going to be an in-ear coach, I need her to, well, coach me. Tell me what to work on for consistency. Let me know when I’m rabbiting my first mile and tell me to slow it down or I might fatigue and screw up my pace later. Right now, Vi’s interactions are limited and you’ll hear a lot of the same prompts and information over and over again. She’s less of a running AI and more of a chatty running app.

Lifebeam also has some work to do when it comes to filtering out road noise. If you run next to a busy road or anywhere that’s windy, Vi’s mic becomes unreliable at best, unusable at worst. This robs Vi of her ability to respond to your real-time prompts. I’m hoping it can be handled with software updates, because any change to the Vi hardware would be criminal. The neckband is durable, sweatproof, flexible, and has great touches like magnetic tips to store the touch-sensitive earbuds. Battery life is solid, lasting through a week of training runs before needing a recharge.

Vi’s got personality; but it left me wanting more interaction. I was hoping for a coach that could map out an ideal training plan for me and help me in real-time on my mechanics. Instead I got a personal cheerleader with bad hearing. Which is fun; but is it worth $250? The Vi team is passionate about the product. I have hope that as Vi gets software updates, she will evolve into the product I want her to be; but she’s not there yet. You can find Vi on the Lifebeam website.

But wait, there’s more! While the Altra IQ and Vi were useful, they didn’t give me exactly what I wanted while chasing my PR. However, there was one piece of tech that rose above all others. More on that soon.