I’ve been using activity trackers as part of “fighting the battle of the bulge” which has turned into a seemingly never-ending battle for better data as well. Because of that I’m still posting this in the “tech” portion of my site – and will use it as a springboard for the “health” sub-site that I will be starting this year. That will become clear if you manage to read all the way to the bottom of this post.

Polar Plotting

I’ve been using a Polar V800 (with the H7 heart rate monitor) for activity tracking for a few years. Before I started down this road I searched far and wide for a device that would capture honest heart rate data – but I also had ambitions beyond that. I wanted to include pool swimming as part of my workout repertoire, so capture of HR data during swims was part of the picture.

There were some false starts along this path, namely the Omate TrueSmart “smart watch”. This Kickstarter project promised to be a fully-fledged Android companion device while also being swim-proof, and able to capture swim metrics as well. However the device was not even water-proof, let alone able to withstand the pressures involved with motions like pushing off of the pool wall (or diving/jumping into the pull, etc)

I looked at Garmin and Polar devices, among those from other manufacturers, and came around to the Polar V800 plus the H7 HRM strap. I chose this because the H7 could (theoretically) communicate with the V800 through the water. For those of you that don’t know – Bluetooth doesn’t work under water. The frequencies used by BLE get swamped out by the density of H2O. However, Polar invented a much lower-frequency method for wireless data transmission, dating back to the 1970s. They still use it today in situations like these – as the longer wavelengths can travel through water and be captured by the wrist-worn device.

I suppose by now you’ve noticed I’ve used “theoretically” several times now. That’s because in practice the connectivity between the H7 chest strap and the V800 is haphazard at best. See the image below.


Here you’ll see the watch “thinks” that the heart rate is stuck on 119bpm for nearly the entire event. And aside from that, there are several drops to zero in the process. After busting your lungs/legs/etc in the pool for 35 minutes this is extremely frustrating. Every once in a while you’ll see how the H7 “snaps up” and catches an actual heart rate reading – usually when I pull up from a set of laps to catch my breath. This implies that the V800/H7 pair are communicating only when I’m partially out of the water and Bluetooth makes a momentary connection. And when you check out Polar’s FAQ, you’ll read how they make disclaimers about connectivity through water due to varying factors, such as salinity of the water, etc. So, in essence all bets are off. Well, I’m not the betting sort – so it’s time to find something that will actually capture data that I can use for analysis.

Enter the Garmin fenix 3 HR

For those that don’t know – switching from Polar to Garmin is like switching from Chevy to Ford, or from Coke to Pepsi, or from Democrat to… ok, not quite like that.

Setting aside the fact that I initially chose Polar over Garmin, I always liked the Garmin wrist worn devices. Since Garmin is more associated with GPS tracking they always seemed to have more of an “outdoors-y” vibe to them – and I like that. The fenix series was especially enticing, because it looked like a nice all-day-wearer, and not just an exercise-only device (such as the Forerunner series). What’s more, the more recent fenix 3 HR has an optical HRM sensor on the bottom panel. This means that it not only captures heart rate data during above-water workouts, but it also “checks in” on your heart rate throughout the day. This is a more accurate measure of activity levels (including sleep) than motion-only devices.

And most importantly, if you wear the device correctly it can collect very accurate data.

GarminHRM_zonesThis is much more my speed. In this example, I was doing an interval-styled stationary cycle session. I had read how certain activities would cause the wrist HR sensor to “drop” values before picking up accurate readings after the wearer had changed positions. I pulled a more detailed view of the plot to see if I could see any obvious drops.

GarminHRMThis is a pretty good sign, as the “mountain” shapes in the values represent intervals I was triggering on the bike (essentially bumping up resistance on the pedals for one minute out of every five). That said, one early spike seems out of place, as the “mountain range” tends to build steadily over the workout. That said, I checked my heart rate on the bike sensors (large stainless plates on the handles) and the heart rate measure there was always within a beat or two of what I was seeing on the watch. I’ve never seen over-shoot happen, unless it’s a *very* early reading before it gets a chance to build up a observation history and perform error correction.

Next up: Getting in the water

Beyond general daily-wear and in-the-gym capture from the wrist, there’s the issue of getting valuable effort data from my indoor pool sessions. The Polar H7 was a good chest strap, but it has its limitations. Aside from the in-water challenges, I noticed some drop-off in accuracy as the chest strap took more wear. This may be a general condition of any chest strap, as the electrical impulses its designed to pick up are subtle at best. But at the same time, if they make the product and put it on the market, I expect it to work. It seems odd, but I’m old-fashioned like that.

Enter the Garmin TRI and SWIM HRM straps. These devices step around the issue of transmitting data through the water by simply capturing it locally on the chest-worn device, and then syncing with the watch after the swim session. I decided to buy the kit that includes both the triathlon and swim models. Given the amount of wear I put on these things, I decided that two (which come at a 20% discount over buying just one unit) would last longer – as I plan to track workouts both above and below the water.

More Better Data

As far as this portion of my site is concerned, it all boils down to one thing – data. I have to admit, it’s motivating to me that I have this kind of capture “hounding me” to perform better than the last outing. I also like to see the trend over time – to give me some idea that even small amounts of effort really does matter.

So, going forward, I’ll be setting up a “health” sub-site here to display and discuss the data I collect over the next few months. I hope to build some hands-free data collection methods, and might even publish them on GitHub if they prove to be useful enough that others might want to take advantage of this for their own tracking purposes. But for now, I’ll stick with my own “n-of-1” analysis and work toward more and more consistent data.