Polar Grit X: wrist-based HR

Generally speaking, chest HR monitors are known to be more accurate than wrist HR sensors. Recent innovations from Polar and other top end watches suggest the gap in accuracy is being closed, making it much easier to track HR with every activity using a wrist sensor.

I've done plenty of manual checks of my resting heart rate, and can see the Polar Grit X with it's wrist hear rate sensor seems accurate to within a beat or so, assuming my manual counting is accurate!

Part 1

Taking the test a step further was to go through an activity and compare the Polar Grit HR sensor readings to a chest HR monitor, which plenty of devices use. We therefore recorded a cycling activity whilst connected to both the Polar Grit and the Garmin Edge 1000 (a bicycle computer with a chest HR monitor) and compared the HR and GPS readings. 

The first test activity was a ride, a gently undulating 19-ish mile route, which didn't raise the heart rate too much - roughly 110 and 130bpm or zone 2 in this case. In short, the results give a very positive impression from this one activity; the Polar Grit X wrist HR performed comparably and the GPS measurements were also very similar between the two devices.

Figure 1 and 2 are screenshots of the activity taken from Polar Flow and Garmin Connect.


Figure 1 - Heart rate comparison: Garmin Edge 1000 (top) vs Polar Grit (bottom)


Figure 2 - GPS comparison: Garmin Edge 1000 (top) and Polar Grit (bottom)

From looking over the plots in Figure 1 and Figure 2, it can be seen how similar the recordings across the two devices were. The speed measured via GPS is very similar in both cases i.e. the graphs have very similar humps and troughs. The Polar Grit closed out at 18.71 miles, whilst the Garmin clocked 18.63 (they are within 0.04% of each other!).

More interesting though, and the bigger technical challenge at present, is the HR recording from the wrist which the Polar Grit is performing. The plots of the HR readings coming from the devices are very similar here in terms of the plot shape and intricacy. The overall average HR measured comes out almost identical too, which at 115bpm for the Garmin and 116bpm for the Polar Grit, seems very consistent.

Assuming the accuracy of the Garmin Edge chest HR monitor, then this all bodes well for the Polar Grit X and general progress in the field of wrist-based HR monitoring.

Part 2

The next text was a 30 minute slow jog in HR zone 1, again using the Garmin Edge 1000 bicycle computer and its accompanying chest HR monitor for comparison with the Polar Grit. Figure 3 shows the HR readings for comparison (taken from the respective Garmin Connect and Polar Flow platforms, hence the different presentation of the graphs).

Figure 3 - Heart rate comparison: Garmin Edge 1000 (top) vs Polar Grit (bottom)


Figure 3 shows very comparable results between the Garmin Edge chest HR readings and the Polar Grit X wrist HR readings. Both hover around 105bpm, with pretty similar spikes along the way. The average HR was measured at 102bpm by the Polar Grit X and 100bpm by the Garmin Edge, which again suggests reasonable relative accuracy of the wrist HR monitor built into the Polar Grit.

Probably what's left is a harder effort involving intervals to test the wrist HR sensor can handle the higher HR rates and the ups and downs.