When the COROS POD 2 was launched last month, we asked 2:39 marathon runner Derek Li if he would try it out, and write down his thoughts. He also borrowed a COROS PACE 2 GPS watch, in order to connect to the POD 2. Below is Derek's review for RoadTrailRun, with thanks for the permission to reprint!
I was very fortunate to get to test the COROS PACE 2 watch as well as the newly launched POD 2, courtesy of the Red Dot Running Company in Singapore.
POD stands for Performance Optimization Device, and the POD 2 aims to do just that. The expectation is that this will supplement your traditional watch GPS to improve tracking metrics. We all know that GPS has its limitations, especially if you live in an urban area where there are a lot of high-rise buildings or if you are running through sheltered sections like underpasses or sheltered walkways. The POD 2 aims to add a second layer of information for the watch to parse such that you have a little more buffer against erratic losses in GPS signalling.
Another key feature of the POD 2 is what COROS coined “Effort Pace”. This is basically an adjusted pace based on the elevation profile and presumably the ambient temperature as well since there is a temperature sensor in there. The aim is to give you an extra metric to help you run at a more balanced effort on uneven terrain. The hope is also that with a footpod, one could get more instant feedback on pace changes when running. Finally, there is the consideration for indoor running. One upon a time, before GPS was in vogue, all running watches relied on footpods to give estimated paces and distances. At some point, the various brands decided to use armswing and the accelerometer in the watch to do the job (especially important for treadmill running), but if you have ever tried to use the indoor function on any brand of GPS watch, you would know that the numbers thrown out are far from ideal. Hence, COROS aims to rely back on the footpod design to give us more accurate data.
For the purposes of this review, I will focus more on testing the capabilities of the POD 2.
Unboxing:
The package comes with:
- 1 x POD2
- 1 x charging cradle
- 1 x USB-C charging cable for the cradle
- 2 x shoe POD holder
- 1 x clothing POD holder
- Carrying case
Pairing the POD 2 is fairly straightforward, by adding it as an accessory through the PACE 2 watch.
The connection is made very quickly and you will see the POD2 ID visible on the watch once you are done.
Once I got that out of the way, the testing began in earnest.
Indoor Running:
My treadmill has recently decided to retire, so my treadmill data is unfortunately quite limited by what I saw was enough to convince me that the POD 2 is at least non-inferior to the other offerings.
As you can see, I still use footpods a lot for the treadmill. This is in addition to my NPE Runn Smart Treadmill Sensor.
The above is a short treadmill run where there were clear steps in the pace change, and you can see that the pace changes are crisp and nice on the graph. More importantly, the reported pace is quite consistent and precise. No issues here.
Running with Inconsistent GPS signal:
The next step involved testing whether the POD 2 could pick up the slack when GPS signalling was weak or disrupted.
There are 2 sections on my usual runs where this is prone to happening. The first is a ~600m section of sheltered walkway on my way from home to my local running track.
The run goes from left to right in the map above. The blue line represents the COROS PACE 2 with POD 2, and the red line represents the Garmin 945 without footpod. The first challenge is running under the Farrer Flyover. This is farther complicated by the presence of two pedestrian crossing on either side, that don’t turn red or green in sync so one is often stuck waiting under the flyover for up to 2 minutes. This part gets my HR up even more than the run because you never know what’s going on with the GPS lock on your watch at this point. Here, you can see the blue line still mostly adheres to the walking path, while the Garmin line tracks out onto the main road for a little bit.
After, you are running along a narrow pavement next to some buildings but it’s still exposed. The next tricky section starts where the Botanic Gardens train station is. This part is completely sheltered almost all the way until where the run terminates. Here, neither watch does a perfect job, but you can at least discern that the blue COROS line mostly adheres to the pavement, but the red Garmin line makes multiple excursions out onto the main road.
Conclusion: COROS wins round 1
The second section involves running on the Green Corridor pathway. There are 2 sections on this pathway with some extended periods under some flyovers and again often presents a challenge to the GPS signal.
The first section involves running under the road junction of Commonwealth Avenue and North Buona Vista Road. Once again, the blue line represents the COROS PACE 2 with POD 2, and the red line represents the Garmin 945 without footpod. Bear in mind this is an out and back route for me, so there are two blue lines and two red lines in each image. You can clearly see that the path plotted by the COROS is significantly smoother and more consistent in the out and back sections than the Garmin.
The second challenge on this route is where I run under the Queensway flyover. This section is considerably longer, to the point that it gets quite dark in the middle section of the flyover. Here, again the blue line is more consistent out and back, but I’d say the Garmin does a little better here and still stays relatively consistent.
Overall, COROS wins Round 2.
My final conclusion is that the COROS PACE 2 with POD 2 does very well in poor GPS conditions. That said, the final cumulative difference in distance covered is relatively small. Often <50m for a 1hour run.
Pace Change Responsiveness:
One of the intended benefits of the COROS POD 2 is for the reflected pace to be more instantaneous to overcome the perceived lag that occurs with GPS-derived pace data. To assess this, I headed to the track. The main aim was to see if I could tease out any differences in pace changes on the track. In theory it should be quite easy to demonstrate if I start both watches at the same time, and incorporate big changes in paces at fixed distance intervals. In this case, I did it at 200metre intervals.
First, to assess the stability of the plotted route by running at a steady warm-up pace. Track running mode was utilized on both watches. (blue line represents COROS PACE 2 + POD 2, red line represents Garmin 945).
Here, the speed graph is most useful, as you can see there is significant earlier spike in the pace with the COROS vs the Garmin.
Then I incorporated 200m of speed with 200m of jogging recoveries:
You can see that cadence and heart rate track quite closely for both watches, but when it comes to the speed changes, the COROS is consistently starting earlier and ending later. Why it’s ending later, I put it down to the Garmin taking fewer data samples and averaging in the slower paces of the recoveries into the time when I am still running fast, and thus interpreting that the interval is shorter than it actually is. Ideally, I would have done some manual laps to get a better read on this, but it really is quite impossible to do manual laps on two different watches without significant lag times.
All in all, I am fairly satisfied that the POD 2 does as it is advertised in that it responds faster than a pure GPS watch in terms of responding to changes in pace.
Effort Pace
The final part of testing looks at the concept and utility of Effort Pace. As mentioned earlier, the aim is to give you a sort of corrected pace after taking into account terrain and conditions. Bear in mind that all this while, COROS has its native run power calculations running in the background.
With regard to Effort Pace, there is instant Effort Pace, Lap Effort Pace and Average Effort Pace, just like the normal pace fields. I played around with a few of the fields, as you can see here.
Before I go into which fields I prefer, let’s look into the activity data to see whether Effort Pace works.
Here, I have a run where I am going at roughly 5:00-5:10/km pace on the flats, then there is a big climb and then the downhill on the other side.
The first graph shows a data point roughly on the flat section, with Effort Pace and actual pace bang on 5:00/km. Nothing out of the ordinary there.
The second graph shows a data point midway up the climb, where Effort Pace increased to 4:44/km while actual pace has stalled down to a pedestrian 6:31/km.
The third graph shows a data point on the downhill section, and Effort Pace as dropped down to 5:25/km and the actual pace is 4:21/km.
I can’t really say quantitatively if the numbers are accurate, but qualitatively speaking, I’d say it is pretty close.
I don’t have a Stryd pod to compare things to, but this is the plot for the Effort Pace vs Run Power as derived from COROS’s power algorithm:
You can see that Effort Pace and Power tend to spike and dip at more or less the same areas, but Effort Pace is a little smoother across the board. I did run this whole section at sort of a constant steady controlled effort so there should not have be any big swings in effort or power.
Is Effort Pace something you need, or is the native Power field sufficient? That really is up to you and how much help you generally need with pacing. I will say this. While it is good to have a system that can magnify small differences so you know there’s a difference, you also don’t want a metric that swings wildly too much such that it unduly prevents you from running a smooth perceived pace.
Back to my preferences on watch fields. I actually found that when I used instant Effort Pace as the main guiding field, the numbers were jumping around a lot, even on relatively flat routes. We are talking 10-20s/km swings every second. This is not a knock on the POD 2. It is simply a reflection of how our bodies produce forces at non-constant rates. Our legs are not wheels, even more so in running than in sports like cycling. The calculated instant Effort Pace is simply a reflection on when in the gait cycle the data sample happens to be taken. This should in theory be improved if we somehow used a POD 2 on each shoe but that may be a discussion for the future.
Looking back at the history of power meters in Cycling, instant power readings have been equally noisy. It is for that exact reason that cyclists commonly use 3s, 5s, 15s or even 30s rolling average power numbers for the purposes of pacing. I would like to see Effort Pace get the same treatment as an update in the future. Of course, you would lose that instant pace feedback in such cases. I think for most distance runners, the benefit of effort pacing would outweigh the pace lag, and something like a 5-10s rolling Average Effort Pace would be very useful indeed.
So to answer the original question, I still prefer to use Lap Effort Pace as the main data field for the purposes of pacing on undulating terrain.
Final Thoughts
The COROS PACE 2 is noticeably lighter than the Garmin 945, even with a regular silicone strap. I have been able to get away with charging the watch once every week at the moment as I am running relatively low miles, and the POD 2 could probably get away with charge every 1.5 weeks, though I opted to charge it before my long run for the week, just in case. I am quite happy with the GPS + POD 2 consistency for the most part, and even the little things like GPS signal capture to start the run are very fast. The POD 2 is a new product and I hope to see more rolling updates and features for this along the way.
The COROS POD 2 currently retails for S$145 and is available for pre-order at www.rdrc.sg.