You don't need to be an 'investor' to invest in Singletrack: 6 days left: 95% of target - Find out more
Okay, I'll try and keep this as concise as possible! I'm using Garmin Coach on my Epix2 to "train" for an event. So I thought I'd best calibrate it against our treadmill. Watch is on left wrist, I've never really worried about calibration before as I've been using Zwift which "talks" to the treadmill on BT. Which the Garmin doesn't hence calibration needed (I don't have the posh Garmin HRM)
Left wrist- all over the place. If I use "virtual treadmill" so Garmin powers treadmill via Zwift (Watch->Zwift->Treadmill) and set the pace at, say, 10km. It takes about 20 seconds of running (so arm movement which is how it works out how fast you are going-it's witchcraft for sure) before getting close to what I'm seeing on the treadmill/Zwift. Then it just drops to 9-8-7-6 over maybe 20 seconds before ramping up again. If I do some funky chicken arm movement I can get it closer to the 'right' speed, but it still drops off.
I've attempted to calibrate the watch as over a 6km run it was 1.5km short! No improvement so I switched it to my right wrist (on Garmin user profile, it's set to left) and now it was OVER reading (ie original calibration appeared to be spot on) and it does not exhibit that 'pulsing' I got on the left wrist.
Also noticed lately on MTB rides, it seems really slow to pick up increases in heart rate. I wonder if I swapped it to the right wrist if it'd fix that as well.
I cannot think of any reason why this should be happening. It's not like I move one arm more than the other when I'm running. I've done some googling but other than extracting myself from some deep rabbit holes and am no closer to a solution.
In short, optical heart rate sensors on watches are hit and miss, they work for some people but you can’t predict that. For any serious training I would get heart rate belt and use that at least occasionally to calibrate the feelings vs device readings.
Cadence readings and running power probably need ‘real’ running to get any sensible values since they are measuring accelerations - on a treadmill the accelerations are slightly muted which messes the input to algorithms.
I recently bought an Adidas treadmill that links to Zwift itself. Whatever speed teh treadmill is set at is the speed that Zwift sees.
Cadence and HR are from my Fenix 7 (I'll occasionally use a HR belt if I'm doing intervals as the watches don't react quick enough). I start the Fenix in "Virtual Run" mode.
When the run is finished I get the option to Save or Calibrate and Save on the Fenix, at which point I enter the Zwift distance. I've been as much as 10% out compared to Zwift but it stays reasonably accurate as long as I run at approximately the same speed each time. So, if I've done a Calibrate and Save at 10kph, and do another 10kph run the next day, the two distances will be within 1-2% of each other.
I've read that you should do some outdoor runs to help the Garmin watch get a better assesment of your stride length (obviously, it can then compare it to GPS distance).
Also noticed lately on MTB rides, it seems really slow to pick up increases in heart rate. I wonder if I swapped it to the right wrist if it’d fix that as well.
Cold weather makes wrist based optical heart rate worse.
Can Garmin watches use a foot pod? Would that help?
Thanks for the replies. To clarify:
I do have a chest based HRM just not a Garmin one. It's ANT/BT and connects to both my Epix2 and Mac running Zwift. The posh HRM (Pro Plus) offloads the cadence and speed metrics from the watch but at £130, it's mad money.
@scotroutes- yeah that's what I tried and it was when I noticed this strange issue where my speed on the treadmill didn't change but the speed recorded by the watch did. It'd point to an accelerometer issue with the watch did it not work perfectly on the other wrist! I think you're right tho, it seems 'better' on a long run at the same pace. But it's still different between arms. It's not a big issue (other than Garmin Coach refusing to accept I've done the workout properly and sending my back to the naughty step to do it again) but it is annoying.
I will defo be doing some more outdoor runs starting next week, so we'll see if that improves things.
I have a Zwift FoodPod and it seems to work well. If I use virtual run, it does pass the cadence back to the watch via Zwift. But it doesn't fix the issue.
The more I think about it, the more I'm sure the strange HRM lag I get on MTB rides isn't related. Different sensors entirely.
Ho hum, another run inside today so we'll try it on the right wrist to see if we can establish a working baseline.
Oh @scotroutes thanks for the link. I have the Zwift one as I said, I don't really want to buy another one! I forgot Garmin made one, for some reason I thought it was discontinued. It's s pain that Garmin won't pair to Zwift Footpad and - I think - vice versa. The joys of open standards!
This is a bargain for £99. Stryd is the benchmark for accurate pace measurement on a treadmill, but bear in mind that running dynamics change when your feet are fired backwards by a moving belt so you will always run "faster" indoors than outside.
You need to tell the watch which wrist you’re wearing it on, otherwise the measurements are wildly wrong.
I thought it was only needed for those ones with gesture controls.
For measurements eg steps the advice is to wear on nondominant hand since thats generally less active picking up cups of tea etc. I didnt think it applied for running etc.
For OP:
As per others I find the wrist based always lags a bit for rapid changes. I also find it generally worse for mountain biking as the arms tend to be more active and I think the wrist flexing can knock it out.
Your left vs right on the treadmill is weird though. Have you tried filming to see if your arm movements are different?
@flaperon - there's a lot going on in those footpads! I'm going to do a bit more research/testing tho before spending any cash. I read something about running being different inside and that does make sense. I'm sure my cadence is faster, I'll go back and look back at some of my summer runs to compare.
Wrist based glances is a thing on my watch, I checked it was set to left and it is. Thought that might be the culprit but apparently not. I have thought of getting the better half to film me, once I've confirmed all is good or as close to good as it's going to get on the right wrist. The idea of reviewing my wonky running technique isn't filling me with much joy tho!
Well update if anyone still cares 🙂
Ran same pace for 40 mins. Didn't change pace at all. Watch on right wrist was within 5%. Then swapped watch to left wrist, ran for a couple of km and it did its pulse/cycle thing from the start. So I guess the solution is stick watch on right wrist... although that doesn't answer the question of "why do I need to do that" #boggles