Spikes and drops in power


#1

Every time there is an interval change the Trainerroad app on my Android tablet shows either a spike or a drop in the power number. I use a dumb trainer and a 4iiii Precision power meter and pedaling at a steady cadence without changing gear.

Is this a bug in the software or is it the tablet HW not keeping up?


#2

Can you share an example ride?

You may be better off contacting support@trainerroad.com depending on the specifics of this issue.


#3

I have the same situation as you with spikes/drops with a fluid trainer(it’s not dumb, it’s just as cycleops made it) and a 4iiii precision left side. I let customer service know about it a while back and got a “must be something you’re doing”(paraphrasing) which was sort of hard to argue. Though from what it looks like spike wise, my guess is that the smoothing resets at the beginning of each interval so if you have 5 or 7 or whatever second smoothing if the interval changes somewhere in between said smoothing it’ll act like 1 second smoothing at the beginning of the interval, basically showing you what your power is at that very moment which of course is not even. Remember the graph is always recording at 1 second smoothing and I found my spikes seemed to correspond with that. It does suck because occasionally I have to chase the average to get it back on track at the beginning of an interval if the spike reads low for whatever reason. Probably doesn’t help much but that’s my thought. A little frustrating but I just ride through it. If there’s a real fix though I’d love to have it just for piece of mind.


#4

I have a 4iiii left side PM , ride rollers and sometimes use my Android phone (Pixel 2XL) for TR workouts and I have not experienced that problem.


#5

You may be on to something there. I guess this is exactly what you would see if the smoothing starts over on zero at the start if each interval. But then the fix should also be easy for TrainerRoad. Why restart the smoothing? As a matter of fact I cannot see these spikes and drops in the data afterwards. They are only visible in the power numbers when riding.


#6

I see the spikes in the graph occasionally. And I agree that it’s hard to provide evidence. The only way I figure is rather than a screen shot it would have to be a screen video of the workout being executed. That’s the only way I figure we could show things jumping around at the beginning of intervals. Like in the middle of work intervals for rampy sweet spot workouts like donner and darwin, having things jump around for no reason when transitioning from 2 minutes to 4 minutes or the like. I should add that I’m on iOS and not android.


#7

I too have 4iiii left sided pm and have also noticed this but simply put it down to a small bug with the smoothing. I just presumed it resets when it hits the next interval.


#8

When I first encountered this issue it was before ride analytics was out. But now that I can drill down to the second by second it becomes a bit more clear the difference between the smoothed power that’s displayed versus what we’re actually putting out that’s graphed below. I did Elephants +4 this past Friday and remember the beginning of the 4th interval being somewhat behind average wise for the first 15-20 second or so when the power that was displayed was telling me I was more or less on target and cadence was steady. When I drilled down I found every other second I was well below target while in between those seconds I was on target which was what was getting displayed so it felt like I had to chase back on to get the average power back on track. Basically it seemed like my 5 second smoothing was keeping me from seeing the power I was really putting out for that first 20 seconds or so. I think this is more or less what is happening at the beginning of some intervals for us. Only that we’re seeing the 1 second power at that moment rather than the smoothed power number. I think in order for trainer road to see this in action would be to drop one of our completed workouts into their development software and play it back like what we see when we’re executing it to show how the smoothing is displayed during the workout versus how it’s recorded. In the analytics it doesn’t give what the average was at any given moment which is what tends to show these spikes at the beginning of intervals. Sorry for going crazy on this but it’s something I’ve been observing for quite some time and to hear about it happening to someone else is a relief in a way. This might not be a bug or an actual problem, it may just be the nature of the software and how power is displayed versus recorded.


#9

Just to add to this, I am running a pair of Vector 2 pedals on a KK with my Android phone, and see the same thing happening at the start of new intervals. Kind of surprised more people have not seen this issue, as it seems to be independent of OS/ power source.


#10

I don’t think it’s smoothing-related because the graph itself isn’t smoothed as far as I know. The power data recorded is just raw power coming from your power meter (unless you have a smart trainer that does smoothing itself). In endless discussions on the vector 3 forums Garmin has claimed that spikes can be a result of the receiver not getting the data for a second or two and then playing catch-up. No idea if that’s true but if it is, I suspect that TR’s interval changes cause a slight delay in processing power data.

But to be honest, I also wouldn’t preclude the possibility that my brain is playing tricks on me and my legs are actually anticipating the interval change.


#11

It’s definitely unwanted “cognitive load” during the Ramp Test!