I have two power meters: (1) a Stages left crank and (2) a smart trainer. There’s a fairly large difference in readings (30-50%).
What would be the best way to determine which is the more accurate? (Ex: go to a gym and ride one - or several - of their bikes; compare with an outdoor ride; compare hr and power zones)
This is a pretty common topic with power meters. Your L/R balance can easily make up a big difference as the left only power meter mutiplies your left leg power by 2, which amplifies any difference. Other reasons may be due to needing calibration, or incorrect settings of some kind (e.g. crank arm length).
Intervals also allows you to separate indoor and outdoor power to account for different readings.
There’s a saying that a man with two watches is never sure of the time. (That’s why Buzz Aldrin wears 3.)
It might disappoint you, but you almost surely have only one Power meter…
Trainers, apart from a couple of really expensive ones, don’t have a power meter build-in.
Almost all of them use a Speed/Power conversion algorithm, taking into account the resistance. You can easily check that by spinning up to 200W for example and then stop pedalling. If the power comes down slowly (with wheel or flywheel speed), you know that it is the case. An exception is that there seems to be some models now where the power is set to zero when cadence is zero. Still you can check by pedalling slowly and not putting in any power while the trainer spins out.
If you happen to have a wheel-on trainer, tire pressure is a significant factor and should be checked before every ride. Wheel-on trainers are much less accurate then direct drive trainers. On direct-drive trainers, the more expensive ones are factory calibrated unit per unit. Every unit leaves the factory with its own speed/power curve baked in. The less expensive ones have an average power curve for a bunch of units. Accuracy depends on manufacturing tolerances. The calibration proces simply sets a fixed value to move the speed/power curve up or down but doesn’t influence the form of the curve. Again more accurate with direct-drive trainers, because those almost exclusively have a linear speed/power curve while wheel-on trainers mostly have exponential curves.
So, if you want to know which one is the most accurate, the Power meter (Stages in your case) is the most accurate one without a doubt and if not fully accurate, the most consistent one.
A mid-tier trainer that is setup correctly and consistently should be good enough for almost anyone when it comes to follow-up of your own performance. If you want to compare with others, only compare power-meter numbers, not trainer numbers.
@MedTechCD Thanks for pointing out that the smart trainer is perhaps “making up the data”. As per your suggestion I’ve pedaled to reach "a stable power, and then stopped applying pressure still slowly rotating the cranks to prevent cadence from dropping to 0. As far as I can tell, power 3s average drops to 0 within 3s, and the flywheel is still turning, so it seems that my trainer (Saris H3) does indeed send power data.
FWIW, my FTP went from 215 down to 150 since I switched to the Saris. I am m69. 205 certainly wasn’t elite but 150 is suspiciously low and doesn’t seem to match my experience (time to complete a course, maximum daily distance, etc.) or other fitness metrics (ex: “superior” Vo2Max) whereas 205 didn’t raise doubts.
On the one hand, it doesn’t matter because after a while power zones will fall to their new equilibrium. Still, i’d like to get a third (or more) source in order to clarify this.
I still have a couple of weeks to decide if I return the trainer.
It seemed fairly accurate in 2019…
Don’t get me wrong. The ‘making up the data’ isn’t as bad as it sounds. A well calibrated speed/power conversion will get you anything you need and makes sure that you don’t have to spend a fortune on an indoor trainer. Look at the price of the trusted power meters and imagine that every trainer would need such a device built-in to give you a reasonable power estimate.
One of the things that you rarely (never) do on a trainer, is just stop pedalling. If the power calculated from speed returns a close estimate, you have a good alternative at a lower cost. Who cares that things may be a bit off for circumstances that rarely or never occur on a trainer?
But it feels a bit strange that a high-end trainer like the Saris H3 is so different to the Stages single side PM. Stages may not be the best out there when it comes to accuracy, it is far from the worst either. Given the single side measurement, a difference of up to 10-12% could be explained, but surely not 30-50%. Something must be wrong with one or the other, or you have a huge leg imbalance.
Do you happen to have a friend that owns an easily swappable power meter? One of the pedal power meters for example? If so, ask if you can use them for a short test on the trainer to compare what they return.
Well, first I’ll rig another her unit and display power readings side by side. Next step in the plan is to hop to the nearest gym and try their bikes. And then, ride an uphill course, once Spring is with us. (Does intervals provide power estimates for outdoor segments)