Stages vs H2 power readings

Hey guys, so decided to compare my stages PM to my H2 PM yesterday. Was doing pioneer -2 in ERG mode so I turned on my wahoo and just watched the Stages 3sec avg power through that and compared it to the 3sec avg power that I broadcast to Zwift while I do my TR rides.

I noticed in the beginnig the power was off, stages was about 10-15w high during the warmup while trainer was cold. As things got towards the 10 minute mark they started tracking closer together, but the Stages was always more “jumpy”. As an example

TR had me doing a 4 minute interval at 220w
Zwift (mirroring the H2 PM) was showing my 3 second power anywhere between 215 and 225 as it “floated” which is normal
Stages showed my 3 second power anywhere between 210 and 230, sometimes dropping as low as 200 momentarily, and as high as 235ish momentarily as well.

If I made an effort to really spin super smooth they tracked within about 5-6w of each other, but normally pedalling the stages seems more “sensitive” than the Hammer, which I guess is to be expected? I just want to know if this behavior is normal. I understand they will never be 1:1 perfectly together but also want to make sure there’s nothing wrong with the PM seeing those larger dips and spikes. I got the H2 new, but the Stages I got used, so if there’s any potential doubt I have it definitely lies with the Stages.

Yes, this is totally normal. If you want to really dig into it you should record the ride on the Wahoo as well as with TrainerRoad and look at the averages over time


I feel we need to make an FAQ for " I have two power devices and they don’t match, why?".

  1. Even with perfect conditions, each device has a power measuring tolerance. That means each one could be reading high or low from the “real” power within their specified tolerance.

  2. Comparing a device that measures total power (like a trainer, hub, sprocket or spider based) to one that measures partial power and estimates the total (left-side cranks and pedals) will be influenced by left-right imbalance.

  3. Despite the manufacturer claims, we see potential evidence that some devices may well be performing outside the tolerance specifications.

  4. The location of power measurement can impact the results.

  5. Power data smoothing will impact eye balled comparisons, even if we do our best to be “smooth and steady”.

There are probably more, but those are the ones I know are relevant and have seen in my own testing.

1 Like