Significant power reading differences between Stages Left and Right crank-based power meter

I have a Stages Left power meter on my road bike and a Stages Right-sided power meter on my gravel bike, both are Ultegra, Gen 3 and have been purchased in the last several years.

Recently, I noticed that riding at 200 Watts with my gravel bike was a lot harder than riding at 200 Watts on my road bike and so I tested the power meters indoors while dual recording the power from my Wahoo Kickr Core (has a power accuracy of around 2%).

My indoor test protocol was as follows:

  • 15 minutes of warm-up for the Kickr Core and then calibrated Wahoo Kickr Core and the Stages L and R power meters using the manufacturers Apps
  • new batteries were installed
  • both bikes have waxed chains and the used the same gearing (straight chain-line)
  • Performed 8 minute test (ERG mode on TrainerRoad) using gravel bike with power recorded from the Stages R power meter (TR app captured Stages power data; Zwift captured Kickr Core power data)
  • Immediately after, performed the same 8 minute test using road bike with power recorded from the Stages L power meter (TR app captured Stages power data; Zwift captured Kickr Core power data)
  • Compared .fit files using DC Rainmaker Analyzer tool

The results were horrendous (for Stages). Assuming Kickr Core is accurate, the Stages Left power meter averaged 30 watts more than the Kickr Core and the Stages Right power meter averaged 50 watts less than the Kickr Core. This means there is a combined 80 watt difference in power readings when comparing the Stages L and Stages R power meter. How can 2 Stages Power meters differ more than 30% when each has a stated accuracy of less than 2%. And I don’t accept my L/R power inbalance to be that high.

I created a Stages support ticket 2 months ago and haven’t received a reply which is quite disappointing. I am wondering if any forum members have any suggestions on how to reconcile the significant power reading differences between the Stages power meter. Frankly, I am almost ready to ditch Stages and look at the Wahoo Power pedals for my road bike and the 4iiii crank based power meter for my gravel bike…

Attaching the graphs for completeness -
Stages Left vs Kickr Core

Stages Right vs Kickr Core

  • note: I ‘failed’ interval 3 as I was well above threshold; as I was running on Erg mode from the Stages power reading.

Honestly hard to say without having a true dual sided. (Garmin Rally, Wahoo PowerLink, Favero Assomias)

If you’re planning on getting the wahoo ones anyway worth running them dual recording and seeing if the stages are accurate and it’s your left/right balance is quite unbalanced.

Otherwise if you’ve got a friend with a set of power pedals, see if you can borrow those and do the above test to check.

Stages right side on Shimano are known to be a bit wack but 50watts is more than I’ve seen elsewhere.

Which model Ultegra crank? R8000?

It is hrs to get meaningful interpretations from the graphs you show and test you did. Consider redoing the test where you ride on a preset ERG workout the increases the step by 25w each 2 mins. For example,
2min at 100w
2min at 125w

2min at 275w
2min at 300w

Then redo the same exact test with the other bike.

Finally, upload that data and compare apples to apples. You have apples to oranges on those graphs.

Edit: I don’t mean to suggest making it a super hard ramp test. Stop it before you work too hard. But you need to have it be replicable from one bike to the next.

Thanks for the reply - Ultegra FC-6800 for the left side and Ultegra FC-R8000 for the right (drive) side.

@jonesy - Yes, I also concluded that dual side power pedals will help determine whether there is a power imbalance at play. I suspect that the difference in how the left and right cranks are constructed may play a factor with the Stages power meter, but I would have hoped that Stages would have figured this out.

Right - I updated the graphs in the original post to show the entire test for Left vs. Right. I did 4 intervals at 4 different power steps, but you will notice that I failed the 3rd interval with the second test as I was pushing 50 watts more. I don’t think this invalidates the findings however.

That’s better. Compute the average for each step for the two different graphs. The erg usually take a few seconds to switch power, so if ignore the first 15 seconds of each new step. You prob want at least 60-90sec of data (after trimming the first 15-30sec) to get something meaningful.

I’ve done this myself. I have a cycleops hammer and two bikes with left side only stages. They do read differently, but more like 15-30w difference. This difference depends on the power level too. It’s not a constant difference. For example, it may be a 10w difference at 200w, but a 30w difference at 400w. It’s not a constant percent difference either.

Your testing is more challenging bc it’s a difference between the two legs, not just two same side meters. So you can’t make much of a meaningful statement about accuracy. Also, make sure firmware is updated on all devices. That can matter a lot

*** you need to run it off ERG based on the kickr power, not stages. That way, kickr power is as constant across the two bikes as it can get

1 Like