In case any wants to see what the actual power meter mean difference is, here's the DC Rainmaker Analyzer set:
https://analyze.dcrainmaker.com/...a5-6729-3ff5a7327ada

This is their "calibration testing workout protocol" they provided me with, and still, same results:
https://analyze.dcrainmaker.com/...83-5d7f-657889237aa8

And this is my friend's Kicker Bike, also 1.0 with a different set of pedals (Brand new Garmin Rally Shimano) and still showing the exact same 4% difference just last week:
https://analyze.dcrainmaker.com/...45-6c6f-e7c0c3a298de

This could all be fixed if they would only provide us with a calibration factor offset tool within the app.

definitely sympathetic to the issue, but I would also say that your daily variance is also probably greater than that 4%. I.e., FTP is really a ballpark number that varies day-to-day based on fatigue, hydration, temperature, etc. I would be shocked if you did MLSS testing every day and saw identical results at a given power output (except for very low power).

Also, is it *consistently* 4% off from your pedals? If it's consistent, then I also think that's less of a big deal. I.e., if it's always high or always low by pretty much the same amount, then I don't know how big a deal that is, since you basically just adjust for the discrepancy.

You have a number of different options depending on the training software you use. On Zwift, for example, you can set your powermeter and your trainer to be different, and Zwift will apply resistance to the trainer based on whatever your powermeter says. I always use my Quarq as my powermeter - because that's what I ride outside - and then just have Zwift adjust the resistance on my KICKR.

I'm guessing you're using your pedals outside - and so you probably should be using those inside regardless. IIRC, Wahoo is itself *estimating* losses - such that the power reported is *closer* to crank power than actual power at the hub. So the idea that your power at the pedals and at the hub would be the same is already the result of a fair bit of approximation. I'd think you'd just want to use the exact same device in all cases. Though even if you do vary devices, I think it's fine, because again, threshold values are way less consistent than people think.

Let's say you have two powermeters that are within the industry standard of +/- 1.5%. Those two units can be 3% different from each other and still be perfectly in line with the top of industry standards. That's not all that different from 4%.

So, basically, yes I agree it's frustrating. But I also wouldn't think - especially given that you have pedals you can rely on for accurate data - that it should meaningfully impact your training.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

Quote:

Our development team simply does not give us even a vague estimate of one, because of the complex nature of their work.

Oof, that's rough. I understand not giving a public estimate. I don't understand throwing the development team under the bus like this. Just say there's no public estimate and leave it at that. This makes it sound like the development team has no idea what's going on, which isn't great to release publicly. Because once a root cause is identified at least a "SWAG" estimate should be possible internally.

2-5% seems totally normal. There will be some inevitable variation around your calibrations based on the components themselves. Battery life and temperature also plays a role. It might not just be "low battery" that is an issue, but a poor quality battery and so on and so on.

Looking for better than 5% accurate from a consumer tool that's not being designed precise life and death scientific lab measurements is just unrealistic.

I agree with their forthright communication that sending you a new version won't solve the issue. And they can't snap their fingers and just make the issue go away with coding as fast as you might think. Building in some kind of compensation measurement algorithm would likely cause other issues and even if it didn't the still need to bug test everything as one fix can often break something else on these kinds of things.

The reality is that the trainer is good enough for you to use and it's all a relative measurement. If you expect to take that exact number and move to another device using that exact number, you need to be willing to factor in some error bars. 1-2% is a number I would personally look at for common medical grade equipment error deviation.

The real question is how is this adversely affecting you? Are you unable to train? Or just unable to say with certainty, my power output is exactly X.

2-5% instantaneous random error might be normal. 2-5% constant bias in one direction that is unfixable with a calibration factor is not normal. And it sounds like that's what's being referred to here.

Exactly, what's worse is, there seems to be a batch that are NOT affected by this, as noted by DC Rainmaker himself, his didn't had any issues. So it's not like the infamous "Neo T2 effect" some other people experience, it's just a certain batch of bikes and basically there's nothing we could do.

If at least they would give us a way to offset this, that'd be great, or at the bare minimum, have the Kickr actually work with a different ANT Power Meter as master, something that's actually IN THE APP, but doesn't work anyways.

It sucks because it doesn't correlate with the great products the company is known for.

My current setup is having my Assiomas on the KICKR BIKE and control the Bike via ZWIFT paring the assiomas as the Power Meter and the Bike as the controllable device, HOWEVER, this won't work in say, TrainerRoad, so what I would usually do then is have the actual .fit and scale it down using fittols.com to -4% but it's a effing mess, no to mention the RPE is all over the place when you see a VO2 of 380 versus a 364.

You're lucky it's only 4%. My first gen kickr is at least 20-25W off. I show up to a Zwift race and destroy the field lol 😂
There used to be a giant thread here back in the day and Wahoo wasn't helpful back then either.
Luckily, with the new software Zwift and Trainerroad etc. they can all pair your power meter and drive the trainer resistance accordingly.

---

What's your CdA?

Why would this not work for TrainerRoad? This is exactly what power match is for in TR

The Stages SB20 is just as bad in terms of accuracy - so its not just Wahoo!

One day it tracks my pedals, the next its was off. I gave up trying to match them, gave up measuring L:R balance too - took the pedals off and I just train on it - its close enough.

How these things, costing this much can be so inaccurate is crazy...

---

Regards, Richard

Considering exactly 0 power meters read the same. A 4% variance that is consistent against some other manufacturers pedals seems completely "fine". Way different if it was 20% which you definitely get from spin bikes.

---

Washed up footy player turned Triathlete.

This sucks. Wahoo has a history of absolutely astounding engineering decisions - ie using a sticker to calculate cadence on Wahoo Snap.
I simply refuse to buy any of their v1.0/1.1/1.2 products. Wait for the 4th iteration or so.

---

Next races on the schedule: none at the moment