Is zpower wattage really over-estimated?

(Jonathan [Grenoble]) #1

Hi Zwift team!

My short question is: is zpower accurate when using properly a modern non-smart home trainer?

The background of this question comes from regular comments I see (on Strava, Facebook, forum…) claiming that zpower is 20-30 watts higher then what you would get on a smart-trainer. Then, I see some more comments claiming that smart-trainers are (again!) 20-30 watts higher then power meters.

I know that Zwift has spent an extensive time to calibrate a LARGE range of trainers, smart and non-smart. I really wonder why you would have introduced such a bias, it doesn’t make any sense to me. I understand that zpower makes mistakes (or cheating) easier, but for someone who does it properly (a modern trainer, regular wheel size, proper tire pressure, reliable speed sensor, right weight/height in profile, …) it should be just right, no?

PS: I have a lot of fun on Zwift (including racing), but as a mountain biker who has already invested too much in a road bike and trainer I can’t picture myself spending more, just to prove my power is not 40 watts over real value. I’d rather get this come from you, free of charge :slight_smile:



(Paul Bestwick ZTAZ) #2

I removed the first 3 B group finishers from the ZwiftPower results on my ride last night. They were all Z Power.  

(Eric C. (Zwift HQ)) #3

So, zPower is still an estimate and it requires the use of math. It also depends on the classic trainer-in-question. When you’re estimating power, you’re doing your best to take into account all the variables that might be happening (fluid heating up over time, magnetic resistance cables that are pulled too taught or not at all, etc…)

As with any estimate, there will be variation but we believe that it works fairly well for training as the power will be consistent from day-to-day, all other things being equal (tire pressure, tire width, tread, etc etc).

Claims are just that - it’s different from trainer to trainer and we will update some of our power curves as time goes on (though we are pretty strapped on developer resources right now).

As for Smart Trainers vs Power Meters - it’s all on what they report. They tell us the wattage that is being generated, we use that data. Some power meters are more accurate than others and some smart trainers are more accurate than others.

They key is to be consistent so your training never varies too much and you can see steady, reliable progress.

For races? The only truly FAIR way to do this is if everyone is on the same trainer/power meters with the same size tired (or wheel off if that’s the case) with the SAME calibration. Everything else is as close as you can get and there will always be variations.

(Jonathan [Grenoble]) #4

Thanks for the detailed explanation, it really helps!

I just found an affordable power meter (4iiii), I guess it’s the best way to go to get further into racing on Zwift. The next question is, am I really B or C category :smiley: (A cat. with zpower).

(Jonathan Whiting) #5

I’ve been wondering this question for a while now as I use a classic trainer. CycleOps Magneto. Just got my hands on some Assioma power meters so did a test. It was overestimating my power by an average of 16%. But then after 550watts it stops increasing the data. If anyone’s interested, here’s the post, with the full data.

Zwift if you would like the  .fit files I would be happy to share them.