Is there any way of adjusting the power reported by the trainer to match the power from your your outside bike so your metrics are the same inside and out. Ie my trainer reads consistently 9% lower than my power meter on my bike.
NB
1 .I know what Power Match is and that is not what I talking about and
2. It is not feasible for me use my outside bike on my trainer on a consistent basis.
Thanks
How do you know that your trainer reads 9% lower?
Probably because you set up your outside bike on the trainer and compared power output from both?
Are we talking about a wheel-on trainer?
If so, this will sound disapointing but with another bike on your trainer, it is almost 100% sure that power values will be different. That is because power from a wheel-on trainer is very dependent on the bike/trainer interface. Two different bikes have a different position on the trainer and thus the tension off the trainer roller on the tire is different. The tires are different and probably the tire pressure is different. So you can’t simply port the power from one bike to another on these trainers.
If you can temporarely transfer the power meter to your inside bike, you can make it match quite closely. Check this post:
https://forums.trainerday.com/t/power-match-not-working-well-on-shorter-intervals/987/12?u=medtechcd
After doing that, remove your Power Meter and do not recalibrate your trainer!
1 Like
Yes I’ve put my outside bike on the trainer for s number of rides and 9 % is the average variance. It’s a wheel off trainer (Elite Zumo) and die to where I live it’s not practical to keep taking the bike in & off it.
There are lots of posts on numerous forums about this issue and it seems to me there is a simple solution. On the the training platform enable a power scaling factor to increase or decrease the reported and recorded power. Xert used to do it but the feature is missing from the latest apps. There are ways round it but it’s inconvenient and time consuming
That simple solution would be confusing because the target sent to the trainer can’t change and the output would be scaled. You will constantly see 2 different Power numbers and the output graph will float above or beneath the target graph…
1 Like
This is interesting. It’s like power match without a second power device. Now that I hear it I am surprised no one else has requested this. This does sound easy but somehow it scares me that it might not be. Let me think on this a bit.
It’s difficult to explain but here goes
Say I’ve got the Trainer and my power meter on together and I’m pedalling at 100W as measured by the Trainer. I’m actually putting out 109W (from the power meter and measured concurrently on say a Garmin). If I was then to put the Power Scaling factor to 1.09 and continue pedalling at the same intensity the App which is showing the power from the trainer would show 109W which would now match the 109W shown from the PM on the Garmin because the intensity hasn’t changed. In this case the .fit file from the app would show the initial period at 100W then the 2nd period at 109W.
one problem is that the % variation can change with intensity and sometimes it can be a fixed offset plus a % variation. My take however is that being out by a variable plus or minus 2/3 % is better than being out every time by 9%
I get it but thanks for clarifying.
Yes, I agree makes sense.
It might be a simple solution to use PowerMatch function to correct a “virtual” power meter with a simple scaling factor.
I know that TrainerRoad does (did) powermatch with several functions. One was a permanent offset.
In this case the target sent to the trainer should be scaled to 91% but displayed as 100%.
Forget about the “measuring” from the trainer. There is not one single trainer that measures power. The trainer takes a target Power and puts you on a predefined speed/power curve. Riding at a specific speed creates a resistance determined by that speed.
Just a note to say that I would very much appreciate and use a linear “adjustment factor” (for example, “+3.5%” or “-2%” with TrainerDay. The reason it matters to me (and the reason I would use it over just adding the +/- % during the workout) is because I like to do a lot of post-ride analysis (calculating TSS, using intervals.icu, identifying best performances over certain periods of time) and when my power meters read vastly different numbers, these analyses are less useful (for example, all my best performances appear indoors, when in reality it’s just because my indoor trainer is a consistent 8% higher than my outdoor power meter)…
1 Like
Some how I have warmed up to this idea. I think I was stressed with bugs or something last time it was requested… Always a little stressed with some bug here or there but they seem to be getting smaller and smaller other than for the person that is affected… Anyway, overall I like this idea, meaning I don’t love power match or at least our power match and really a linear adjustment is all that most people should need that use power meters to get it close enough. So let me have my developer analyze this and see what we can do.
1 Like
@Alex Man, you’re the best. I love this app/program/community so much. THANK YOU for everything you do.
1 Like
So you are suggesting % difference but I think the more likely scenario is a watts difference? I don’t know about this, I know mine seem to be closer to 10w difference and most people I hear talk about watts difference.
So I am thinking something like this.
@MedTechCD I would take a strange guess that you are more of an expert in what what is most common and best 
Keep it simple is my ‘moto’ 
Y = a * X + b
Where X is the original value and Y is the modified value.
When you see a fixed offset over the whole range, a is 1 and b is the offset value.
When you see only a slope change where lower wattage values have less/more offset then higher values, a becomes for example 0.98 or 1.02 and b is 0.
In case you have a combination of slope change and offset, enter both a and b.
This will work perfectly for most smart trainers that have a (almost) linear speed/power curve.
For trainers with an exponential speed/power behaviour, it’s likely best to keep a = 1 and only use a fixed (b) offset. It could become chaotic with a multiplier.
My difference averages out at 7% and varies depending on fatigue, cadence, torque and intensity. For me a straight forward @% would do, a fixed number of watts would be next to useless. @MedTechCD idea satisfies both scenarios and would be ideal
That’s simple for math guys but confusing for less math specific but for sure it is the most flexible. For us it’s a balance of flexible and simple. Need to think about this.
Yeah the more I think about it, you are 100% right. Glad I asked. Anyone doing this will either understand or ask, which is fine. Really this should compensate for most power meters with out the need for power match.