New to Xert, and knowing that my outdoor power curve is far superior to my indoor one (both using the same PM pedals), I am looking for an option to differentiate between the two.
Yes, I could start scaling the signature or the PM pedals and use the indoor training power to try to align things.
But why not have two fitness signatures — one for outdoors and one for indoors? Intervals.icu lets you set separate outdoor and indoor FTP values to account for this.
Interesting way to look at this problem of power meter inconsistency. Another thought that comes to mind is that this could be used to effectively calibrate the power data Xert takes in. Each “gear profile” or whatever it would be called should obviously develop the same fitness signature over time so it seems like it should be possible to use that to work out automatic power correction for each profile.
I struggle with what to do about this problem every year because my indoor trainer is old and probably not very accurate from an absolute standpoint but also likely much different that my outdoor bike (a mountain bike of all things) power meter. It seems like the industry answer is to keep working towards higher accuracy but I’m not sure everyone has agreed on a standard for assessing that yet.
No doubt power accuracy for PMs is a worthy goal but the most helpful thing for me, and I suspect a fairly large majority of Xert users, would be if we could bring our outdoor bike indoors on the trainer and have both power meters going with Xert automatically generating a correction profile that it would save and apply to incoming data automatically.
This request is not related to power meter accuracy — I use the same PM pedals indoors and outdoors. It is about the well-known performance difference between indoor and outdoor power output.
If my fitness signature is calibrated for indoor use and I do a hard outdoor group ride or VO₂ intervals on Monday, Xert may detect a breakthrough of ~30 W relative to my indoor signature.
What should happen next? Accept the breakthrough and end up with an indoor signature that is no longer accurate for my next indoor session on Wednesday?
Alternatively, consider that Monday’s effort may represent only a small breakthrough (e.g., ~5 W) relative to an outdoor profile. In that case, the outdoor signature would update, while the indoor signature would remain unchanged.
Likewise, detraining would shift both signatures downward over time.
For these reasons, having two separate fitness signatures — one for indoor training and one for outdoor riding — would allow more accurate tracking and reduce the need for manual adjustments.
Having two separate signatures is equivalent to scaling up your indoor power meter power so that it matches the same “effort” as outdoor power. Keeping two separate signature profiles and translating one to the other across all 3 training loads is no small feat. Better to just scale up indoor power to match. This is something we have in the feature backlog.
My personal thought is that the evidence to suggest indoor training applies additional training stimulus compared equivalent outdoor power isn’t definitive. I, for one, found I’ve been able to achieve similar power numbers with a focus on hydration, increased sweat rates and adequate ventilation. If I didn’t address these, I don’t think upping the effective power of my indoor rides is accurately representing the training stimulus.
For me, I find the placement of the fan to be critical. If it’s blocked by my arm or handlebars, for example, body temperatures will rise and HR can be 5-10 BPM higher for the same power.
I like your points, however, unfortunately, Xert is not very precise about the breakthrough. Just for example, a few weeks ago I had a breakthrough during an outdoor race, raising my FTP by 30 watts. This kind of increase is completely unmanageable during indoor sessions.
However, I think Xert should also use the heart rate data to correctly balance the fitness signature rather than creating two different ones for indoor and outdoor activities
Hi Armando,
Two separate profiles is not the same as simply scaling one of the power meters.
Firstly, to scale correctly you would need to know the exact scaling factor — but that factor is unknown and likely varies over time. The indoor–outdoor difference also depends on training. When doing the first indoor sessions after the outdoor season, the gap is usually much larger than at the end of winter.
Secondly, this approach assumes that the indoor profile is a linear scaling of the entire duration–power relationship: that your 5-minute indoor max scales the same way as your 1-hour max.
But that is not true. The indoor–outdoor offset is not a constant ratio across all intensities and durations.
This is why two separate profiles — one indoor and one outdoor — would provide a far more accurate representation.
This add significantly more complexity to an implementation, having to track separate signatures and translations. Does a breakthrough indoors get applied to your outdoor profile? Is that scaling linear across all 3 signature parameters? Will that actually work? If you delete a breakthrough in one profile which then causes a breakthrough in another profile and that then has to be applied to all subsequent activity XSS and signature calculations for both profiles.
It’s an interesting idea and if the team had less going on, it would be a fun project. It’s a ton of work though and would be challenging to test and ensure it’s working correctly.
I go by the principle: don’t make up data that isn’t there.
That being said, if a user knows that their outdoor performance is at least equal to — or better than — their indoor performance, then it makes sense to increase the outdoor fitness signature (per parameter) when the indoor signature surpasses the outdoor one.
This could happen, for example, after a successful indoor training block during the winter.
If your PM pedals are paired to Power indoors on EBC, the Powermatch feature is automatic. Do your pedals and trainer auto-calibrate?
How many watts difference are you experiencing on a percentage basis? Example, 10-12 watts at ~4%. Or is it 5-10% lower indoors based on past experience? If so, how did you determine that? Is “30 watts” hypothetical or actual?
A significant difference is more likely due to equipment, position/movement, and environmental factors. Otherwise, you are the same physiologically indoors and outdoors but performing under different conditions. Consider a long hard climb on a cool autumn day versus a mid-summer heat slog in +75% humidity. Same ride. Same engine. Different performance.
Unless you’re into heat training 
your pain cave should resemble a cool autumn day if possible. Even so performance will differ due to constant pedal stroke and bike handling/position indoors.
If you ride Z2 easy for 1-2 hours how much sweat pools under your trainer?
Should be near zero with sufficient cooling. If not, there are several ways to address that and reduce the outdoor/indoor difference.
It’s not unusual for riders to express a Peak Power (PP) 100+ watts higher outdoors, but TP is usually close enough for indoor training purposes. If not, you can effectively adapt by lowering % intensity or simply ride in slope/resistance mode which many coaches recommend.
I’ll flip/back and forth between AUTO and Slope on EBC when required on hard sets. Or free ride in Slope mode with XMB to meet today’s goal which I can perform at my own pace at variable watt targets. Doesn’t need to be exact to accomplish the intent and XSS goal for the day.
A 30-watt difference is based on my 5-minute power: 420 W outdoors vs. 390 W indoors.
TP is much closer — around 10–20 watts — and peak power is way off, simply because I don’t want to do full standing sprints on the trainer.
My power data always comes directly from the pedals, and yes, they auto-calibrate.
For an untrained eye, an indoor trainer might look identical to cycling on the road, but it’s not. Outdoors, the bike rocks from side to side, and cyclists who rely on upper-body engagement to produce power can’t reproduce that movement indoors. Besides that, there are numerous other factors that can limit peak performance indoors. That is something different from performance occasionally being lower due to sub-optimal conditions.
I would like to keep this discussion on topic — the request to have two fitness signatures.
not possible as the signature model used by xert says, there are no different power signatures, just a different ability to reach higher difficulty scores (peripheral fatigue is equal, central fatigue might be higher indoors). Modeling always means imperfection.
I agree with this model. It’s amazing the difference 2 good fans and cool indoor temps can make. My trainer power isn’t accurate enough to for me to feel comfortable dismissing the OP’s claim of an obvious need for 2 Fitness Signatures (indoor and outdoor), but it just doesn’t make intuitive sense to me at all. It’s simply more difficult to fully express your fitness indoors IMO. I don’t think your training suffers due to any potentially inaccurate signature in this case because the effort is still the same. Your numbers might just be a little higher outside and likely only on the top end. 
You asked why not have two signatures? 
I’m making the “not” counterpoint 
and suggesting most users shouldn’t need this.
If greater than 4% at TP (measured or percieved) riders should exhaust all possible remedies before assuming dual citizenship will be of benefit.
I can think of six contributing factors plus some related aspects unique to Xert that obviate the need.
Is this a good place to discuss those factors or should I start a new topic?
I agree you can’t thrash around indoors the same as outdoors, but you should be able to rise up and stomp on your pedals when required. No need to match your outdoor PP. 