Signature keeps resetting

I’ve conducted a Threshold test today and entered (and saved) my FTP into my signature to be calculated and as soon as I select to recalculate do all of the numbers reset within the selection (threshold, max power) etc. This throws my data way out and makes this application useless to me. Is there a glitch? All was fine until the latest update.

Hi Mark. A progression recalculation will take your current signature and use it with your first activity and trickle down from there. If you have a long progression, it won’t likely make any difference to your current signature.

Looking at your data, you may have some data anomalies since your Peak Power is very high.

We’ll have a look and see if we can help spot the issue.

Looks like you had a couple of activities with some data anomalies that our filter didn’t pick up. Their signatures have been reset and locked with the same values as the previous activity and it looks like things have now sorted themselves out.

Thanks for letting us know you were having issues. Let us know if things are better.

For some reason my FTP on the signature is 235w, I’ve tested 3 times over the last 4 months at 275/276w (yesterday was the third test) but still the data hasn’t changed from 235w and forecasts 240w in 6 weeks. There’s still a glitch present and it doesn’t reset when i reset the signature or manually try to change it. The 2100w high anomaly has also come back (lost traction on a TT and the back wheel spun as I accelerated out of a corner) where it should be closer to 1200w. Can you resolve this please?

Please remove the activity with the 2100W anomaly and then add it back in. If the data spike is still present, let us know.

After reviewing the activity, it appears like the majority of the ride data is faulty so removing and adding it back in will not resolve the issue. There was also another activity with data spikes. We’ve saved/locked their signatures to more reasonable values. Let us know if the progression is better. If you perform a progression recalculation in the future, their signatures will get reset and you will need to manually save/lock new values again.

Your comment “the majority of the ride data is faulty” - Please explain? I have no issues with Training Peaks/Strava/TrainerRoad reading the data and this is the first I’ve heard of any data being faulty? It’s either recorded via a Wattbike, Stages powermeter, or Powertap pedals to one of two different Garmins, or via a direct data download from TrainerRoad or wattbike. That’s different sources of data and different equipment recording the data and either a laptop or a pc is used which are evn different computing systems. What’s the common denominator in the ride data being faulty, do you have example activities or dates? I (did) do a lot of sprinting so there may be the odd wheelslip on wet roads which will give a spike (hence 2300 watts for 1 second) but I’m not aware of other anomalies. I think that’s example has only happened twice, maybe 3 times at the most.

Currently, the FTP shows as 264 on Xert since you’ve reset, although it’s actually 275 (see above) - I was last tested on 27th December 2016 and the activity is viewable on my XERT activity of that date.

Whe re do we go from here? What else can be done? I like Xert a lot, but if it’s not accurate then it’s of no use to me. I’d really like to work with you to resolve this but if it can’t be resolved please let me know and I’ll discontinue my DD, although I’d rather not reach that conclusion.

Hi Mark. The issue was with just one of your rides. The majority of its data was faulty.

How are you obtaining your FTP? If you’re using an 8 minute or 20 minute test, these will overestimate your FTP because your HIE is above average. That, is the 90% or 95% factor is likely a bit too high for you. These tests don’t measure your HIE which is vital in determining your actual FTP. If you look at your patterns over more variable rides in the past and open the power duration tab, you’ll see that your FTP is less than 95% of your 20MMP for those rides pretty consistently.

Thank you for your feedback and apologies for all the confusion.