XERT, your take?

[Stefan]

I've been a paying subscriber for a few months now. More out of curiosity than an actual need. As a long time GC user I actually have all the tools available that I need for my training analysis.

I must admit I'm not overly impressed from what XERT delivers me right now. There seems to be some misconception on what XERT offers at the moment and what it may offer in the future. For example taking into account fatigue/strain/stress into their fitness signature extraction. I signed up with the believe that experienced load during a workout would impact this extraction. Unfortunately not, this is a feature to be implemented. Of course my bad, but this is just one example. But judging from the XERT facebook community this does not seem to be clear to others as well.

Furthermore, I had hoped that their model/algorithm makes testing less important. Not really, while they do not call it testing, they say you "just" have to do maximum efforts weekly to keep your signature up to date. Since it is a three parameter model you'll probably have to test different durations. You you actually do more testing which I find quite stressfull. But once again, this is just me. Others may enjoy this.

Once you've done your testing you can't be certain that your new fitness signature is true. Many posting on FB on this. People are advised to "play around" with those three parameters until they believe it describes reality better. I don't really see much different to just setting your parameters based on your experiences in your training, e.g. your power distribution curve.

This is not rant, I know some people are very passionate about XERT, I just would like hear other opinions. Especially from the GC community since here a people who do have experience with Wbal modelling, PD Curves and so on. So XERT may not look so revolutionary here.

And as GC useres can we learn something from XERT? What are your use cases? Do you use both?

[Mark Liversedge]

Can't comment on utility, will leave that to others, but I can confirm we will be enabling upload/download directly with Xert in 3.5.

[Harold Heyermans]

I have been using XERT going on for just over a year. I started in the fall and played around with its App indoors and using the smart workouts. I didnt pay attention too much to the threshold number until after a few months. I had hit a few breakthroughs and my threshold was climbing to a point that I knew was correct from previous seasons. I did have to try a few hard efforts go get my MPA up as normally my trainer workouts don't involve sprinting. Once I established this I did not do any 20 min efforts to gauge my fitness. The threshold was pretty spot on for me going into the outdoor riding season. 

My only issue I had was with the training status form. I was always in some form of overtraining the whole year. I would have to comment as last year was not my best. I really only had one stellar ride and the rest was OK.  So maybe it was correct and I just ignored the fact. This indoor season I will be paying closer attention to it. 

I am just now starting to use the new fitness planner. I have it populated to keep myself on a good linear increase without putting myself in the red. My first event is not until mid summer and have two lesser events in late May, early June. 

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-cheetah-users@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to the Google Groups "golden-cheetah-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-users+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Daniel Holmes]

I've been using it for a few months and find it interesting. Although i'm not sure if I find it accurate or not. 

It's telling me my 5 minute power is 404w which is pretty big for a threshold of somewhere around 305-310w. I know my 5 minute sustained power is around 360w because I have quite a few criterium races and general PB efforts maxing out my sustained 5 minute power at around 360w. It's as if it's a wall I hit each race or hard ride I do. So I'm pretty confident that's my current 5 minute sustained power. The guys on the Xert user group just tell me that I'm not pushing hard enough and I have to believe in myself.

I want Xert to work but I can't trust it when it's suggesting those sort of numbers.

[Will]

I tried it for a few months as well. I agree with the experience above. I find the CP/W' model to be roughly equal for super hard interval work. Similarly, you need good data to construct the model. But, I've definitely seen a few KJ of variation on any given day just based on how I'm feeling, time of day, nutrition, etc.

All you can do is all you can do, but having the W'bal in front of me is a useful guide. Eyeballing the MMP curve and knowing the context of the data points is probably the best ballpark estimate of what you could do, rather than a totally theoretical model, IMO.

So, no real-time model is going to be perfect and I don't think Xert is superior to GC. I could say the same about WKO4, for sure, and probably any other software.

Will

[Stefan]

I've tried to update my "all-out database" recently. I want to see how XERT MPA and CP/W' produce differences for me. As can be seen below in the two charts, they behave similar.

Really one of the challenges for me is getting robust parameters from CP/W'. This is not as straightforward as it may seem. Depending on the input durations I obtain quite different model parameters. Even more variation when I consider the recommendations from here: http://www.jsams.org/article/S1440-2440(17)31817-0/fulltext

XERT simply gives you a set of parameters, whether these are more robust is not known for me. If I do some "Advanced MPA analysis" I can obtain quite diverse parameter sets as well.

What XERT does very well is the Garmin field MPA/Power. When doing all out efforts this is really motivating to push it a little bit further. However, I haven't tried the available W' Garmin field yet.

Just listened to a recent podcast on XERT. There it has been said that one of the great advantages of XERT is that no testing is required. Parameters/fitness siganture changes automatically with every ride. I find this statement misleading. I don't do 5 min all-out-want-to-vomit-afterwards-efforts by accident. This is just a matter of semantics if you want to call it a test or not.

[Mark Liversedge]

On Monday, 25 December 2017 06:08:44 UTC, Stefan wrote:

Really one of the challenges for me is getting robust parameters from CP/W'.

You need to follow a reliable and repeatable test protocol

Merry Christmas

Mark

[Stefan]

Getting adequate all out efforts in the respective time domains is actually one of my least problems. This is pretty straightforward for me and looking at the data points very plausible.

I'm struggeling more with all the different models and their different outputs. Then adding this recent publication linked above on required minimum durations for good estimates. More uncertainty. So as usual as in sports science: "it always depends ..." For a "naive" practitioneer not necessarily useful.

But of course, this is just my non-expert take.