Busy winter for the crew at Xert and they are rolling out lots of improvements, just in time to pack up the rollers and get outside.
One the new features that I feel hasn’t gotten much explanation yet is DURABILTY. I’ve heard that what really seperates the pros from us punters is not just their high threshold power but their ability to do work after burning 6000kj.
How is everyone using this new metric in Xert?
For starters, I went to my activity list and filtered the minimum time to 180 minutes. I don’t have many races, most of my long rides are endurace pace and don’t have many with hard efforts at the end. I started going into each activity and clicking the “Access Durability” button. Most activites weren’t very intense and gave me a very high (meaning low) Durabilty Score.
The article on Durability Scores says normal ranges are between 5%-15%. After searching for a while I found a gravel ride from Aug 2024 that returned 12%.
Sooooo, now what? Is the Durabilty Score in my profile somewhere I can look at? Does it only pertain to that one ride? Is it just part of my fitness profile behind the scenes? Are we supposed to be trying for durability breakthrough as well as fitness breakthroughs now?
First we got Smart Workouts and then AI Planner. Now we got Magic Buckets and Durability Scores. Tadej better watch out because we’re coming for him!
I also see the durability score is based on kJ and not XSS. Do we expect to see the same decline in capacity after 3 hours at 200 W vs 2 hours at 300 W?
Thanks for the link, Scott. That doesn’t really give any new info on the durability feature though.
It feels like the durability score is generating “data” but not “information”.
Maybe in the next round of updates it will be further developed into something that is actionable.
Hi @blairc , it’s still so new that I’m not sure if there isn’t anything actionable, yet. Do you have any thoughts on this? Perhaps a start is to analyze past rides and seeing where your durability falls. Moving forward, maybe you can start tracking your calorie intake (specifically Carb) during rides. AFAIK, the largest contributor to durability is nutrition. If you’re not eating, don’t expect to have excellent durability.
Bigger picture idea here… Perhaps allowing Race AI to account for durability would be one way to expand this in the future. Let’s say you need to do a hard effort later in the race, so you need to ensure that your signature is sufficiently high - with durability considered - to be able to achieve that effort.
This is a good idea though obviously requires a good durability estimate and carb intake info.
If someone is motivated they can specifically test and track this over time, with this approach. Arguably removes the need for an all out effort at the start of the ride (which is the alternative approach to measuring durability). You could also test fueling impacts as you say, but think as big a factor (or bigger?) is training load?
You can presumably also measure XSS during a ride vs the ‘durability decayed’ signature for a better estimate of load and fatigue from long rides? Goes in the direction of the Xert 2.0 we hear about…
I’m less sure if it necessarily changes training prescription though…? I hear about doing intervals at the end vs start of a ride but don’t know if that’s proven to be better for improving durability? Presumably it’s more strain if you can hit high power at the end of a long ride, but if you can’t it’s less…? And what about the impact on recovery (vs doing intervals early, or vs not doing intervals at the end and just riding longer)?
We have some ideas on what can effect durability, outside of the elephant-in-the-room: training load. These have representations in 2.0 but would require a lot more testing and validation that these are indeed factors affecting durability. For example, there’s a difference depending on whether anaerobic power is fueled by blood glucose or muscle glycogen or some split between them. You may be able to ingest a lot of carbohydrates but if the work you perform isn’t able to be primarily fueled from blood glucose, your durability will be compromised. High carb ingestion rates have to be matched with higher use of blood glucose in order to see improvements in durability. You can possibly see the effect of this if you switch your Durability Metric to Carbs Used. If you’re seeing poor durability scores when you replenish lots of carbs, this could potentially explain it.
I took the chance to try this feature out yesterday. I put in a hard ride over 3 hrs long and added some a few all-out attacks at the end. Overall I was pretty happy with how the ride went: strong till the end and was pleasantly surprised at how much power I could still eek out. I was patting myself on the back until I ran the durability eval and this ride scored an 18.1. My ego might be more sore than my legs today. Despite that sad number, the durability adjusted MPA curve and how it reacts to efforts later in the ride is definitely much more representative of how I felt during some of the efforts than the unadjusted MPA, so there’s something to it.
Next I tested the other durability methods since the default was set to total work. The Carb based approach dropped my score down to 13.5 after inputting the correct fueling number. Not sure which is more accurate TBH.
I’ve been waiting for this durability concept in Xert for quite some time now but I’m not quite sure what to make of the outcome I’m seeing. On the one hand, the tracking of my MPA deeper into a hard ride was much more reflective of reality after running the assessment. This seems good. This is what I want. Knowing and improving this decay rate in a quantifiable way is meaningful when preparing for longer hard races. But on the other hand I’m kind of hung up on the absolute value of the score itself if normal range is 5-15. I went back in time to some of my other hardest rides over the last couple years, including races I’ve won, and all received durability scores that were even worse than yesterday’s ride and the results exhibited quite a bit of variation. Am I reading too much into the absolute value of the number itself? Is this number really ride dependent and I should be extremely selective on rides I use to evaluate? What happens if the inclusion of the durabilty adjustment triggers a breakthrough? Is that possible? How does that change fitness signature if it is possible? If it doesn’t change the signature, why not?
