Have you tried any of the SMART workouts, by that I mean:
“Workout players also use advanced mathematical models, in real-time, to adjust either interval power or duration to assist the athlete in meeting the goals of the workout”.
Thanks
Have you tried any of the SMART workouts, by that I mean:
“Workout players also use advanced mathematical models, in real-time, to adjust either interval power or duration to assist the athlete in meeting the goals of the workout”.
Thanks
I’ve seen it recommend openers with an event date several weeks into the future. I was just playing with it a few days ago with that new Continuous feature, and it recommended 4 openers in one week. Keep in mind, when I say recommend, I was clicking “automatically recommend” in the Planner. That’s the other thing I find odd. Xert often “recommends” a workout using that button on the Planner that is completely contrary to what it recommends on the Training Page, on the left side bottom, the verbiage “Your current training status is fresh and you should consider…”. I guess where I’m at with Xert is I have dwindling confidence in it. It started with first impressions (they’re everything, right?) from the helter skelter UI, and it’s gone down from there.
Manually search. Compare Xert search to TR’s. It’s like the designers didn’t even try with Xert.
No, but it will still “recommend”
“UI is not a strong suit.”
The ability to search isn’t UI. It’s function. There’s no filtering in the workout page. You can sort. You have one text field you can search on, which often doesn’t bring up the workout that I know has the word.
I think that’s incorrect. There’s filtering for the recommended workouts on the Training page (it’s not sticky), but on the Workout page, the master list of workouts, there is no filtering. There is sorting, and one text field.
We’ll have to agree to disagree. A search box and sorting columns is not filtering. After a proper filter is applied, one should be left only with workouts that meet the filter. As for the search, ok, maybe that’s a filter, but it’s “weak.” I don’t even think it brings up results that I KNOW are there.
Nope, but you’re starting to sound like The Religious. I am explaining what contributed to my lack of confidence in the program. If the Xert gods can’t implement a proper filter, what else can’t they do? That’s where I am, and that’s why I started the thread. But…Thanks for your contribution…
Ok, it’s “usable”, but it shows a lack of attention to detail. Again, I’m just stating the impressions that I receive…they’re not confidence inspiring. Did you notice that when you click to go to the next page, you have to scroll to the top each time? Certainly not a huge issue at all, but I’ve never seen that in a professional website. It’s shows a lack of attention to detail. It’s like meeting the attorney, and first thing you notice is that he has a big stain on his tie and his shoes aren’t polished. He might be a great attorney though.
Wrapping it up…Thanks for all opinions…and help understanding features.
Cheers for the badge but I started using Xert only a year ago. Not sure if it’s the right badge
I do find Xert an interesting approach and I’m keen to understand it more… but still trying to grasp some of the Xert concepts as everyone else.
Been there, done that. Stay away.
So much back-end tweaking required you need a science degree to understand what’s happening. Designed by nerds for nerds.
That beginners guide doesn’t teach you how to use Xert well. The users and owners are constantly admitting it’s hard to understand and that it takes a lot of work to learn and understand. Many even recommend paying a coach. Tuna himself says below something like, “I’ve been using it for a year and am still learning”. That’s my argument in a nutshell. Also, “if you join a group ride at 2 in the afternoon on a work day you can get some advice” is not a helpful model. They’ve been asked many times to record those sessions so you can watch them at a time that is convenient.
There is a great backbone here. I love the idea. I love the FTP estimation and find it very accurate. I really like some of their workouts, including the LSD series and the Ronstadt stuff. The Garmin Connect fields are great. I just don’t find myself having the motivation to sit down and take notes through 25 podcasts or take a day off work to get some live advice when there are so many other user friendly apps out there and business that understand the need to focus on the consumer. I paid for a year, but I’ll be letting my subscription expire.
Yeah, totally agree on everything you said.
Eric Min from zwift was also seen trying it out about 18 months.
That’s actually insightful. Thanks for sharing.
Whilst I agree that the interface could do with a bit of a fixing up it’s much better than it used to be now that most things are available from the tabs on the My Fitness page. To me it makes sense, but I’ve used it for a few years now…
When Adaptive Training was announced there were a few folk saying that TR were just copying Xert etc, but really they are fundamentally different approaches that, in my opinion, is driven by one thing: the number of users that don’t have power meters on their bikes.
To get power duration models like Xert’s (or WKO for that matter) to work well you need to feed it with a lot of data - much easier if you actually have power meters on your bikes.
Beyond the power duration modelling you need to have a good data to drive the fitness and fatigue modeling, which in turn drives the workout recommendations.
I wouldn’t even like to take a stab at what percentage of TR users have power meters on all of their bikes, but for the more casual user, I would guess that most are limited to the trainer and a large number of those will be virtual power.
This is the reason I believe TR went down the road they did with machine learning and classification of workouts coupled with feedback forms etc - they just don’t have a high enough percentage of users who could make a robust power duration model work. Progression levels and ‘energy systems’ are a fudge for a system that can’t handle power duration modelling. They even had to add another 600+ workouts to the library just to fill in some gaps.
Personally I think machine learning is a bit of a sledgehammer for the job and that’s why we’re not seeing unstructured rides being classified yet - it’s a nightmare of noise to try and assign to a single ‘energy system’ and then rate how well you performed. Xert’s approach to this is very elegant - here’s what you did and how the three energy systems contributed to that performance and if it showed that you got better then you get credit for that.
Mike
Probably didn’t think it was worth the asking price.