Alternative to Trainer Road
Moderator: Moderator Team
-
- Posts: 734
- Joined: Sat Jul 24, 2010 9:41 pm
I don't do anything structured unless I'm racing and I build my training plan leading up to C, B, and A events.
Base training is polarized and then more specific as race season approachesHexsense wrote: ↑Tue Mar 02, 2021 7:24 pmDo you all do Polarized training year round?
Like Polarized training and then switching what to do on the interval days by base phase, build phase, specialty phase 's appropriate interval.
Or it's mostly base phase things, and not quite applicable to other phases?
We're lucky in Australia in that our climate is so mild we can ride and race outside all year round
Edit: Well apart from when half the country is on fire.....
Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓ Broad Selection ✓ Worldwide Delivery ✓
www.starbike.com
I took the plunge on WKO5................ but I think I'm gonna stick with GC. Amazingly, GC's interface is more intuitive which is saying something
Has anyone already tried this adaptive, AI based new Trainer Road?
As far as i understand, it is not rolled out for the public yet, but a couple of thousand athletes were able to use it during development.
Something similar is done by a german company named "enduco"....not sure if this is comparable, but they seem to have some nice features....will give it a try starting from next week (They offer a 2 week free test period)
As far as i understand, it is not rolled out for the public yet, but a couple of thousand athletes were able to use it during development.
Something similar is done by a german company named "enduco"....not sure if this is comparable, but they seem to have some nice features....will give it a try starting from next week (They offer a 2 week free test period)
Interested in buying Carbonsport Lightweight wheels with broken spokes.
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
There is a thread on slowtwitch discussing the "AI" based TR programs and its probably not actually AI just a "best fit" based on the huge data sample they have
So basically will be recommending a plan based on a few variables (such as training time, years training, age, etc) and matching that to data of what worked for others with the same basic variables, but it wont be actually AI based where it caters to your person situation and needs (such as a coach would do)
So basically will be recommending a plan based on a few variables (such as training time, years training, age, etc) and matching that to data of what worked for others with the same basic variables, but it wont be actually AI based where it caters to your person situation and needs (such as a coach would do)
According to the TR Podcast (https://www.youtube.com/watch?v=nAYVLch8tEA) they claim, that their algorithm actually does exactly that.
Monitor your training in "real time" and adjust accordingly.
I will look up ST. Thanks for that hint.
Monitor your training in "real time" and adjust accordingly.
I will look up ST. Thanks for that hint.
Interested in buying Carbonsport Lightweight wheels with broken spokes.
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
Curious how this AI thing is going to affect the performance especially for amateur athletes. As the algorithms get fine tuned and empowered with even more data, i guess it's gonna be a huge breakthrough in training plan offerings. Trainerroad is a huge step forward in terms of historical data of training and that's gonna help them improve their algorithms quickly i guess. Didnt know that Xert is on a similar path. Great days ahead of us... especially for us time crunched indoor enthusiasts
I guess I will pose my skepticism over TR's "AI" addition as a series of questions:
- Are FTP tests better than real world power curve data for determining training intensity?
- Is there value in individualizing interval length/intensity to a rider's specific strengths and goals?
- Can adjusting training intensity more frequently yield faster adaptations?
Basically my feeling is this is still a missed opportunity. For $20/mo I want customized workouts built around my individual strengths/weaknesses, goals and rides, as well as intensity determined from training data (so I never have to do an FTP test again). From what I'm seeing this still requires regular FTP testing and only provides suggestions to canned workouts. Maybe their workout library is extensive enough to not need to build things from scratch. Maybe they are really wed to that IP. IDK. Either way I'm skeptical.
- Are FTP tests better than real world power curve data for determining training intensity?
- Is there value in individualizing interval length/intensity to a rider's specific strengths and goals?
- Can adjusting training intensity more frequently yield faster adaptations?
Basically my feeling is this is still a missed opportunity. For $20/mo I want customized workouts built around my individual strengths/weaknesses, goals and rides, as well as intensity determined from training data (so I never have to do an FTP test again). From what I'm seeing this still requires regular FTP testing and only provides suggestions to canned workouts. Maybe their workout library is extensive enough to not need to build things from scratch. Maybe they are really wed to that IP. IDK. Either way I'm skeptical.
My view is that by joining the adaptive training schema of TR you just participate in the biggest ever athlete training survey. I believe they are gonna make good use of the data they gather, as it is clearly a very exciting opportunity with huge financial potential as well.
However, I wouldnt expect any serious changes in their training regimes from day one, not even in the first year. AI algorithms will take a while to work well and till then it's just trial n error. That's just my take of course as a tr subscriber.
However, I wouldnt expect any serious changes in their training regimes from day one, not even in the first year. AI algorithms will take a while to work well and till then it's just trial n error. That's just my take of course as a tr subscriber.
- HotWireMyHeart
- Posts: 37
- Joined: Sat Nov 07, 2020 10:46 am
It's not like they're starting with no data at this point and TrainerRoad will need to implement the adaptive stuff and then work on the adaptive product with the data they get. They already have a huge amount of data- another year isn't going to tell them anything new. Their algorithm either works or it doesn't when they launch. Though they can tweak it after customer feedback as time goes on of course.
TrainerRoad has been one of the least sophisticated training systems for years, well behind the Xert/WK04/etc curve. Pun intended- because if they're still basing everything off that one point on an athletes power curve and ignoring differences in lower threshold power and maximum sprint etc then they're still off the pace for me.
TrainerRoad has been one of the least sophisticated training systems for years, well behind the Xert/WK04/etc curve. Pun intended- because if they're still basing everything off that one point on an athletes power curve and ignoring differences in lower threshold power and maximum sprint etc then they're still off the pace for me.
Well, than maybe you should have a look at their podcast, where they explain what they are doing.
It seems, that they actually measure/determine your capability in 7x3 dimensions....should that be enough?
Will look into Xert as well
It seems, that they actually measure/determine your capability in 7x3 dimensions....should that be enough?
Will look into Xert as well
Interested in buying Carbonsport Lightweight wheels with broken spokes.
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
BMC Teammachine SLR01 Disc Team 2018|Ridley Helium | Kuota Kross|Cannondale Scalpel 29 Hi-Mod
The issue is that some believe that what they say they are capable of in the podcast and what they are truly capable of delivering may be substantially different.
Have a read of this thread on slowtwitch and decide for yourself
https://forum.slowtwitch.com/forum/Slow ... ead#unread
Yes, they are def underutilizing the dataHotWireMyHeart wrote: ↑Thu Mar 11, 2021 8:30 pmIt's not like they're starting with no data at this point and TrainerRoad will need to implement the adaptive stuff and then work on the adaptive product with the data they get. They already have a huge amount of data- another year isn't going to tell them anything new. Their algorithm either works or it doesn't when they launch. Though they can tweak it after customer feedback as time goes on of course.
TrainerRoad has been one of the least sophisticated training systems for years, well behind the Xert/WK04/etc curve. Pun intended- because if they're still basing everything off that one point on an athletes power curve and ignoring differences in lower threshold power and maximum sprint etc then they're still off the pace for me.
I feel like there is so much more potential. For example I would love to be able to plug a ride or route into an app, and have the app analyze the route in the context of my power curve and FRC, and develop individualized workouts and even a training schedule to get stronger at that ride. The same could be estimated for different kinds of races and events as well even if you don't have the actual route or past performances to judge on. That would be worth paying for and to me seems like something that's possible based on FRC/power curve data.
FTP = 20mins steady maximal work * 0.95. The problem of FTP is that it has an assumption of the same capability/capacity to work above FTP (i.e. FRC or HIE, or W' capacity, similar ideas) among all riders. If two riders have the same FTP, would they necessarily sprint the same?cheapvega wrote: ↑Thu Mar 11, 2021 6:02 pmI guess I will pose my skepticism over TR's "AI" addition as a series of questions:
- Are FTP tests better than real world power curve data for determining training intensity?
- Is there value in individualizing interval length/intensity to a rider's specific strengths and goals?
- Can adjusting training intensity more frequently yield faster adaptations?
Basically my feeling is this is still a missed opportunity. For $20/mo I want customized workouts built around my individual strengths/weaknesses, goals and rides, as well as intensity determined from training data (so I never have to do an FTP test again). From what I'm seeing this still requires regular FTP testing and only provides suggestions to canned workouts. Maybe their workout library is extensive enough to not need to build things from scratch. Maybe they are really wed to that IP. IDK. Either way I'm skeptical.
If TR "AI" workouts are based on FTP, it is flawed fundamentally, or the "AI" is not I enough to understand different riders and prescribe more appropriate workouts based on rider's goal (to sprint better or to climb better).
Rikulau V9 DB Custom < BMC TM02 < Litespeed T1sl Disc < Giant Propel Advanced SL Disc 1 < Propel Adv < TCR Adv SL Disc < KTM Revelator Sky < CAAD 12 Disc < Domane S Disc < Alize < CAAD 10
Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓ Broad Selection ✓ Worldwide Delivery ✓
www.starbike.com