I hear what you’re saying and I totally appreciate the points you’ve raised. Maybe I’m looking at those points from a different angle?
Personal coaching might have many positive aspects and I’m not in a position to discredit it. What I would point to is a very similar progression to possible burnout, regardless of how the training is prescribed. We, the athletes are almost always the problem.
It has been mentioned on this forum that personal coaching is based on the relationship. That relationship is built on trust and your personal goals. If your coach is setting you workouts that smash you up and down but, your too proud or stubborn to admit that something is not right, you’re the problem. The coach is (I’m assuming?) eventually doing to pull you up because your numbers are all over the place. That coach is going to be thinking ‘why didn’t this athlete say something…?’ Yes, the coach set you the work and it was too much but it’s your responsibility to raise your hand and say ‘hold on a minute but…’
In my mind, we, the athletes are always the weak link. We have to let go of our numbers and our egos to achieve our goals. There’s no shame in saying ‘right now, I can’t do that.’
I also think that experience is a double-edged sword. We get annoyed with ourselves when we can’t finish a workout we did two or three months prior. Why? Why do we beat ourselves up? Maybe you were absolutely flying two months ago.
Yes, I know that I sound harsh. I just can’t get onboard with all the X said I should do Y, even though I knew I was feeling Z thinking that’s floating around. We have to be accountable, regardless of which training platform we use.
I’ve just noted on the DJ thread about “ego”. It’s invariably that.
Just because there’s a sledgehammer in the toolbox doesn’t mean to say you have to use it especially if you are only trying to fit a pane of glass.
Part of the point of having a coach is that they are a dispassionate observer, if they are experienced they’ll be able to spot symptoms before you do, they aren’t working from your N=1 but N=100 or whatever. They’ve seen things before.
At the moment TR is “dumb” and all the subsequent analysis is on our shoulders. AT should lift some of that burden but it still won’t know about the kids waking you up in the middle of the night; your mum being taken to hospital; your boss being a PITA; etc. that you would be able to discuss with a real life coach, so there’s still responsibility on us.
Even a real life coach won’t prescribe you a “perfect” plan from the outset, you’ll get a standard plan adjusted according to whatever circumstances and goals you’ve told them. Your feedback will help the coach focus in on what you need to improve on and the plan will alter.
I think you’ve clarified your points perfectly. I can’t say that I disagree with anything you’ve said. That said, your reference to that balancing point is really interesting.
Lets take a few things that we on the TR forum accept as known. Lots of people have little or no experience with Structured or Interval training. Six hours per week out on the road isn’t the same as six hours of intervals. Mid and High Vol Plans do, for most athletes, need to be built up to. How many other things in a sporting context, do you jump straight into the deep end? Yet how many people post on this forum saying that they’re new and jumped straight into a Mid Vol Plan. Quite a few. Why do that?
If they’ve used Plan Builder and been honest with their inputs, then yes, there is an issue.
So why, with regards to cycling and specifically TR, are people so determined to find their balancing point ASAP? If you lack experience, how can you gauge that? My next thought is why do some athletes need to balance right on the very edge? Do the possible gains outweigh the losses if it goes pear-shaped? How long can you realistically balance on that edge?
Slow (Low Vol) and Steady (Consistency) really do win this race. It’s like learning your on-bike craft. Make yourself a well rounded athlete. Take the time to learn about yourself. Enjoy the journey. Experiment.
Reading the last few posts here makes it clear to me that there is little consideration of other’s experience outside one’s own here. In effect, “It’s easy to do what AT is going to do, all it takes is experience with structured training.” Maybe, but then deprecating those without experience making ill-advised decisions ignores the fact that there is no other way to go about gaining experience. Experience is experience. Expecting people to jump to the experience level of someone who already has it figured is not limited to cycling but also makes little sense when we stop and examine our own assumptions.
So how to get people more experience with structured training in a volume-appropriate and consistent way so that they can avoid all the wrong turns and road blocks usually encountered by people starting to train? Enter ML and AT.
The above aside, my personal opinion is that “make it easier when you know you should” or “make it harder when you know you should” is a fallacy. If this process were so simple or clear, then coaches would become obsolete because athletes would be able to just do it themselves. This is obviously not the case, though. Plus, the suggestion that AT is going to do what a coach already does is, in my opinion, exactly the point. Coaches, and good ones at that, are the exception, not the rule. Millions of people worldwide have precisely zero access to a “good coach.” Developing a ML tool to allow a personalized level of coach-less training and making it available at scale is truly a new paradigm. We won’t even go into the fact that a ML structure could potentially make better recommendations than a coach.
I feel like the TR crew might need to open their next podcast with a Mitch Hedberg quote, “Ya know, you can’t make all the people happy all the time…and last night, all those people were at my show.”
Oh, well… If the only task of your coach is to tell you that you failed a workout then I totally agree with your point. We can definitely replace that with algorithms.
In my opinion the problem with AI training right now (and that doesn’t only goes for AT) is that you’ll have an incredibly hard time to actually quantify load of a session. After you have that you’d have to see how someone reacts to that load. I don’t think this is simply covered by the fact if you can finish such workout or not. Then, you’d need some kind of physiological model of your athlete and check if the load and reaction of that model goes into the direction you’d need for your target event. If you have that you can possibly compute an adaptive plan that leads you from the starting point to your race.
As far as I know no “AI” comes even close to that workflow and I believe it’s partly because there just isn’t the data yet. No doubt I’d love to try out such system and you can easily imagine with loads of users it could lead to better recommendations than a coach.
I do pay attention to IF but it is pretty unreliable for me. I think the end goal in a year or so is for AT to know better than you do what workout at what intensity will lead to the most progress. I agree though, to start it will probably function as a tool to help reduce failure rates
Edit: imagine a world where you feel fresh and ready to nail a workout but based on a million other rides AT thinks you should back off and take it easy and that ends up being the correct move to maximize gains. It’ll happen at some point, just a matter of when
I don’t think that anyone at TR is selling AT as an algorithm to tell you when you failed a workout. If we are all being real about what a coach does regarding the fitness portion of training, it is, at it’s essence, data analysis, and primitive at that. In no way am I saying that coaches have no utility, only that there are better data analysis tools out there. In my opinion, coaches should be able to focus on the more human, psychological, technical aspects of training that are all-too-often neglected, and leave the fitness/data analysis portions to the ML experts.
My understanding after listening to @Nate_Pearson on the podcast is that this is precisely the expectation - use the vast, untapped powers of ML analysis to better quantify the training load not just based on subjective feedback, but on actual power numbers, workout completion, power zone progression or regression, etc, with a goal of building a physiological model of each athlete. Nate mentioned on the podcast that they had identified over 100 individual factors that played a determining role in an athlete’s performance. As we know, there is no human coach on earth that can match the data analysis powers of ML, especially at this scale. Finding patterns in a sea of data that have real-world applications is something that hasn’t been done, and something that will require us all to shed our assumptions about cycling training.
100 million workouts is probably approaching the the point at which the data is there, but I am mosdef no expert
I think you’ve misunderstood my post(s) and that may be my fault.
You are however, correct, when you say that I have little consideration of others experience. Training for me is individual. I have to find what works for me. I tweak, I fail, I tweak again and maybe I fail again. What I am doing is gaining experience. That experience is personal to me. My lifestyle, my life stress, my goals etc, etc. This is why I wouldn’t dare try and tell another athlete that their training is less optimal than what I do.
What I’m not doing is looking for someone or something else to blame.
Compared to many on this forum and cycling in general, I’m still green behind the ears. I have limited structured training experience and if I’m honest, I bet a lot of that experience would highlight just how wrong I’ve got things in the past.
Not for one moment do I propose that I can do what the latest offering from TR can supposedly do. I’m in the super excited camp. That said, I think that I’ve finally found a way to make TR as it currently is, work for me and my goals.
What I don’t understand and I doubt that I ever will, is why a certain section of new TR users believe that jumping straight into a High Vol Plan is a good idea. If you have little to no experience, I can’t see how you expect this to end in a positive experience.
TR staff (Chad) confessed that there’s no evidence whatsoever that that plan is any good, but people requested it. This is one of those instances where you see comercial interest getting in the way of proper training.
For some it works for some it doesn’t. It all eventually comes down to how good you know your body. Dogmatic advice won’t be of much help there.
Obviously a personal anecdote but I am one of those users you don’t understand. I started off the couch with mid volume and after one or two plans went for high volume. To both mid and high volume I added and add endurance riding and also running. Going for low volume never was an option for me. I had and have 10+ hours to train and those I want to use. The resulting 500-700 TSS is not out of the world. Some guys do that on the weekend climbing some peaks in the alps.
I can’t say that it works. I can’t say that it doesn’t work. Even users who have access to the beta can’t say that it works…or that it doesn’t work.
We know some tantalizing details. None of us can say what the real deal, practical implications of the technology will do IRL.
The two blog posts about adaptive training posted on 25feb by Meaghan and Sean serve to frame the approach pretty well, IMO. Workouts are going to be sorted into buckets (already are if you check) according to the goal of the workout. Workout history will be ranked by the same buckets. Workouts in a given bucket will be ranked by how challenging they are. If your last threshold workout was too hard…well then the next time you see a workout in that bucket it might make sense to go down the difficulty slider to a workout you can handle.
So…no machine learning in sight…that’s already a great improvement. I like that idea. None of us need a muh cheen to tell us that sounds like a good approach.
The ramp test is a great diagnostic test…but it is almost true by definition that it gives many riders the wrong FTP. The ramp test gives you your MAP. Your FTP is estimated as a portion of that MAP. @Nate explicitly said on the podcast that FTP is distributed around 75%. If you’re a rider that is a few standard deviations away from the mean that’s bad news.
That’s one of the issues TrainerRoad is targeting for improvement with this approach…not just improvement…it’s probably closer to the truth that they have resolved that problem. We’ll see.
Potentially, this could be a material improvement to TR software. I’m cautiously optimistic.
Sorry but this is false. TSS for 5 x 10 and 1 x 50 are the exact same for these intervals. Factoring in the rest periods in a 5 x 10 workout will likely give more TSS, but if you really want to compare apples to apples, you would have to add rest time and intensity to the end or beginning of the 1 x 50 workout…giving the exact same TSS.
Fair point. If something like that happens to me eg. 6x4min VO2max is too hard, I’d either just skip one or two intervals at the end or choose one 6x3min. I agree that there is a benefit if I don’t have to search for it in the library.
Ehm… you can’t just throw data at ML and it will tell you all what you want to know.
If you want to make a model based on your power numbers you’d need regular all-out tests in various time ranges. That’s what WKO4/5 does for years. It suggests you interval duration and power based on your PDC. However, if you don’t maintain your PDC it obviously won’t work. This might be different for others but I find it hard to integrate 3-5 different all-out sessions every month just so the algorithm can tell me what I know by doing the workouts anyway.
Serious question. How do they know about your performance?
Do you mean FTP derived from a ramp test?
Analysing a power duration curve?
Power file from a crit and the result?
I was refering to features of the available dataset not the amount of workouts.
Agreed, unless you have built a model at which you can throw data, then yes, you can throw data at it and it will tell you exactly what you ask of it
Again, one of the points of the AT they bring up in the podcast - there is no need for these periodic assessments, after which your workouts are static/benchmarked until the next assessment. How long is an all-out assessment accurate? A day? A week? Not much more for sure. Thus, using more advanced data analysis to track your progress and fitness continuously is useful. As they repeated several times on the podcast, the right workout on the right day based on your entire training history. Not just your most recent PDC, ramp test, 20 min test, or what-have-you.
Nate specifically mentioned this as proprietary, thus we will likely never know. It’s not hard to imagine though - some combination of ramp test results, indoor/outdoor workouts, outdoor rides, etc, going back “X” number of days. Or forever.
It’s interesting to read these comments (the whole thread, not just one or two comments/posters) and realize how difficult it will be to get people out of their own limitations and into something new. I am not an early adopter, but to me this seems to be so obviously superior to other training platforms, including TR’s current. Probably quite obvious from my posts, but hey, I should probably mention it
But to my simple mind you are now comparing a 40 min wko with a 55 min? Throw in your 15 min z2 before or after the interval in example 1 and you get the same IF?
I’ve been thinking (uh oh). I would be really hesitant to adopt Adaptive Training until other rides/workouts can be accounted for. I really don’t think it should be publicly released until this is available.
I’ll get this out the way - I’m training for an Ironman. However, I’m not necessarily asking for the runs and swims to be supported off the bat. It’s a cycling platform that offers very basic multisport stuff as an aside. I know that, I can’t wait until it’s integrated more but until then I’ll keep this cycling focused.
Since it’s announcement, I’ve been more mindful of my training. My worry is the AT system is being beta tested with TR staff and, as they said, specific TR users “you probably know who you are”. Surely this is an oversight. Maybe this is still very early beta but the way I see it, to prove your concept you’re going to need a much more diverse beta group other than just hardcore TR users.
I point this out as I, like a lot of people, have been using TR for my structure but using zwift and going outside for races and greater interaction. Just last night I did half of my plan builder prescribed workout intervals before I jumped on to Zwift with my TTT team for a course recon and finished my workout there. Yes, not as structured but it was a similar effort. As I understand it I’ll need to justify why I stopped that workout early with the new platform and it will in someway effect future workouts until the zwift ride that I did straight after is accounted for. The same goes for the long rides outside. Am I hell doing 3+ hours on a turbo, just not going to happen.
As far as I understood the timeline, it’s closed Beta (full AT system), mostly for basic bug testing and other obvious issues. Then, once they grow more confidant about stability they keep adding more and more users to the beta (which should also be more diverse).
But for the public release, as I understood it anyway, it won’t be switched to full on AT overnight at some point. Instead they will release smaller subsets of features slowly one by one. Presumably the full system will be active behind the curtains, but I guess the effect on the user will be limited at first.
