We generally don’t recommend doing this because it’s too easy to miscommunicate information to the software.
I would advise you not to get preoccupied with bumping up your Progression Levels. Even if a Workout doesn’t impact your Progession Levels, it will still be considered by Red Light Green Light and it will be considered for your AI FTP detection result.
Your Progression Levels ensure your TrainerRoad workouts are prescribed at a precise intensity so as to meet the intentions of the Workouts and more broadly, your Traning Plan. When you manipulate the software in a way that is not intended, you interfere with this. For that reason, it’s best to let the specific prescribed TrainerRoad workouts inform your Progression Levels.
I want to be sure not to discourage you from riding outside though! There is lots to be gained: fitness and otherwise!
I would advise you not to get preoccupied with bumping up your Progression Levels.
Your Progression Levels ensure your TrainerRoad workouts are prescribed at a precise intensity so as to meet the intentions of the Workouts and more broadly, your Traning Plan.
I edited it as I understood you meant only a “bumping is PL” isn’t a concern.
Could you elaborate a little more on what is the relation of this? I think that is my issue. I don’t feel comfortable knowing that the PL used as a base for my workouts is not accurate, as it seems that my next workout/plan isn’t reflecting the correct status/fitness/ability.
Hey! I’m not totally sure I understand your question. But let me try to do a better job of explaining what I mean to see if that helps.
At the moment, unstructured Outside Workouts won’t affect your Progression Levels. This is because we don’t have a really accurate way of quantifying the work you have done during them as it relates to time spent in specific zones and how that influences specific aspects of fitness (yet), at least in comparison to the carefully controlled TrainerRoad Workouts.
Therefore we don’t include unstructured Outside workouts in the calculation of your Progression Levels. This ensures we apply the principle of Progressive Overload in a very deliberate and precise way, that is not skewed by random inputs.
I’ll try to give you an example so that it might be more precise.
One did a group ride randomly, a hard one, 50min TiZ FTP pretty consistent. My PL “understands” that I’m only level 2, so the workouts given to me are “easy”. Let’s say a 4x6 @ 97%. In this case, in order to have a progressive load, that person might need more than 4x6 @ 97%.
So, what I meant was, without considering what one did “unstructured”, how can I be sure that my workouts are accurate in relation to the level of fitness I have? And I know your first answer will be “Do the 4x6 97% and rate it easy, AI will catch up and adjust”.
Yes, but it might take 2, 3 weeks. Or, I might’ve made another hard effort again, and my plan will always be behind. No progressive overload.
For now I’d be content with a WL/PL 1.5 system in which the “AI” is able to read output of trainerroad workouts done inside and change the levels and type accordingly. This won’t effect people who use erg mode for everything and do all workouts by wrote without increasing/decreasing intensity but for those who put a workout up or down by % during or ride in resistance mode and perhaps overshoot the targets I wouldn’t mind knowing what that PL would be. I’m reminded of an old Coach Chad saying from the podcast in which people asking about increasing/decreasing intensity or rest intervals should consider the question “are you changing the nature of the workout?”
As an example let’s say a user wakes up one morning after a night of unbroken, sound sleep and feels amazing. Some how everything feels like it’s falling into place for nutrition, timing and low fatigue. But they’re wary of changing the workout whether it be because they’ve felt similar before, tried something harder and got burned or are just tentative about going against the plan put forth. So let’s say the workout is Vancouver-1, 3x15 @ 92%. They get going and partway through the first interval they give the intensity a nudge, then another. Next interval they bump it some more. By the end of the workout it looks like they could have done Rainbow -1 judging by the output with higher TSS and KJs. But they would still only get credit for a 6.4 sweetspot workout with a minor nudge to threshold where at they pretty much completed a 5.4 threshold and didn’t get credit. And I’d say just telling the post-ride survey it was easy/moderate doesn’t tell the story that the rider did threshold instead of sweetspot. At the least if the workout credit isn’t changed then the AI should in addition to the sweet spot level give a larger bump to threshold PL if a good portion of the workout was done at that output. The same would go for burst workouts like taylor or gendarme, maybe even Bluebell. It’s easy to overshoot on these in standard mode and not kill yourself so why wouldn’t the AI read the higher output and determine it deserves a higher PL?
I figure TR would benefit from this middle step where the AI would have something to compare to with the set workout and judge levels accordingly versus jumping straight to trying to figure out an outside unstructured ride with no basis for comparison other than a set FTP and an estimate of the times spent in zones. I think they’re most worried about the message board filling up with posts that either read “I just did (enter hard outside ride type) and my levels slightly/didn’t move” or read “I just did (enter hard outside ride type) and my levels went through the roof and AT is suggesting workouts I can’t do.” And honestly it would just plain feel like progress at this point since I would agree it feels like this system has stagnated. No need to argue that things are going on behind the scenes because I’m sure they’re trying super hard to bring updates but outwardly it’s been the same WL/PL system that we’ve had before.
if there is some recognisable structure to my unstructured ride that I think will translate to PLs then I sometimes retrospectively create a TR workout that matches what I did and then link the ride to it. A recent example of this was joining a friend who was doing hill repeats, didn’t have time in advance to find a suitable TR workout, but hit my lap button at the start and end of every repeat, did a quick TR workout to match to those times and numbers, and got a big jump in VO2 PL as a result as I hadn’t done any TR VO2 workouts in quite a while
If it’s just a hard group ride with lots of time in zone I’ll be less inclined to try and replicate it and will instead just make a judgement call on what level I can handle next time I pick a TR workout. Might mean picking a Stretch or even Breakthrough workout, but if you get it right it gets you to your “correct” PL more quickly than the 2-3 weeks you mention. Does take some experience though and on occasion I’ve got it wrong. I do find that it’s much harder holding steady ~threshold watts solo than it is amassing time at ~threshold in a group which typically involves a lot of micro recoveries
If the unstructured ride is the exception not the norm and most of my hard training is on TR (I never bother with outside workouts for endurance rides) then I will also generally happily just let TR catch up over those few weeks. Always better to slightly underdo the training than slightly overdo it. The 2 examples above are what I do when I’m doing a lot of unstructured (or at least non TR) rides and then occasionally dipping back into TR structure
This is one of the few drawbacks of what is overall a very good adaptive training program by TR. I am hoping TR addresses this stuff soon for new riders. When you’ve been doing it for a while, you kind of know what PL to start with.
Just wait until you are smashing level 8 V02 workouts with intervals at 130% FTP, and AT still thinks your anaerobic PL is 1.2.
You can also manually adjust if you feel like you have gained fitness during unstructured outdoor rides. But in my experience, unless you are talking about endurance rides, unstructured rides are unlikely to push one particular PL close to your current level.
I think of structured workouts akin to strength workouts that target one specific muscle group. Unstructured outdoor rides are analogous to composite movements that tax many muscle groups. Their benefit is to work on the coordination rather than pushing one of them to the limit.
There’s no analysis of out doors rides at the moment (No WL V2), its only your Pass/ fail survey that changes the PL’s, The PL increase was embedded in the workout and subject to your response, and nothing to do with the spike.
Super snarky mode on - TR is busy adding weight training enhancements, cause that is what Nate is in to now. - Snark mode off.
Don’t actually believe this, but all the hype around the weight training kind of put me off. Anyway, I am back on TR after a many year layoff with self coaching. Have a broken clavicle, so inside for awhile now. But once healed I do most of my longest and hardest training outside. I often do structure within or as part of a long ride on the weekends. If TR can’t recognize those, and my prescribed workouts are always behind, I guess I won’t be long for TR this time.
If post workout feedback is new since you last used TR, then it might help the adaptive training system to take those outdoor rides into account a little better than previously.
The theory being that if those outdoor rides are so effective to your training, then when you come to do a prescribed workout, your post workout feedback input will be that the workout was easier than expected, so the system will crank things up accordingly for future workouts.
I doubt it can get it spot on right every time, but it is another input that helps the system to adapt.
I think this is slightly too pessimistic view. TR has said that WLv2 was quite near to be released but real life testing did not prove model behind it being fully correct (or something in those lines).
So, lets consider outcomes if TR would have released WLv2 regardless and keep fixing it while we can already enjoy work-in-progress:
you get less than optimal workout: you don’t progress as fast as you’d like
you get overly ambitious progression and as result, will burn out
As a customer, I think they did best they could in this situation:
release proven stuff (RL/GL), which does work with unstructured rides;
restart internally with other stuff to get fully validated product in the end
and while focused team works on “restart” project, other teams can meanwhile still deliver unrelated features. For those saying “throw all devels on WLv2 refactoring”, increasing team sizes rarely improves progress and more often actually slows development down.
To add to that: the “restart and double up the person power” suggests that lack of dev time was behind it when it may very well be that TR found out the (reasonable) approach they have had was missing something. Understanding that something doesn’t work and why something doesn’t work are two entirely different things. Explaining the why means you have gained insight into the problem space and that could facilitate a solution. Or tell you that a solution won’t be forthcoming anytime soon. (Think self-driving cars it is an “easy” problem if all cars on all roads are self-driving, and humans are absent.)
I’m glad that companies have the guts to not ship something, especially after its existence had been disclosed.
Come on… people have asked for various levels of consideration and/or integration of strength training in TR since about the creation of the forum. I get that it’s a feature not needed by everyone, but this wasn’t something out of left field either.
This is going to be TR’s problem going forward: each new / incremental feature / enhancement will be wanted by a slice of its user base, while other slices would have preferred TR to work a different feature / enhancement.
For example, there is a small slice of us who would really like to TR to come out with a modern Workout Creator to replace the existing one. TR has basically said this isn’t a priority.
So everyone has to decide if the current TR feature set meets their needs better than the alternative. If so, then use TR. If not, use the alternative that better meets your needs today.
Well that is where we are now, regardless….at least if you ride your bike outside for any decent amount of time.
PLs work fine if you constrain yourself to only TR workouts. But once you start to add in outdoor rides without structure, they fall apart and don’t accurately reflect the work you have been doing.
The fact that it has been 3+years and there is still no way to accurately reflect work done outside of TR is all the evidence you need that it is a flawed concept.
I simply ignore PLs for the most part now…I’ll pay some attention to them once I get back to “trainer season” (Nov - Apr), but once I start to get outside, they have no value to me.
