šŸŽ‰ šŸŽ‰ šŸŽ‰ Introducing Adaptive Training! šŸŽ‰ šŸŽ‰ šŸŽ‰

Thanks Chad. I chose ā€˜too intenseā€™ (as thatā€™s how it felt!). Be interesting to see what happens to PL if Iā€™m completing 5ish level SS workouts and rating them hard.

1 Like

So at the very least I can just add the same workout on the day I want to do it and still have the downstream AT effects. Thanks

2 Likes

Anecdote:

I did a 6000ā€™ climbing day on Saturday, 4000ā€™ Sunday on my 40 pound enduro bike. I do have a PM, but I didnā€™t notice the TSS as I donā€™t care. AT didnā€™t care at all about those rides. I still had the same workout scheduled on Monday.

2 Likes

Was away from home the past 2 weeks, so no vlogging while I was away, so caught up today with a video on my past few weeks. Managed to get a full training load using a stages spin bike since I didnā€™t bring any of my bikes. I had access to some primo gravel roads too, which was sad lol

Is there a way to tell AT to ignore a particular workout? I have a couple of custom workouts that have really messed up my PLsā€¦ For one I went back and changed the survey result to ā€œvery hardā€ but it didnā€™t seem to do anything, and now I am out of the 7 day window. The other one was a fail, but maxed out the PL anyway.

1 Like

This is such good news, loads of us only train indoors and donā€™t care about outside ā€˜issuesā€™, it seems like AT works great for us and weā€™re still waiting to get onto the beta. Happy days and letā€™s hope new update doesnā€™t cause too many bugs. Cheers for update Nate.

2 Likes

Hi All - sorry as always question may of been answered beforeā€¦

Today I did a ā€˜Productiveā€™ vo2 max workout (Baird +2 modified in workout creator to add some z2 at the end for 30mins).

It had a PL of 6.7 which was a step up of +.3 from my last vo2 max workout, Kern +4 and as such I was expecting to see an increase in my overall PL level for vo2 maxā€¦but the Baird +2 workout today shows a negative -2.1 PL adjustment.

I am not on a plan (using TrainNow and self selecting workouts if that makes a difference).

Is this normal behaviour and if so what am I missing?

Thanks,

I had a very similar issue with a Progressive Starr King Threshold workout today (negative rating, and an Endurance progression). I just heard back from Support and there is an issue that popped up overnight and they are waiting for the Developers to have a look at it

1 Like

Thanks @PWeber - please keep us posted on the outcome as could be a similar issue as you note.

I had McGregor +1 (3 x 20 @ 92% with 1-minute breaks during each 20-minute interval) on the calendar today with my usual modifications to lengthen the warmup and more intensity during longer recovery intervals. Due to three days in a row of below average sleep I had to dial it down to 88%. AT still rated it as a pass even though I was expecting to get the ā€œstruggleā€ survey at the end. Also, it shows as a Breakthrough tempo workout (probably because I dialed it down to the low end of sweet spot) but downgraded my tempo PL by -0.2.

My sweet spot PL has been at 8.3 for a while even though I havenā€™t completed a sweet spot workout at that level in three weeks (26 June). Now, in fairness to AT, I inserted a recovery week due to family visiting and did a longer outdoor ride last weekend instead of the prescribed workouts.

No complaints, just find it interesting. Below is a screen capture of the completed workout from the TR app as well as two custom graphs that Iā€™m using to track changes in my PLs over time (both show the same data, just in different formats).

1 Like

@IvyAudrain
I accepted the beta invite and have been dutifully following a training plan, making use of AT, etc

Is there an ā€œofficialā€ venue for feedback, or is it sufficient that I simply use the product and you can collect the usage results on the backend?

Want to make sure Iā€™m actually contributing to the beta, rather than just using the features!

I do this as well. I often also add TrainNow workouts if I am looking for a little extra work that week or day. If you move your scheduled workout, it will no longer be adapted or changed if an adaptation is accepted. However, as I understand it, once accomplished, it will contribute to future adaptations like adding a workout should do (again as I understand it.). If I move a workout I do so at the last minute. If I skip one I will drop a planned day off on top of it at the last minute.

2 Likes

If you find any anomalies / bugs, you should report those to support@trainerroad.com for them to look into

1 Like

Iā€™m sure this has been mentioned (and Iā€™m equally sure Iā€™ve read it).

Started a new plan today. Added to calendar via Plan Builder, no shuffling of workouts, all workouts being done outdoors.

Did Ritter (Threshold 5.2) and pretty much nailed it - the ā€œoversā€ in the first set were slightly low as I eased off before the interval ended but those in later sets were all well above the target power.

Got home, saved the ride and it was uploaded to Garmin then pulled into TR. I refreshed the Calendar via the website and was offered adaptations. AT knocked 0.8 off the level and changed the next workout. At this point I hadnā€™t even been offered a survey. When I opened up the ride, I got the failure survey, chose ā€œI didnā€™t struggleā€ then selected ā€œModerateā€ from the pass survey. No further adaptations were offered.

Is there an assumption that the survey (pass or fail) has been completed prior to accepting adaptations? I donā€™t use the app away from the trainer so it may be that this is a ā€œraceā€ condition in the calendar: refreshing the calendar instigates the adaptations but without refreshing it you canā€™t get to the workout to fill out the survey to say you didnā€™t fail!

without seeing the power graph my guess is going well above target is the reason for fail survey. And since I joined the beta several weeks ago, it appears that once AT deems it a fail, selecting ā€œI didnā€™t struggleā€ will not update PLs.

(side note: IMHO fail is the wrong term)

speculating based on your reportā€¦ it appears that once AT deems a workout fail then it proceeds down the adaptation path without regard to the survey.

ā€œWell above targetā€ was about 7%, hereā€™s the workout: https://www.trainerroad.com/app/career/bobw/rides/114291833-ritter, ignore the big hill at the end, thatā€™s just me getting home!

it does look like you hit some traffic lights mid-intervals (I hate auto-pauseā€¦). And that is a tough over/under structure to hit outside if AT is using strict compliance to target as criteria for pass/fail survey.

Here is Echo -3 in Garmin Connect so you can see the targets:

and zoomed into first half to remove the middle traffic related hard acceleration:

and second half without another traffic related hard acceleration:

Same in TR with tempo highlighting:

From TR description:

ā€œEcho -3 is 50 total minutes of Tempo work broken into 8x6-minute segments ranging from 75-85% FTP.ā€

Looks pretty compliant to meā€¦ However there was a lot of Friday night traffic and therefore a lot of short stops and slowdowns.

AT is so strict it gave me the fail survey.

@AlphaDogCycling is right, support@trainerroad.com! Thanks for contributing, the more feedback we receive, the better we can do as a team to isolate issues and identify common points of confusion that are easily preventable! Cheers.

Actually I was riding the same road back and forth, no lights but I had to pause the workout near the end of the road before it met a busy main road then turn round and start things up again.

Being able to tell AT that it isnā€™t a ā€œfailedā€ workout before it makes its judgements would seem to be rather pertinent.

Last night I did some 2 minute intervals, had to turn around because of a busy road:

About a 9 second turnaround from threshold back to threshold power. My preference is to (always) keep the clock running and add some time at the end (= shorten the recovery interval). No significant drop in heart rate or breathing on that turnaround. Again Iā€™d say that is personal preference. On the other hand, auto-pause drops data and distorts the picture so I turn it off.

Although its software, not sure how easy that would be for TR to implement. Unless it uses that info to zero out the workoutā€™s PL and plan adjustments.

@bobw in any case, I agree with your point and would fire off an email to support.