I can confirm, the survey pops up right after you hit the end of the workout timer. Once you click your selection, it returns to the app like always.
And you have the option to edit your answer up to 7 days after, if you made a mistake or need to change your response for any reason.
It would be useful if the survey included choices that are unique to outside workouts. I see reasons for over or under performing outside can include: weather, terrain, traffic.
Iām curious about how to ātakeā the survey when someoneās RPE may be different at the beginning of the workout versus the end of the workout. How do you handle that type of situation? Especially an over/under type of workout, which is where I usually struggle on the later intervals.
I bet its been launched today, I havenāt been checking my emails 
I believe they talked about this on the podcast. Youāre not taking a snapshot of one minute of the workout and analyzing your RPE of that, but how the whole thing felt throughout. Itās a given that the first or second interval will feel easier than the last.
Just be consistent on how you rate your RPE - let the AI/ML figure the rest out.
@mcneese.chad could you share the progression level on Spanish Needle -1? Curious how that one is regarded. Have done that one and Mills (4.9) this week just to try seeding some data for AT and they feel so different I honestly canāt guess which one TR would rate higher.
For now I guess Iām going to just compare upcoming general build vo2 max rides vs trainnow intensity recommendations and try to see if trainnow suggests anything that is clearly faster or slower ramp up than the stock general build and just hope AT is released before plan is over.
Just caught up with the podcast on this 
Sounds great, looking forward to making a huge splash in the market. Which makes me wonder you should have kept quiet about it till it was closer and just picked random users for beta and slap a tight NDA on them. Now all the other players will be on it like a car bonnet.
I still want you to partner with RGT though for training, as the RGT team are also good for cycling and not a bunch of elitist ***** ****** like zwift devs
Really looking forward to it now, so keep up the great work you are doing for TR and cycling in general
TR subscriber for two years now and having not used it much since COVID started once all events got canceled because folowing entirely a plan just doesnt makes sense to me at the moment. Have been riding the last months and racing a lot on other platforms for a more social and engaging experience.
Getting second thougts on keeping my year subscription on TR for next year this new feature allowing to get all my workouts and rides done in all platforms and outside considered at all times so whenever I get on TR I can have a specific prescribed training that fits in my training with some King of Logic may just be a game changer to keep on subscribing.
Hope all goes well with beta and by the summer we all can get this new feature available. 
How was your RPE for that point in the interval vs earlier in the interval?
One thing Iām curious aboutā¦using a dumb trainer with speed / cadence sensors and bluetooth connection to my iPad, there is a bit of lag between the power im putting out and the power TR sees, especially on short intervals like vo2 or short sprints. Sometimes it will show that I didnt hit my target for the interval because by the time it was over, TR had just caught up.
So hereās the questionā¦Will adaptive training see this as me not hitting my workout targets and scale my training back incorrectly?
Super excited about the development and I love everything TrainerRoad is doing. Thanks yall!
Havenāt looked for the post, but Nate answers this above - these would be counted as successful intervals. Iām guessing it would see the interval and providing itās in the ballpark of whatās required by the workout, would assign a positive value.
Yup, as mentioned, the ML will tend to learn your style.
But on the main topic, I have wondered as well. My take is to look at the entirety of the main work intervals. Some will feel worse at the beginning than the end or the inverse. So I try to āaverageā the feeling and pick with relative consistency on how that feels from workout to workout.
But like any subjective measure, it will be individual to a point. So trying to rate things similarly is the key, in order to give the ML a basic measure of how you feel with respect to the performance that it measures via the power (and HR?) data.
Question for any of you currently with access to the Beta: I did Baird +6 last week and tomorrow Bashful +6 is scheduled. Baird +6 was an All-Out RPE for me. Seeing as Bashful +6 is a slight step up, Iām wondering if anyone could check the ratings of these workouts and give me a suggestion on something to replace Bashful +6 with? Big thanks in advance!
These look to be similarly rated.
Looking at Baird +6 and some alternates, here is one quick option that may be better for you.
FWIW, I just went through the Baird/Bashful combo and found them pretty much the same - the extra 2% in the intervals is amply compensated for by the longer recovery valleys.
Iāve started using the rating in Training Peaks in order to get in that mindset. I think Iām going to have to remember those last harder intervals should not tell the story of the whole workout, even though they are the freshest in the mind. Maybe wait until a little later instead of rating right away.
Over the weekend I did 2 days in a row of big mountain, 3hrs z2 rated at 7.3. FYI to get to level 10 endurance one needs to do bandeira, which is 6hrs. My rear end is not ready to be a level 10 on the trainer lol
Fortunately, I have been using a default note format to add notes to all my rides. The very first two lines were subjective survey data that I had for current and future reference:
*** EFFORT ( 1 LO / 3 MD / 5 HI ):
*** HAPPYā¦ ( 1 LO / 3 MD / 5 HI ):
This was formerly 1-3, but I just swapped it to 1-5 since the TR RPE uses that. Point being I have a habit of already doing this basic step, but need to apply a slightly different scale now. My general approach was to do a copy/paste with my default note right after finishing the workout. I used the basic feeling in the intervals and my general feeling of exhaustion in that final moment afterward to set my values.
As my usual self, I am over-thinking it all right now and considering the ānewā rating with the finish of each interval. But I need to just roll with it and relax, let the workout happen and make the call at the finish. Itās necessary and helpful, but also is possible to alter after the fact, if you find your initial call was not the best. Just try to get it close and I think the system will figure it out, but that is a guess with minimal experience at this time.
Brain experimenting. Can/will anyone in the beta share which is the highest rated (hardest) VO2, 45/60/75/90 minute workouts, and the rating? I am curious how the length of the workout relates to the ratings. Since, it sounds like, a 6 should feel like a 6, whether it is 45 or 90 minutes, Iām wondering how high you can go in short workouts and whether there is an āidealā workout length given an individualās rating