I used the AI for like 2-3 months now but i’m also frustrated. The workouts are good but if you missed a workout there is no way to get back at the FTP you had before…
You didn’t “have” that FTP. The system was predicting you could have it at the end of the AI window.
Interestingly the other day I did my sweetspot then a hour of endurance both TR workouts in the same ride (time pressure with work) … if I workout matched the outdoor ride to endurance it dropped my AIFTP prediction by several watts, when I swapped it to the sweetspot it boosted me back up…I’m not convinced the system looking at the outside workout at all, merely getting a box tick of completion…
I’d be happy to take a look at your recent training history as well as what you’ve got lined up now to see if there’s anything I can help out with.
Let me know if you’re interested in that and if I can share what I find here.
I think I’ve looked at your case in the past. 
I’d be happy to take another look and/or dig up what we’ve worked on if you’d like. 
Which is your A race, the road race or TT?
If you’ve built your plan around a rolling road race, then we’re going to prioritize that type of fitness (shorter, more puncy efforts).
To be perfectly prepared for one of those events, you’ll likely lose a little bit of specificity that would be useful for the other.
Had you already filled out your post workout survey?
Telling the software that your Sweet Spot workout felt moderate brings different results than telling it that your endurance workout was moderate.
Thanks for the reply, I can’t remember but I can see how that would make a difference. I was expecting that the software would pick out what was done in the workout rather on the basis of what it was associated with tho as the file was the same both times
@eddie you’re welcome to take a look. Username: rkoswald. I have been on vacation and a business trip for the last 10 days, four completely off but was able to get a few spin bike / Z2 rides in. That plus some VO2 failures and I don’t disagree with a slight FTP decrease, but only about half of what TR AI prescribed.
And the complexity that different environmental factors bring. I wonder if some of the analysis that was shared in the most recent podcast around elevation gains while training was done to support the predictions. And if your route makes a difference in the prediction, not sure how that will easily be addressed.
Had 350 and because I missed one interval workout it dropped to 348. Instead of that interval workout I did a longer endurance workout. I don’t think your ftp drops that’s fast;)
I would probably make that assumption too. I’ve asked the team about this, and I’ll follow up when I hear back.
@JasperGr Are you talking about FTP Prediction or Detection? replacing a hard workout with an endurance ride could move your prediction 2W, especially if the prediction was weeks away.
350 is my ftp and the 348 is the prediction for Wednesday. Tomorrow I have a workout but I’m sure after it the ftp drops even more while I completed the workout good.
Only 0.6% difference. Within the realms of power meter accuracy limits.
I wondered about that too.
I’ve also considered taking HR out of the equation. just to see if removing those data makes any difference.
My A race is a rolling road race. That makes sense that training can’t focus on everything. Thanks!
I’m seeing a lot of season-best PRs, but also a lot of failed hard workouts as well.
It looks like you’re often using survey responses that avoid pointing at workout intensity as the source of the problem, such as Sleep, Nutrition, Training Fatigue, etc.
Are the workouts actually the right intensity, but you’re just not prepared for them, or do you think the workouts are a bit too hard?
Regardless, ensuring that you’re able to get through those hard sessions either by recovering better, coming in well nourished, etc., or allowing TR to lower the intensity of your hard workouts through different RPE survey responses is going to be really critical to improving your fitness.
In terms of fitness gains, it doesn’t really matter why you’re not getting through those hard workouts. Missing that key stimulus is having an effect on your fitness, though.
Let me know your thoughts on how things are going and where you think we could make some adjustments to ensure that you’re able to get all of that productive work in, and I think you’ll see good results.
First, I do appreciate you staying engaged with the users while this outdoor workout bug is resolved. Must be frustrating to field all these questions.
Here’s another one ( question, that is) :
In outdoor workouts, does the software parse the rest intervals? Could that be an issue?
Example: a VO2 workout might alternate between 300W (work) and 104W (rest) indoors. On an indoor ERG trainer, you have no choice, it’s 104W. But on Garmin outdoors we get a range of 290-310 (work) and 0-104 (rest). I’ve always assumed that as long as the pedals are turning during rest, it doesn’t matter if you hold 104W, 80W or 60W. As long as you hit the work interval targets and don’t go above 104 during rest, you’ll get the stimulus. But rest intervals make up a larger total time than work intervals in a VO2 session. So if my rests outdoors are significantly lower than 104, even with perfect work intervals at 300 does this look like a failed workout to AI?