Kinda related to this subject is an issue I raised a ticket for with TR - just interested if anyone else has observed the same issue.
I recently did VO2 Max Thimble +2 workout but I didn’t really pay much attention to the PL rating of 7.4 for this workout.
If I had checked my current progression levels, I would have seen my current PL for Vo2 Max was 3.4, obviously this is a very big delta.
Since I had not done any VO2 Max workouts TR reduced my PL from 4.3 to 3.4 on 18/Oct. My last VO2 Max workout was on 21 Aug with a PL of 7.1
So it makes sense that TR drops my Career PL since I’d done very little VO2 Max workouts in the past approx 2 months, but throwing a 7.4 session at me was a little cruel I reckon 
Anyway, off I went into the first set of intervals (3 sets of 7 in this workout), I got through them (target was 367w and I hit that except for the 7th one where I got 360w) but felt that they were excessively hard.
Next interval I only got 358w and really struggled with it so I knew then that I’d have to drop the intensity (to 90% which equates to 330w) in order to complete the rest of the workout.
Nailed the rest of the intervals (possibly could have upped the intensity a few % but said I’d leave it) and just to spite myself I went hard on the last interval to average 384w.
So in terms of scoring for the workout, my PL did not increase at all (still at 3.4)
While I did drop the intensity to 90% after the first set of intervals, I still hit 347w average out of an original target of 367w, this equates to nearly a 95% completion performance. It’s a similar figure when you compare TSS and IF target-to-actual figures.
Thus I expected my PL would have increased somewhat above 3.4.
I got a reply from TR advising:
”Since you did not complete the workout fully, you won’t get the Progression Level increase from this workout. However, it will be taken into account for workout recommendations, AI FTP Detection, and you will, of course, get the physiological training benefit from the work you completed. ”
Essentially what TR seems to have done is to ignore the workout performance completely with regard to PLs.
I would suggest for a future enhancement that as opposed to ignoring the workout completely when a user eg; drops the intensity, that TR algorithms take into account the actual performance-to-goal of a workout, in particular for the intervals if the primary benefit of the workout is completing the intervals.
A point of note is my upcoming VO2 Max workouts, do you think they seem like they have taken my Thimble +2 workout performance into account? appreciate any comments 