I’ve just completed SSBLV1 starting with a TR FTP estimation and using adaptive training. I’ve never been a particularly compliant user as I like to ride outside, MTB, Zwift and play Ice Hockey.
I used the FTP estimation tool at the end of week 4 and was delighted to see a 5% 14w increase, so with a recovery week ahead decided to take it really easy before the start of SSBLV2.
I’ve been careful to make sure that most of my other indoor riding over and above plan has been either TR endurance or Zwift focused on Z2, to consolidate base, trying to nail low volume as is the guidance.
I’m naughty but I have used the FTP estimation several times during this week and it’s indicating my FTP is consistently dropping during the recovery week, to the point that I’ve lost 7 of those hard earned 14w. A steady 5% gain has been eroded to less than 3% during recovery.
Now I know it’s not the end of the world, and probably won’t make a significant difference in the longer term but would this be typical of FTP estimation and adaptive training?
I can imagine that if I’d made smaller gains FTP estimation might have me a lower FTP than when I started?
I love not having to do a ramp test and am generally feeling pretty motivated by the recent enhancements to the TR product and look forward to progression levels picking up riding outside of TR.
I can’t add anything in the way of concrete answers, but last block I checked and it predicted 218 last hard day of the block, it fell several but I did the test (I wanted to prove it to myself, I don’t like the tests 
), I tested 218 exactly. One block later it’s estimating 244 on the last hard day (today), I’m not gonna check before doing the test in just over a week, whilst I’m not expecting it to be exactly on the money twice, I don’t think it’ll be far off based on the improvements I’ve seen outdoors.
There’s definitely something, not necessarily a bug, to be considered, planned recovery obviously isn’t bad, it’s necessary, but it seems to be treated harshly in FTPD, judging by my very limited sample it’s probably most accurate during the work weeks of a block, it would be interesting to compare, for people still doing ramp tests, detection on days leading to a test against the actual test results, but that’s something I can only do a limited sample for…
Remember, it’s not predicting your Lactate Steady State/Hour Power (hard for me to separate it from after studying Ex. Phys.), but rather trying to give you just the right level of workout.
There’s possibly some data in TR saying that generally athletes don’t hit the same intensity workouts as well first week after recovery and so there’s some gentleness there.
My advice is to only check when scheduled - or if you’re going to shift it to pre rest week or whatever, do it the same each time. So you’re comparing apples and apples.
Pretty much same thing happened with me. From week 4 of ssb1 to after my recovery week AI ftp gave me about a 50% reduction. So I’m not really sure what is right/what is the better number to use…
You must not be accepting the FTPD value if you are able to see it repeatedly? Maybe your responses for difficulty based on your current FTP setting are causing the system to doubt the higher FTP estimates.