Kind of. If you do some work >FTP you can easily add 10-20w or more to your ramp test result without actually changing your FTP.
A handful of people complaining in short order could very well indicate a problem. And thats what we are disxussing. If you are not interested in discussing, then i suggest you move on and ride your bike for example 
So i was in base, like i said before my vo2 progression actually went down.
Rode my bike twice today, thanks……
Ride it a 3rd time! No typing and riding tho! Thanks
I”m pretty sure @Power13 has every right to type while they ride if they care to do so. Gotta do something to pass the z2 time. Furthermore, they have contributed quite a lot to this thread if you care to listen.
Your ramp test isn’t telling you whether or not aiftp is accurate, or what your threshold power is. It is telling you you got better at taking a ramp test. That probably indicates better fitness. Your ramp test might give you a good number to train from, or it might not. Unless you know what your anaerobic contribution is, it’s kind of a waste of time imo. Better to do 2x20 at suspected threshold and see how they feel. Then you get a good thresh session and know for certain what you can hold for 40ish minutes.
You kinda started with the assumption that aiftp is incorrect, but it doesn’t seem like you have a very good standard to measure it by.
Just my .02, ymmv.
Oh boy i didbt start with that assumption. Before my training block in november my aiftp was 272 and i could not complete the structure, so i did a ramp test and got 252. I adjusted it down and went upfrom there. I purposely did not bring this up previously, as the aiftp was fed unstructured ride info from outside. So i wasnt surprised it was off (this would have been in my pogajar favor…) .
Then i started the structure and got initial fine progression, ramp tests and aiftp correlated, then after custom plan builder agressive, more vokume aiftp went down and as you can see the ramp test continue to go up consistent with my other biometrics. The aneaeobic component worked against me here, since plan builder cut all of it out in the last 2 months, that was discussed if you cared to read.
For your interest last weekend i did a 1:45 threshold session with 4x20 mins of over unders and that went also just fine, hence my surprise when i did the aiftp and saw my ftp went down.
Anyway i have shown data and all you guys have brought is feelings and push the facts that support a pote tial aiftp problem aside. Look at the words im usong potential aiftp problem. Thats all im trying to get across. I have no ftp goal, but i do want to be training in the right zones. Not sure why you guys keep on waffeling, “show me a ramp test thats what we all want to see”. I do the ramp test… “wait no no thats not right, you need to listen”
Seriously?
Grow up.
That got you attention quick! Ganging up on a forum is lame, hence my comment.
Oh boy.
I am not going to go re read every post in this thread but people are not blindly defending ai FTP, there were at least a few other people who doubted the numbers it gave them like you are.
You can either decide it’s not accurate for you and ignore it, do a ramp test that gives you a higher number and decide that is correct or do a longer effort at what you think your ftp is and validate if it’s accurate.
If your recent workout was 4x20 at 280 then I would be very confident in that being your ftp but I have no idea what watts you did for your recent workout at.
I would not at all be confident in your ftp being higher than what you did 4x20 at however, TTE is very much a thing and riding longer at FTP does not always mean your FTP has increased.
The people you are telling me to listen to are blindly defending aiftp, and thats what im responding to.
Im convinced there is an issue period. I can be convinced otherwise by data, not feelings. I have done my part. If you dont like it fine, keep using aiftp.
As I commented above I don’t use ai ftp…. I even said you should question it lol. Keep using ramp tests if the number they give you makes you happy. If you continue to solidly complete all your workouts then it’s probably accurate enough 
Can we please unban The Cog for a day so he can tear us all to shreds about how none of this has anything to do with actual FTP… which would then morph into an argument about what FTP actually is. 
Different methods have different pros and cons, im very well aware. If different methods paint the same picture like they largely did in the past for me its a very good sign you are in the ballpark. If the methods start diverging, it is time to start questioning a few things to understand what is going on. Espescially in the context of high adherance to the plans.
No one is ganging up on you……but please, continue with the childish, homophobic insults.
Hear! Hear!
Cool, so the solution is to completely ignore ai FTP and do a ramp test then. Glad we got there after ~140 posts.
Or… we can investigate the issue and optimize the AI/ML learning model, or you can keep trolling posts for a TR feature you have never used youself per your own words . And derail a thread.
You are best off working with TR directly in that case, they are the only ones that would have the actual knowledge of how the model works.
All we can do is verify it on an individual basis.