Genuine question to get a sense of how people measure and interpret performance progress. There seems to be a lot of questions on the forum on merits and accuracy of AI FTP/ predictions ( too high too low just right), but without a lot of discussion on power curve PRs or route PRs (personal KOM).
Just wondering the split of people benchmarking themselves more based on primary data ( power curve ) vs algorithm generated number ( AI FTP for instance).
Watts! 
I usually stick pretty closely to a plan of some kind so a lot of the same types of workouts are always in the schedule. I know what watts I’m usually pushing in these workouts so I can easily track improvement. I like to watch my Endurance + wattage going up. My Over and Under watts as well are numbers I like to see going up. FTP prediction is a fun carrot to shoot for but I don’t really track with it. I watch the watts.
I’m adhering to a specific plan…masters aggressive…and do use the ftp score as ‘one’ method to gauge fitness and improvements.
I also use Strava…it’s AI does a pretty good job, subscription level, of describing the day’s workout/ride and comparing it to previous rides based on intensity, elevation, distance, duration, mph, etc. and comparing to the past 30 days. IMO it’s a pretty good method of charting progress but like all plans it really doesn’t dive deep into an ‘off day’ or a ride where you feel tired…perhaps from lack of sleep, life stress, etc…it just shows a lesser intense ride and comments on the results not knowing the reasons behind the results which can then be misleading.
I’ll typically look at power curves and “like for like” workouts. And on workouts, it can be looking at the the identical/similar workout from prior season at about the same time, or just comparing similar stuff week to week (ie - last week I knocked out 9x3’s at x watts, this week I’m doing them at a little higher wattage or finding the same workout more manageable).
I think you need to consider what training phase you are in. For base, you would normally have a SS, over/under, and some zone 2 that are on the higher end of the zone.
I track all three of those type of workouts and monitor if my watts are going up at different durations and how it feels to previous workouts. I consider it progress if I do the same watts, but it feels easier, and obviously if the watts go up at similar effort.
Once you get into build, watts tend not to go up as much, but your ability to repeat higher efforts goes up, so I tend to look at NP a little more here. I’d like to see NP for different durations rising during this period.
Depends on what you are progressing at this time.
FTP work is easy to track improvements in, other types of work it depends on your goals.
I know it’s April 1st because this morning I’ve read 3 different replies telling people not to worry about FTP increases in Base, Build, and Specialty because “that’s not the phase when you gain as much”.
Ha, yeah - and then there are the folks that say FTP increases don’t matter at all!
Seriously though, if you do a proper base, you most likely will get most of your FTP increases during this phase - at least that’s been my experience and other folks I know who do structured work.
The new AI FTP only has a small history. Comparing old FTP values to the new AI FTP doesn’t make a lot of sense. Going forward, you could easily compare AI FTP values over time.
To compare pre- AI FTP ability to now, I would simply look at power curves. Break them down by season, year, whatever.
That’s my experience as well, but I’m typically coming into base after a decent off season where fitness has dropped quite a bit. But for people who train throughout most of the year, I’d assume base make less of an impact and they probably don’t see big jumps in any phase since they never drop too far.
I like to use the Season Match feature in the Power Records
PRs (power or time), KOMs, wins. Any or all of the above.
I don’t really care or pay attention to AI FTP. I care about real world numbers and performance.