I had a similar observation. I have just compared the training plan I have completed (SSB HV, SusPB HV and spec) and they seem quite a bit easier in the new plans than the old?
Having had 100% compliance rate I am concerned to do the same plans next year if they have become easier (less hard!). Although will AT be able to spot this and adjust future workouts if this is the case?
Thanks. I am excited by AT and the almost gamification of being able to ‘level up’ although do struggle to comprehend how it will be better when the harder workouts resulted in 100% compliance this year. Hmmm
One point that bears mentioning is the intent with recreating these plans.
If the goal was to make the plans easier, we could have just whipped out a whole plan of Baxters real quick
The goal was not to make the plans easier or to make them harder. It was to analyze data about completion rates, difficulty, productivity, and do so at a really high resolution so we could create data-driven plans that increase the odds of making all of us athletes faster.
Another point on that: Just completing a workout doesn’t mean you get all of the intended benefit. We’ve talked about this commonly on the forum and podcast, but do the same plan with more sleep, better nutrition and lower life stress (ultimately better recovery and less stress) and you will likely get better adaptations from that work.
These plans are designed with that context in mind – not just looking at individual workouts, but how to set athletes up to get the most improvement.
So I bet 100% on the new plans will get you even further than before.
Maybe ‘subjective’ is a better description than arbitrary - there is reasoning behind how hard a ‘10’ is for each category/energy system. (I’m assuming a 1-hour interval is not a 10 for all energy systems - that would be arbitrary.)
I think for threshold (and above) the limits are more readily apparent - 2 hours at 102% of FTP is likely not humanly possible for correctly set FTP. It might be useful to flag workouts that are completed but that should not be possible and suggest the user update their FTP For efforts below FTP I think the ‘humanly possible’ limits are likely more squishy.
I think normalizing things like you have done makes the levels intuitively useful - things that are close to 10 either don’t need work or my FTP needs updating. (or at least that’s my take-away of how to think about them.)