My approach would be to pick a softer endurance workout that you know to feel easy. And if you don’t know in advance that your workout becomes moderate than you could reduce intensity on the fly so that it stays RPE easy.
1 Like
I wouldn’t do that. Either self select slightly harder workouts or raise the power targets manually while in the workout. Then rate consistently. If you change your definition of hard, easy, etc - that will impact you in the future when you get to the right level.
1 Like
Yeah. Trying to game it is probably counterproductive, but testing to see what happens when you choose a different response to see how the system responds and questioning why is legit, IMO. I usually have some difficulty deciding which is the “truthful” response, like could I have done another interval? Yeah, maybe or maybe not. I can see where seeing a change in the predicted FTP could influence someones choice.
2 Likes
That’s a useful test for people: “if seeing a change in the predicted FTP influences your RPE survey choice, then switch off AI FTP Prediction”
(…because you’ve succumbed to gaming it)
2 Likes
Not necessarily. You can respond to the survey as honestly as possible, and then see what would have happened had you answered differently. This provides information about you and about the system. It’s only if you change a response to get a more appealing AI prediction that you’re gaming the system.
1 Like
In which case, it’s not influencing your survey choice then…
You’re gaming the system if your selection is influenced by how the model changes in response to your choice, rather than making your choice honestly. What you’ve described is different to that - it’s just being inquisitive about how the model works. If after scratching that itch you still provide an honest response, then obviously you’re not gaming it.
In 6 months time once the novelty’s worn off a bit, we’ll probably all be in “Jonathan Mode” in any case, barely even bothering looking 
2 Likes
It all evens out in the end. You rate one too easy and rate another too hard but get the vast majority accurate. When the FTP detection comes it will all come out in the wash.
2 Likes
I’m in the same boat as you.
I feel like my rate of perceived exertion is weighted much too heavily in the system. How I feel on any one day is quite arbitrary. Power and heart-rate doesn’t lie, though.
For example, I have VO2max workouts that I would rate 9, because the intervals are the absolute maximum power that I can hold. At the end, I could possibly do the whole workout again, but I couldn’t hold 3 more watts on the hardest intervals. Also, some of the endurance workouts kind of suck (i wouldn’t rate them “easy”), but I could do higher power for much longer at a similar level of suck.
I’m thinking of just skipping the surveys.
1 Like
When they said “keep the easy days easy, and the hard days hard”, I don’t think this is quite what they meant. 
2 Likes
Perhaps the problem is in gamifying and focusing on an estimated gain that is wholly new to you, rather than following the workouts prescribed to you, along with your actual training experience and data and RPE and how you feel on the day to steer your training from day to day.
2 Likes