Time to Recover

Is there a way to make the model speed up my recovery, it’s been overestimating it all year. Like I finished a big block last week before taking time off on vacation, came back did one hard(er) workout and now I’m “Very Tired” for the foreseeable future… I’ve read in a recent forum post(Help me understand the tiredness star) that the model needs time to align my actual fitness with the fitness it knows about, but it’s been 2 full months now since I resumed training.

any settings would allow me to tweak this?

HI Jim. Try changing your time constants in Account Settings/Profile to 22/12, 22/12, 60/5 for peak, high and low resp. from 42/7 and see if that makes things better for you. These are the values we’re using for new users and it appears to do a bit better.

Armando, after changing time constants and clicking “Save Settings” down below, a banner shows, saying profile has been updated. Is that it, or do I have to click on “Recalculate Progression” also ?

Nope. After hitting Save the system will do all the work for you. You can see the update to the XPMC chart happen if you open it right afterwards.

it does look slightly better, care to explain what are those settings and their effect? and also what is the seed value setting underneath them

Unfortunately, it’s a bit overly complicated to explain in this box. We’ve been meaning to blog about it but even then it’s not that simple. There are articies on the web that talk about increasing/decreasing time constants based on if you recover faster/slower but unfortunately they don’t have the math quite correct when they make those statements. It doesn’t really operate that way. We’re looking to add a new feature that will allow you to tune things a bit more. If we do it right, it should help you make your own adjustments without have to understand all the math behind it.

Oh, that would be great!

It would, I agree.

For the interested user if you could provide a reference to the relevant litterature, it would be also be great

Googling will provide a number of hits, none of which provide really help hence often hearing the argument that it requires as much art as science, something we don’t believe in: https://www.google.ca/search?ei=yQ6bWqXNA4GKtQXi75jwDA&q=time+constants+atl+ctl+42&oq=time+constants+atl+ctl+42 We just need to find a way to help people map themselves and how they feel to the data better.