PMC and CTL/ATL Constants...

Porkyboy said:

Hi

I'm using the PMC to manage my training load and hopefully bring myself to peak form for certain events later in the year, this is the first year I have tried to do this in anger.

The degree of freshness/form or whatever will be dependent on the TSB at a certain point and the TSB figure displayed will in turn be dependent upon the applied CTL/ATL constants, the defaults being 42 and 7. My problem is won't people all recover/adapt etc. at different rates and is it not therefore possible that the mathematically derived TSB you are working to is actually mismatched to your own rates of recovery/adaptation?

Is there a way of working out or estimating what your own CTL and ATL constants should be set on to try to ensure that the TSB figure is in reality a true reflection of your current freshness/form. It just seems to me that if the CTL and ATL constants are not personalised and accurate then you might as well just go on your gut feeling of how you feel?

Thanks for any advice!

PBUK

Well, this is where having a history of power data becomes very handy. You can analyse past seasons and see what time constants best match your performances and/or perception. Typically however it is the ATL TC that you'll want to focus on. Changing the CTL TC a bit doesn't have that big an impact.

As a suggestion, clone your PMC and enter in different TCs for each. Then track how each matches your own sensations.

It's also part of where the art and science of training come together. Everyone is different.

There are some (e.g. Dr Phil Skiba) that have taken the concept further so you could read up on his work.

As a general rule of thumb, shorter events need more freshness than longer ones. Older riders need more time for recovery than younger ones. So that may point to a need to lengthen the ATL TC when targeting short events or if you are of masters age.

[quote="Alex Simmons"]
As a suggestion, clone your PMC and enter in different TCs for each. Then track how each matches your own sensations.

It's also part of where the art and science of training come together. Everyone is different.

QUOTE]

Hi Alex

Thanks very much for the advice, I'll take it. I know this is a bit of a piece of string question but typically how much would you change the ATL constant for this experiment? I was thinking of setting up charts for say 5 days and 9 days, 2 days either side of the default, or would you change it by more than than, I'm almost 50 BTW :eek:

Thanks.

PBUK

Porkyboy said:

Thanks very much for the advice, I'll take it. I know this is a bit of a piece of string question but typically how much would you change the ATL constant for this experiment? I was thinking of setting up charts for say 5 days and 9 days, 2 days either side of the default, or would you change it by more than than, I'm almost 50 BTW :eek:

I think for the purposes of assessing the PMC and in particular, training stress balance, I'd start with an ATL TC of five, seven and 10 days.

One other consideration, of course (just to complicate things), is that your ATL time constant may indeed vary depending on what stage of the training season you are at.

Easy, huh?😉😄

Just try not to lose sight of the forest for the trees.

Alex Simmons said:

There are some (e.g. Dr Phil Skiba) that have taken the concept further so you could read up on his work.

Phil's program simply implements the impulse-response model of Banister et al., upon which the Performance Manager is based. IOW, rather than taking things further he's really "returned to the roots".

acoggan said:

Phil's program simply implements the impulse-response model of Banister et al., upon which the Performance Manager is based. IOW, rather than taking things further he's really "returned to the roots".

OK. I was under the impression that it was one way to nail down one's TCs. I could never make much sense of the output of my trial version since my historical data doesn't contain test data of the standard Phil's software seem to require - and therein lies the problem in my eyes - the type and especially the frequency of maximal testing required seem to me to be more than is practical.

Phil did offer to do his own analysis of my data but I never progressed further with it. I have enough trees to look at at the moment 🙂