Do recovery rides figure into the Performance Manager?
- Thread starter DRAwpt
- Start date
DRAwpt said:
Sure. A recovery ride is going to have a low TSS, which will bring down your ATL. Your ATL drops significantly more quickly than your CTL, in turn resulting in a positive TSB.
OK, so I guess I want to know if doing a recovery ride will increase my TSB more quickly than simply resting, according to the Cycling Peaks algorithm. Does anybody have any references to a studies that show the efficacy of recovery workouts?shawndoggy said:
It´s quite simple: TSB is CTL minus ATL. By doing a recovery ride, you alway increase the ATL in comparison to a day completely off, therefor TSB rises more slowly - in the performance manager.
In real life, it strongly depends on the circumstances what is more beneficial, but in most cases, a short recovery ride will be better. This is one of the points when one has to remind oneself that the performace manager is only a mathematical model and has some caveats.
In real life, it strongly depends on the circumstances what is more beneficial, but in most cases, a short recovery ride will be better. This is one of the points when one has to remind oneself that the performace manager is only a mathematical model and has some caveats.
The short answer is no.
From a given starting state, the optimal strategy to maximize TSB is to train at the limits of physical breakdown, then hard rest (preferably stay in bed) until the event. The duration of this rest is shockingly long, which depends on the precise values of the time constants for ATL and CTL.
Dan
P.S. I use a different formula for TSB, just for fun. I have two components of TSS: one based on the integral of (P/FTP)^4 (the high intensity part), the other based on the integral of (P/FTP) (the low intensity part). These are added (no square root, per the usual formula, as that results in the non-linearlity problem which has been well documented in this forum) to get net TSS. But I calculate a separate CTL and ATL for each part, and use shorter time constants for the high intensity part (21 and 3.5 days) than the low intensity part (42 and 7 days). The effect of this added complexity is that, to maximize TSB, the optimal schedule becomes first, a reduction in volume with maintenance of intensity, and later, rest. However, even this approach fails to recognize the benefits of "active rest".
From a given starting state, the optimal strategy to maximize TSB is to train at the limits of physical breakdown, then hard rest (preferably stay in bed) until the event. The duration of this rest is shockingly long, which depends on the precise values of the time constants for ATL and CTL.
Dan
P.S. I use a different formula for TSB, just for fun. I have two components of TSS: one based on the integral of (P/FTP)^4 (the high intensity part), the other based on the integral of (P/FTP) (the low intensity part). These are added (no square root, per the usual formula, as that results in the non-linearlity problem which has been well documented in this forum) to get net TSS. But I calculate a separate CTL and ATL for each part, and use shorter time constants for the high intensity part (21 and 3.5 days) than the low intensity part (42 and 7 days). The effect of this added complexity is that, to maximize TSB, the optimal schedule becomes first, a reduction in volume with maintenance of intensity, and later, rest. However, even this approach fails to recognize the benefits of "active rest".
