Using CTL & TSS to Monitor Training Load and Adjust Workouts
So, I read Friel's blog today about CTL and TSS and using both to avoid overtraining.
I know how to import my workouts into wko+ and monitor the CTL curve, but I don't really know what do with that curve. I assume that if the curve starts to flatten out then I'm not seeing a fitness gain either because I have ratcheted back training loads, or I am not able to perform the prescribed workout at the specified intensities (the latter being an overtraining indication?). The other aspect of this that seems to be significant is the rate of CTL increase (e.g., how many TSS points increase in the curve over a certain time period).
All that being said, I was wondering if anyone has any thoughts on how I can use this CTL data to adjust my weekly training load to ensure that I am either getting enough recovery, or, less likely, pushing hard enough to build the CTL/fitness curve at an appropriate rate.
Thanks, John
Comments
Of course, the big caveat to doing this is that you have to ensure that the figures in your PMC are in fact accurate. The goal of training is not to see who can rack up the most TSS points, just as my goal in an FTP test is to not to get the highest number, but rather in both cases to get the most accurate representation rather than the highest. This requires either frequent testing, or a good understanding of your true thresholds and an ability to adjust them somewhat fluidly in accordance to your changing fitness. Personally for me the entire process is still more art than science, and I'm far more amateur hour at this than many who have been doing it for years.
With that being said, there are a few things I look for in the PMC. For one, I try to understand the phase of training that I am in and what the primary intent of that phase is. Not all TSS are made equal and different types of training certainly elicit starkly different physiological responses than others that is not necessarily being captured in your resulting metrics in WKO. For example, the VO2 max block from the OS or short course plans definitely can take a much greater physical toll than the TSS would indicate.
From there, I try to keep an eye on my TSB, how fatigued I am, and how long I am that fatigued, then look back at my modest collection of data from the last 2 years and see if there are any patterns that have emerged that would help me from digging myself in a hole from which I am unable to recover. This historical data is perhaps the most important to me, if I look at a large sample of data it is always peaks and valleys, a solid build phase in which my ATL and CTL begin to show very solid progressive climb, usually coupled with extended duration into deep negative TSB, ultimately ending in a localized trough essentially representing a burn-out due to excess fatigue before I eventually recover and begin to build fitness again.
In this regard, I think a phase of more gradual CTL climb, or even flattening, is entirely acceptable and even desirable, the body needs time to recover, and this is certainly preferable to a burn-out and an actual trough (CTL decline). But, if my CTL began to flatten for an extended duration, it would signify that a more profound change is necessary, whether that be going to a more advanced plan, or speaking to the coaches and smarter people than me to ask for advice.
So in the day to day of it, I essentially monitor my TSB and/or ATL (since they highly related), but more importantly the CTL and longer trends about where I am in a build-recover-build cycle. If I'm at -25 TSB and I have my long run tomorrow morning, but I'm on the upswing of a fresh CTL build (i.e. I'm at the start of the build and haven't been fatigued consistently for several weeks) then I won't worry about it. History as shown me I'll bounce back pretty quick. But if it's Wednesday, and I'm wiped after my workout that day, deep in the negatives on TSB and have been for weeks, I usually can tell that I'm not going to recover quickly enough to pop out of bed at <5a and go get a 2 hour run in. In that case, I'll probably let myself get that extra sleep on Thursday morning and move my run to the evening and/or Friday morning. <br />
Final word of caution, all of this is n=1, i.e it's just what I've observed that seems to work for me, and it's definitely a work in progress. But I think if you make an longer term effort to map your estimated fitness (WKO+) with your day to day perception of how you feel, it's ultimately very powerful. No coach, no matter how much you pay, knows as much about you and how you feel as you do yourself, so even with the best athletes and best coaches, the athlete needs to have considerable input into the process.
Agree with Trevor. I use the data as well and make sure I keep it updated consistently. It is usually right on compared to how I feel. When I see a -20 TSB I am also feeling the load through fatigue. When it begins to show -30 or worse, I feel like my eyeballs are sunken into the back of my head from the fatigue. At -5 or -10 I am starting to feel frisky and at +5 I begin to feel like a caged tiger. I do think it is a useful tool if you know how to use it.............
The way I prefer to think of this is that TSS (and NP, from which it derives) is only a MODEL for the physiological stress of training. As such, it behaves pretty well in a certain range. It does a good job of comparing one 2-hour ride to another. But it doesn't handle extremes well. It'll let you compare tangerines and oranges, but it has a lot harder time comparing grapes to watermelons. In my opinion, for example, this is the reason that the "allowable" IF on an IM ride hits a floor and doesn't keep getting lower and lower - it's an extreme limit of the model where it's starting to fail (long time and low power). Trevor's example of VO2 invervals is another extreme: high power and short time.
Regarding the ATL and CTL.... it's a lot easier if you remember exactly what they are:
CTL and ATL are basically moving averages of your TSS points per day (though they are exponentially weighted). Because of the different time constants, ATL tells you roughly about the last week, and CTL tells you about the last several weeks, and it does not move very much from day to day.
TSB is just comparing the two. If you are in the middle of a build up, it will necessarily be negative. When you are tapering, it will necessarily be positive. If you have a couple days off because you are sick, it will be positive. Not because you are "rested" but because you didn't rack up TSS points for a couple of days, and that drops ATL. If you have a month-long heavy training load, toward the end of that, the CTL will be high, but the TSB will be closer to zero (assuming it was a hard but steady load).
Once you really realize the simple mathematics of what it all means, then Trevor's recommendations make a HUGE amount of sense... but remember ATL, CTL, and TSS really aren't magic figures...they're just moving averages based on an approximation of workload. In that context, JF's discussion of how much CTL should go up per week is a more sophisticated version of the old bromides about increasing running volume by no more than 10% per week...it's just a rough number past which you're taking a risk.
Super helpful comments. Much appreciated and gives me a better understanding of how to use wko+ as one of several tools.
John