Guys:
Sorry for overlooking this thread. Steve just brought it to my attention. The subject of TSS and the other TP metrics that fall out of it are somewhat involved and not completely straight forward so let me summarize here. Steve did do a video on this subject https://uphillathlete.com/performance-management-chart/ that helps explain the various scores like TSS, CTL, ATL, TBB but perhaps we need to to do another more in depth video on this topic.
Points to keep in mind:
1) TP uses a sophisticated algorithm that looks back at the past 6 weeks of TSS. The model uses a weighted, running average of TSS wherein the more recent workout’s TSS count more than the ones you did several weeks ago. All these metrics are only as valid as the TSS inputs to the algorithm.
2) The TP algorithm can calculate TSS using POWER for cyclists using power meters, PACE for runners on relatively flat ground, or HEART RATE. Power TSS is very accurate because it is a direct measure of the work being done/stress of the training. Pace is the next most accurate because running pace is a very close proxy for the work being done/stress on the body. HR is the least accurate way to calculate TSS because HR is not a particular good proxy for the work being done/ stress on the body. Due to the variable nature of training for mountain sports we have to default to HR. This makes our input into the nice TP algorithm suspect to begin with.
3) As mathematician George Box said: “All models are wrong…..Some are useful”. So even with our model being flawed it still offers us some kind of measuring stick. But we need to be able to make our hrTSS inputs as accurate as possible so that the model’s output will be more accurate. This entails some guesswork and fudging which has its drawbacks for sure. Being consistent is what counts the most. You want to be comparing apples to apples in your own training first and foremost. Forget about comparing yourself to other or some arbitrary number like TSS of 100. That is not a magic number. It’s a very rough guide for an individual. That individual will be fitter with a CTL of 100 than when they have a CTL of 80.
ADJUSTING TSS
This is where the magic happens and so hold on.
1) TP’s model was designed to reflect the TSS for aerobic endurance activities. As soon as we start using it to measure things like strength training/technical climbing, ME workouts,
2) Aerobic Training:Up to ones anaerobic threshold the relationship between HR and effort/pace is linear. In this aerobic zone HR is a good predictor of TSS making the TP model for aerobic workouts using hrTSS pretty accurate. Over AnT the relationship goes non linear the harder you push the more non-linear it becomes. What this means is that in the more anaerobic realm of intensity each increase in pace/effort results in a smaller and smaller increase in HR. The TP model understands this and again does a decent job of adjusting for this. However the non-linearity is not the the same for everyone so this begins to introduce some potential for error.
3) Strength Training: This is where the model goes completely off the rails. That’s because HR bears almost no relation to the effort or training effect of strength workouts. By “strength” I mean: lifting weights, hill sprints, hard climbing moves, down hill skiing or moving a piano. Anything that you do that requires very high effort and lasts for a short duration. Yep. those terms a pretty nebulous. High effort means you can’t sustain if for long. The higher the effort the shorter the duration. A pitch of 5.10c is going to require less strength (effort) than a V7 boulder problem. Most us could could not string together 100 feet of V10. So, figuring out how to count the TSS of strength workouts is a grey area. I’ll get into how we do it at UA in a minute.
4) Muscular Endurance: The water becomes even muddier here because ME is a mix of strength and endurance and spectrum along which your workout could be place is long with no great measuring stick here. All we can say for sure is that one end of the spectrum requires greater strength so will allow fewer repetitions (sustained movements or endurance) and the other end requires much less strength so allows many more movements. But no matter where you are on that spectrum; from a 3 move crux to a vertical K run to a gym ME workout to a weighted uphill water carry, you will be operating at, or very close to your endurance limit for the amount of strength you must put out. HR may or, in many cases, may not be a good proxy for the effort. The farther you are toward the strength end of the spectrum the less HR reflects effort. Anyone who as done a well executed weighted water carry ME workout or a hard climbing session where your legs or forearms are are limiters can tell you that their HR does not reflect the effort expended.
5) Adjusting TSS for aerobic work: Here’s what we typically do (of course there are exceptions). For sub aerobic threshold workouts hrTSS will typically net you a hrTSS of around to 50-70/hour with the less aerobically fit being on low end. Aerobic workouts done at one’s anaerobic threshold yield a hrTSS of 100/hr. For all these aerobic efforts we add in 10TSS for every 1000 feet (300m) gained and lost in these workouts. This is admittedly arbitrary but be consistent with this fudge factor and you’ll keep things apples to apples. When we add weight we add 10TSS/1000 feet / 10% of body weight carried. Arbitrary again but pretty effect.
6) Adjusting TSS for strength and ME work: With Strength workouts we typically award 60/hr for one hour of our general strength/conditioning sessions. Max Strength we award 80/hour. Same for a good hill sprint session. For ME we use the residual fatigue from the workout to set the TSS. Workouts that have a strong ME effect will leave you fatigues (doesn’t matter if its forearms or quads) for several days before you will begin to recover to the point where you could potentially complete another workout at the same intensity. For these we generally award 100-200 based on fatigue.
Hopefully this helps.
Scott
