OK Sam, prepare to get down in the weeds with me here on this one.
The TSS numbers I gave are UA derived. Here is how I came to figure out this system.
A while back I was at a coaching conference and some very experienced Triathlon coaches were talking about Ironman prep. The TSS numbers that TP can calculate were solid enough for them to be able to say pretty with some high probability of being correct that to compete the Kona race under 9 hours one needed a CTL of 150. They also had enough data to do similar correlations for other races and marathon times. This got me thinking about how helpful it would be for UA and the clients we coach if we could have a similar yard stick.
Soooooo…. Here is how we came up with the our methodology.
Chronic Training Load (CTL), when calculated using hrTSS (heart rate Training Stress Score) only considers heart rate when calculating TSS for any given workout. TP is very clear that TSS is most accurate when using power to determine TSS (as in cycling with a power meter) or pace to determine TSS for road and track runners as in your marathoner example. Both Power and Pace are very good reflections of the actual local muscle work being done by the athlete so TSS in those cases paints a really good picture of CTL. Unfortunately for us, as mountain athletes, going up and down hill, we can’t use Power because no good power meter for trail running, hiking, skiing exists yet and pace is meaningless on such variable terrain. So, we are left with using HR as our metric for gaging the training load.
The problem is that HR is an imperfect measure of training load. On flat ground and in the aerobic realm (AnT/LT) the relation of HR to load is linear so there it can do a decent job on TSS. However, as soon as you add in a significant muscular strength/endurance component such as a 3 hour trail run that include 2500 vertical feet of elevation change, then HR no longer gives a very accurate assessment of TSS because TSS is the combined stress of the workout (metabolical stress, cardiovascular stress and local muscle fatigue).
As for Strength workouts hrTSS is meaningless because HR is no indication at all of how hard a strength workout is.
So what is an Uphill Athlete to do: Here is what all the coaches at UA do. We introduce some fudge factors to the hrTSS scores for daily workouts. The importing thing to keep in mind for the rest of this article is that you remain consistent in how you apply the fudge factors. That way you will always be pretty much comparing apples to apples.
Here is the system I have developed for us and it seems to work pretty well:
1) For any purely aerobic run/hike/ski we calculate the TP hrTSS and add 10 TSS for each 1000 vertical feet of gain for an athlete carrying no to minimal weight.
2) For the same workout but with a significant weight, say more than 10% of BW add 20TSS/ 1000 feet.
3) For hard muscular workout that has a very high local muscular endurance factor with a (disappointingly) low HR such as an uphill ME workout then I pick a number that reflects the recovery time before the athlete feels ready for another such workout. An ME workout for a high level athlete like David Goettler or Luke Nelson will take them days to recovery from and based on the actual workout I have assigned I will give these a TSS of 150-200.
4) For the TftNA general strength and core workout I give them a TSS of 50-70/hour.
5) For the TftNA Max Strength with core warm up I give these a TSS of 80-90/hour.
I will grant you that this seems arbitrary and some is perhaps seems too formulaic to apply across broad swaths of athletes. But these fudge factors do reflect the actual pretty well and it works well enough when we are assessing fitness and preparedness for certain undertakings. It take a lot of time for us to fudge the numbers on all of the athletes we coach but we feel it is worth it and because we have been doing this for years with many dozens of athletes I was able to say that 100 is pretty much the price of admission for an O2 assisted 8000m peak ascent.
Obviously we are not talking accuracy to the third decimal place but this tool is the best fitness and recovery monitoring tool I have every come across in 30 some years of trying to find a better solution. And I was highly skeptical of it when I started using it but now I am convinced that with suitable modification it works well enough.
Sorry for this long winded explanation.