Chill Runs, very thoughtful post! This is the type of discussion I’ve been waiting for since TSS was first brought up!
As far as I can tell, in trying to discern how Andrew Coggan derived TSS, it seems he used the percent of 60 minute race pace (which he called FTP or Jack Daniels would call T-pace) and the percent of the lactate value at that pace (based on using a power function to model the lactate curve) to assign stress points. Then scaled it so that the score for a 60 minute race effort would be equivalent to 100 TSS. Sirpoc shared with me some old cycling threads Coggan posted in and he himself states that TSS is not a system meant to represent physiological stress, strictly training stress. I share all that only to set up this next point.
As I interpret your post and the general validation for using TSS from others, it seems to reduce to maximizing training load, so CTL. That long-term training stress is thus dependent on the TSS equation. So, with that in mind, consider this. A score of 60 TSS can be achieved in many ways, three of which are 55 minutes at 80% FTP, 36 minutes at 100% FTP, and 30 minutes at 110% FTP. To put that in terms of Jack Daniels 80% FTP is E-pace, 100% FTP is T-pace, and 110% FTP is I-pace.
So, according to the TSS equation, those three workouts would be equivalent in stress. Now, I don’t expect strike zone accuracy, but being in the ball park seems a reasonable expectation. Does any runner think those are even close? If I made it to 30 minutes at I-pace, I’ll have also made it to heaven, because I’d be dead haha. Contrast that with 55 minutes at E-pace (which I think many feel is more like a moderate pace). That’s something I could do regularly, particularly if I’m not doing any workouts.
To be fair, and I do try, Coggan designed this system based on training by power for cyclists, so there is that consideration. However, if the underlying equation which is quantifying training stress is so skewed, how seriously should it be treated in an attempt to generalize an explanation on? I think the poster lactate_guided pointed this out… TSS, while directionally accurate, is likely not a predictor of race performance nor physiologically accurate.
Another way to possibly frame this, if training load, as determined by TSS, is the driver of fitness, one should be able to achieve the same race results on the same CTL score, no matter the intensity distribution. I imagine that if most people took their weekly average intensity, it would fall between 75-80% FTP? That’s probably a little slower than Daniels E-pace for most. If you simply ran that pace, for the same amount of hours per week, do you think your race results would be the same? I think the workouts matter, but so too does the individual and their physiological make-up.
Another observation, and this may just be that many on this thread are from a cycling background and/or are new to running? There seems to be this suggestion that runners have no concept of training load or haven’t thought of this type of approach/mindset before? I can think of several examples just on LRC, going back to the early 2000’s, that would suggest that assessment to be unfair.
Tinman started a thread, “no interval work success” where the whole discussion basically promoted and shared experiences of runners training exclusively in the easy to threshold pace range. John Kellogg practically wrote a book, “The Training Wisdom of John Kellogg”, which a poster compiled somewhere, on how he coached Wejo to a massive 10km pr of 28:06 doing almost exclusively base work (i.e. volume and threshold). Hadd, whose original thread may be longer than this one (?), outlined what is essentially this method, but based on HR instead of lactate. And though it wasn’t on this site, to show that Andrew Coggan isn’t the only PhD with a training stress point system, Jack Daniels published his own point system for tracking training stress in his second edition. Which, bringing it back to LRC, a user programmed into an Excel workbook at the time and shared on this site (jiggymeister, you may want to look for that version haha). There are even more flamboyant (though likely unknown) coaches like Lyle Knudson who’ve promoted near daily threshold running (sorry sirpoc and shirtboy, the experiment has already been run, literally haha).
Lastly, sirpoc leading on that he’s not scientific is a bigger troll job than JS Wizard ever leveled on this thread haha. He’s far more keen minded than he credits himself as.