Yes I just started it and find that the training stress for running seems to be significantly understated. I'm only running a couple days a week so it's not a huge deal but I don't think the training peaks approach would be very useful if your primary focus was running. It's a shame because just counting up mileage is kind of a barnyard approach to measuring the stress of running too but maybe that's the best we can get.

Running TSS is not TSS. Running TSS is rTSS. 1 rTSS does not equal 1 TSS. Same goes for swimming (sTSS). Track cycling TSS, rTSS, and sTSS independently of each other.

If you want a better measure of rTSS, get a stryd.

---

blog

What zone system are you using, and are you looking at your zones from pace, HR, or RPE?

I had a Stryd before and never really got into it too much. Maybe now that I only run it may prove beneficial. Especially as I often can't find flat ground. Thanks.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

When I discuss I usually talk 5 zones. I actually only use 3 currently.

The zones don't really count for TSS anyway, as far as I know it only takes into account activity compared to threshold, regardless if where you placed zones in comparison to that.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

Maybe you don't have your threshold properly set in Training Peaks?

I notice a difference in runs, although its not a huge # difference. (example, easy 10 miles is around 100 rTss, a tempo run of 2 x 2 miles @ threshold with the rest easy was 110 rTss).

That would make it the other way. My Threshold in TP is currently 3:45 per km. I was running 3:35 intervals, so a fair bit above threshold; which should have (if thats how it works) made the TSS higher.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

I use Strava's "relative effort" with heart rate for all activities (no power meter), which is not exactly the same as TSS but it's similar. Anything heart-rate based is going to saturate or become less useful for harder and shorter intervals, where you may not be in steady state or are bumping up close to max heart rates. So I've found relative effort does not always accurately indicate the difficulty of interval sessions. In particular, for very short intervals, like fast 200s or 400s, heart-rate-based TSS becomes much less useful, because you are putting much more impact stress on your body than your heart rate may indicate. The same might be true for your 1600s.

Running Training Stress Score (rTSS)

• Applicable Workout Types: Run only (TSS can not be calculated for bike workouts from speed)
• Accuracy Level: rTSS is the second most accurate method of calculating a Training Stress Score, and is the default method for Run workouts when there is enough data present to calculate
• Requirements to Calculate: Workout time series with pace and elevation data, user threshold pace for run/walk (entered in pace/speed zones)
• What is rTSS? Running Training Stress Score (rTSS) is similar to Training Stress Score (TSS), but instead of being powermeter based, it is based on pace. Since running pace is a function of power, and for most runners, speed or pace is the measure of greatest interest, we can use the same principles applied in the TSS system to quantify training stress in running. In particular, the elements of the normalized graded pace (NGP, see below), relative to the functional threshold pace. Coupled with the duration of the workout, we can determine an appropriate rTSS score for each workout. rTSS is a metric that can only be used with workouts that are runs, as the algorithm only accounts for the unique physical demands of running. Further details on rTSS can be found here:

Run Training Stress Score (rTSS) Explained

• Normalized Graded Pace (NGP): The adjusted pace reported from a global positioning system (GPS), or other speed/distance device, that reflects the changes in grade and intensity that contribute to the physiological cost of running on varied terrain. For more information on NGP, please see:

What is Normalized Graded Pace?

• Intensity Factor (IF) for rTSS: For any workout or part of a workout, the ratio of the Normalized Graded Pace to the runner's functional threshold running pace, which gives the user a relative intensity in relation to their threshold pace. IF is used to calculate rTSS.

Yep, I have read that. It's doesn't mention Normalized Pace being adjusted like Normalized Power. It mentions

".. For any workout or part of a workout, the ratio of the Normalized Graded Pace to the runner's functional threshold running pace, which gives the user a relative intensity in relation to their threshold pace. .."

So there would seem to be an awareness of the need to adjust for intensity, but they only do it for hills (grade) and not intensity of intervals within a workout I think. I think this kind of screws TSS (or rTSS) being of any use for runners who include intervals unless you start throwing in an arbitrary figure.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

In your running example you gained 15% more TSS per hour whilst doing intervals than whilst doing your easy run. Therefore I would say that in the run example you quoted, the TSS was adjusted for intensity...?

I am a former triathlete turned runner and I use TP Premium for my training. As long as my form is +25 on race day I will always have a peak A race performance. I wear a Stryd pod for all my runs but never look at the data

But intensity is already accounted for given that on flat ground your actual pace is directly relative to threshold pace. The issue with hills is that effort up or down is not as directly relative to threshold pace (so the adjustment is being made, whereas hard efforts on flat ground are already accounted for). Unless I'm misunderstanding what you're saying here.

Tend to agree with above about the difference seeming about right given the difference in activity time (and, what were the efforts between intervals like relative to your steady easy run?)...

It is adjusted for intensity, but not weighted for intensity. It takes exponentially more effort to go faster. This isn't being accounted for. I think it is for the bike, at least that's what is suggested by the NP figure. But there is no Normalized Pace. Just Normalized Graded Pace (NGP)

I think NGP accounts for hills and adjusts the IF accordingly. Easier to go up hills, harder to go down.

But NGP doesn't account for the demands of increased effort intervals, like NP does / should / might.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

Pace should not be directly relative to threshold. That's kind of my point. What we are discussing is Training Load. TSS. The Training Load above threshold is exponentially higher than Training Load under. The current system gives an equal weighting, which doesn't reflect the stresses involved.

They cover this, I think, with biking by using Normalized Power to weight higher efforts with more Intensity Factor.



On the face of it, it may seem about right. But, I could have have done 4 of those easy runs back to back; whereas on the interval run I finished actually on my back. I was due to do 5 intervals, and only did 4 (partly because I was experimenting with paces - which is what lead me down this rabbit hole in the first place).

When you look at numbers, distance, time .... it might look reasonable. If you go out and do those two workouts and physically experience what is involved, they are miles apart. Both in physical toll at the time and recovery. And (I think) that's what TSS is supposed to account for ..... but it doesn't seem to.

I appreciate that there are shortcomings and that's it's all an arbitrary figure anyway. BUT ..... Training Peaks is kind of built on TSS, I just thought / hoped they'd do a better job of accounting for interval sessions which are an integral part of any runners schedule.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com