Yeah, I only looked at the Strava one to compare… I don’t usually look at it either.
I’m not saying there’s anything wrong with the TP HR one - they’re just different. The Intervals HR one is easy to understand as well, especially given it’s free. Any HR based method is more open to variability but that doesn’t mean it’s not possible to train with - people did it for years. I did similar to you in 2016 using nothing but Strava Suffer Score . I started training with power (=using TR) in October that year, but I’ve still not beaten some of my segment PRs from that year.
(I might ask @davidtinker why the proper TSS is lower in Intervals sometimes though - it could be as simple as it’s using my eFTP for the calculation, which is 4 watts higher than the one I’ve set, but the TSS only varies on outside rides, not on inside TR workouts so that would be weird).
Tx @bobmcstuff . Intervals.icu TSS excludes coffee stops because it doesn’t make sense to me that your TSS will be higher for a ride if you add 30 mins in the middle having coffee. So that would be one reason why it is lower sometimes. Apparently this is the burrito break discussion on wattage (before my time). It always uses your manually set FTP.
With regards to HR TSS in Intervals.icu you can choose between 3 models:
Time in zones: Regression model based on time in zones for your previous rides with power and HR. Works well if you have those and your HR and power zones are aligned. Sometimes overestimates low intensity activities if you don’t have a lot of Z1 riding with power
Average HR: Simple model that looks at previous rides with power and HR and works out TSS/hour for a given average HR.
HRSS (normalised TRIMP): This uses your resting, threshold and max HR and apparently is what Elevate uses. It’s nice because you don’t need any power data. For me time in zones produces better results but I have loads of power and HR data and my zones are aligned.
Intervals.icu will also estimate TSS using pace data for running and swimming.
You can also always choose to enter whatever load you like. The dialog shows the estimates from power, HR and/or pace as a guide.
This will be why I got such a high estimate for the low intensity ride last Monday then - makes sense.
Is that also likely to be similar to TP?
I usually turn my Garmin off if I have a long stop (cafe or shop) - I assume that would also be excluded from Garmin, TR calcs too (looks that way in the TR ride profile). But this ride included 5-10 mins waiting at the start plus maybe another 5 waiting at the top of a climb, so probably 15-ish mins of stopped time which I guess would be excluded.
. I started training with power (=using TR) in October that year, but I’ve still not beaten some of my segment PRs from that year.