How much stock do you put in the daily training score?
About half a chickens worth.
It’s more sensitive to intensity than the traditional TSS score.
For example, 1 hr at threshold will give you 100 TSS, as will 2 hrs at 70%. But obviously 1 hr at threshold is much harder on your body, even though TSS suggests they are equivalent. The Garmin/Firstbeat will at least give you a much higher score for 1 hr at threshold.
An even better example would be a 30-40 minute session of anaerobic or VO2Max intervals. The short duration and recovery intervals means the TSS won’t be that high, while the Garmin/Firstbeat score will better reflect how hard these short very hard efforts are on your body.
On the flip side, for long steady-state efforts, Garmin/Firstbeat can give you a score that is too low.
It is based on an estimate of Excess Post exercise Oxygen Consumption (EPOC), on which there is little transparency. I had a Suunto Ambit3, which at least allowed you to see EPOC, and how it was responding to the intensity of effort, and the duration of recovery intervals. Unfortunately the Garmin devices don’t allow you to see EPOC.
Personally, I look at both TSS and Training Load. I think both scores have their limitations, but together they should capture the impact of your workout.
About half a chickens worth.
Sounds delicious
.
I think it works like the Magic 8 Ball. Random answers float around in blue liquid and pop up at the end of each workout:
ProductiveMaintainingResults Fuzzy: Ask again later
I generally don’t put any stock into these sort of metrics as they don’t seem to have much truth or consistency. TrainingPeaks still has my threshold pace at 7:55 per mile and I ran 6:40 per mile for the last hour of the last marathon I did. Garmin has my VO2 Max in the mid 40s with a 19 minute 5k. Also apparently a hard hour on the bike is only half the TSS of an easy hour swim. It’s all nonsense if you ask me.
About half a chickens worth.
Sounds delicious
Sorry, half a tofu chickens worth? 
I think it works like the Magic 8 Ball. Random answers float around in blue liquid and pop up at the end of each workout:
ProductiveMaintainingResults Fuzzy: Ask again later
That’s funny!
What you described is what prompted me to ask the question. Short trainer ride that prompted a high score followed by a hard group ride that showed a low score…
Sooo tofu chicken it is!
I generally don’t put any stock into these sort of metrics as they don’t seem to have much truth or consistency. TrainingPeaks still has my threshold pace at 7:55 per mile and I ran 6:40 per mile for the last hour of the last marathon I did. Garmin has my VO2 Max in the mid 40s with a 19 minute 5k. Also apparently a hard hour on the bike is only half the TSS of an easy hour swim. It’s all nonsense if you ask me.
Ummm…
Training peaks training load metrics aren’t in the same category as the garmin nonsense. Tss is simply a math of your pace / threshold.
TP won’t change your thresholds automatically. That is your job. So, if you think your run threshold is 6:40 or so… Then change it.
Swim TSS is based on you swim threshold pace, which should be set to your all-out 1000TT pace per 100. So, if the TSS for an easy swim is too high, fix it.
Yes the epoc (firstbeat crap) based stuff is mostly garbage.
Perhaps this information is somewhere but i haven’t found it.
I’d like to know what Garmin’s criteria is for calculating training effect to understand what they are going after.
Is it time, intensity, workout mix, frequency, combined load, a combination plus more?
It might very well have some science behind it (which we may or may not agree with) I’d like to know.
https://www.firstbeat.com/en/science-and-physiology/white-papers-and-publications/
Last paper in the list.
I think it works like the Magic 8 Ball. Random answers float around in blue liquid and pop up at the end of each workout:
ProductiveMaintainingResults Fuzzy: Ask again later
Yeah my Garmin is usually mad at me and says I’m “Overreaching.†Lately it has said “Productive†after I rode 407mi in 5 days so that’s something. I see “Peaking†too.
The most useful metric on my Garmin watch though is the performance metric when running. If it is +2 or +4, I know I can amp up the speed or distance no problem. It goes down to -4 for me and I know I have to hold proper form and be careful as I know I’m more injury prone. Sometimes the -4 number has pissed me off and I’ve extended the distance just to say ‘f u Garmin’…
As noted above I think the firstbeat metrics are mostly trash… For Multisport.
I’ve been singlesport /running all fall, and I’ve seen a much better correlation during this period than when I’m SBR. given the papers on their site and how they generated their algorithms, that probably makes a lot of sense.
Most of their metrics are based on using r-r intervals to estimate respiration rate, vo2%, and epoc. Note that their reported r**2 for epoc is ~0.79. There is quite a spread in the data.
You’re going to make me read? 
Edit: Thanks, it’ll take me a while.