Hi
Does anyone know the formula used to create the TRIMP value?
Thank you,
Karl
If you click on TRIMP in the activity, it opens the explanation and a link to a research paper.
https://www.researchgate.net/publication/20910238_Modeling_human_performance_in_running
Yes, I’m aware of that paper but unless I’m using the formula incorrectly, the value I calculate is significantly higher than the one shown in Intervals.icu.
Formula used -
w(t) = (duration of exercise)* ((HRex - HRest)/(HRmax-HRest)) * Y
I have also used the above formula without the weighting value, Y, and this gives a much smaller value of TRIMP.
For example - HRmax = 166, HRrest = 44, HRex = 139, Exercise duration = 240 mins, male,
gives 240 x .779 * e^(1.92*.779) = 833. The value shown is 534.
For the activity I shared with the TRIMP value of 98,the following works out to 98.39022
Duration (D) = 66.2 (minutes)
HRmax = 187
HRrest = 44
HRex = 138
HRr = (HRex - HRrext)/(HRmax-HRrest)
y = 1.92 * HRr
In Excel, TRIMPexp = D x HRr x 0.64 x EXP(y)
I used the same Excel file to calculate a 3h30m ride, and I get the same TRIMP value (rounded down).
Knowing that my HR was clipped on that ride, I pulled a run instead.Same result.
Thanks for showing your formula, there is a scaling factor of .64 which is not shown in the Morton/Banister document. I think it is used to align the TRIMP value to the values created in other software (TrainingPeaks for example) - see Possibly wrong TSB zones in PMC/Fitness Trend if using HR instead of FTP · Issue #341 · thomaschampagne/elevate · GitHub.
Using the corrected formula gives me the same value in Excel as Intervals.icu.
Regards,
Karl
2 Likes