TSS and IF? What is the difference?

Hej everyone!. Can anyone tell me what TSS and IF stands for??



Always go with the original source:

ETA: In this case, especially important, since the TR article perpetuates the myth that FTP = 60.0000…min power.


I still don’t understand how you arrived to this classification, why those specific cutoff points?

  • Less than 0.75 recovery rides
  • 0.75-0.85 endurance-paced training rides
  • 0.85-0.95 tempo rides, aerobic and anaerobic interval workouts (work and rest periods combined), longer (>2.5 h) road races


So if you were to do a constant power workout at your FTP (so IF = 1) for exactly one hour, TSS needn’t equal 100, it could be more or it could be less .

No. Definitionally if IF = 1 for 60 minutes, then TSS = 100

This relationship doesn’t care what you think the definition of FTP is, it is how TSS is defined

Not mathematically.

ETA: Perhaps your intended point was that if you pedal as long as you can at exactly your FTP, your TSS might be more or less than 100? If so, the answer is obviously yes, but so what? TSS still helps put everything on the same/a similar scale, much better than, e.g., total work. That’s doubly true when you consider that TSS is based on NP (with all its attendant limitations), not average power. IOW, you’re already talking about an edge case here.

1 Like


1 Like

Can you elaborate? How do you go from observation (of what?), to precise %’s?

I (and others) analyzed many power meter files from many cyclists, observed how things played out, then described the result of this analysis using easy-to-remember guidelines:

“Typical IF values for various training sessions or races are…” (emphasis added).

How typical? Well, 20 y ago on the wattage list we used to play a game of “name my ride”, in which someone would provide their TSS and IF and folks would take turns guessing whether it was a race or just a training session, the terrain, etc. More often than not, it was possible to do so just from those two values.


Why don’t the typical IF values quoted above align with the power zone definitions? I.e. Based on the power zones I’d expect a recovery ride to be <0.55IF not <0.75. And an endurance paced ride to be 0.56-0.75 not 0.75-0.85.

1 Like

If one accumulates 500 TSS in a week, is that the equivalent of 5 hours at FTP?

If FTP were equal to 60 minute max, that would be so, but we know that’s not what FTP is. So that interpretation doesn’t work at 500 TSS, or at 400 TSS, or at 300 TSS, or at 600 TSS. But you’re saying it works at 100 TSS.

Maybe the riders being observed very usually sprinkled a handful of hearty 15 second sprints into their endurance rides?

That would do it :slightly_smiling_face:

Because, as I have pointed many times before, my original (descriptive!) training levels were skewed downwards to account for the variability in power output that normally occurs when riding outdoors.

In contrast, IF is based on NP, which attempts to correct for such variability “up front”.


If two cyclists can generate the same power at maximal physiological steady state, but one can maintain that power longer than the other, who experiences greater physiological strain as a result? (Obviously that’s a rhetorical question.)

If FTP were defined as exactly 1 h power to satisfy the mathematical pendants, what would that mean when FTP is employed for power profiling? For quadrant analysis? For elevation-corrected power?* For calculation of dFRC? (Also rhetorical questions.)

*Still something that only WKO does logically.

1 Like

I live and train at 8.5k feet. I was getting excited, however, it’s useless for my purpose.

WKO’s elevation correction assumes that mFTP is based on data collected at or near sea level. If most of your training is done at elevation already, using elevation-corrected power (ecpower) will not produce accurate results for you.

Any suggestions?

Elevation-corrected would still be a cogent estimate of what you could do at sea level.

The article is also wrong in claiming that mFTP assumes that the underlying power data are from sea level. The model is in fact totally agnostic.

What you don’t want to do is *mix-and-match".

For example, if someone like you travels to sea level for a race, you’d be likely to generate some new power PBs, which would alter your P-D model parameters. Such data would need to be excluded from the curve fit to get valid results.

Similarly, someone like you wouldn’t want to use elevation-corrected to calculate TSS. In that application, the effect of elevation is essentially self-canceling, as it lowers both your FTP and the NP of training sessions or races.

There are fairly simple solutions to these issues, but my concepts for dealing with changes in elevation were never fully implemented in WKO4.


You are misinterpreting what I said: I said that TSS is defined as IF * 100 * (time in hours). And IF = NP / FTP.

These are mathematical definitions. Argue however much you want over how to define FTP, that doesn’t change the fact that TSS = IF * 100 * (time in hours) definitionally

thank you.

IF squared x 100 x hours.

1 Like