CP - Mean Max NP Curve

Originally Posted by djconnel

what an NP-based optimal pacing says is after that 30 minute near-maximal interval, you want to rest for a fraction of tau to get smoothed power down, then you can bring it up to no more than (f * n)^(1/n), where n is the exponent and f which is the additional fractional contribution to NP. For the second half-hour to contribute no more than 0.1% to the NP: for n=32, this is 90%, for n=12, this is 70%, for n=4, this is 25%.

Originally Posted by acoggan

...which is why the normalized power algorithm can suggest impossible pacing strategies if you try to apply it to courses that haven't been divided into small enough segments. OTOH, the alternative smoothing and weighting that you suggest would result in grossly overdamped power when applied to the modeling of courses that have. Again, this conundrum stems from the fact that the power-duration curve isn't well-described by a power function, at lesat not over it's entire span. That's why I say it's more logical to approach the issue from a physiological perspective, and see what falls out (which turns out to be an algorithm that works quite well).

Originally Posted by djconnel

Right. Which is why I started my plot at 50 seconds...

Originally Posted by djconnel

In any case, a feature of high exponents is they essentially pick out regions of maximal smoothed power. An exponent of 32 means that if an interval has a power only 5% less, it counts only 19% as much. So basically it's estimating FTP on the hardest interval long enough to poke its nose above the smoothing. Even if you did 5 intervals at the same power, the multiplicity of intervals would boost NP only 5% versus one interval at the same power. This is all about estimating FTP and optimizing pacing, not about traning stress. Optimal pacing still involves non-uniform power for changing conditions, but that's more playing with power smoothing than trading off steady-state power with such a high exponent.

Originally Posted by jetnjeff

I think it may over estimate TSS as well. I have had teaching rides where I coach new riders on technique, where I through in a few VO2 intervals and wait for them at the top. These rides come off with a pretty high NP, but I do not feel nearly as wiped as after a hammer ride with my training partner with very limited stops. FYI this is MTBing. The technique ride was 10/21 NP= 230 for 2:24 IF= .999 AP = 214. Hammer ride was 10/21 NP=265 for 2:53 IF = .975 AP =230. Both days I had top tens in 1, 2 and 3 minutes for the last 17 month time frame.

Originally Posted by Alex Simmons

There's something wrong with your numbers.

How can you have:
NP=230 for IF=0.999 and then have
NP=265 for IF=0.975 ?

Can you really ride with IF = 0.999 for 2.5 hours? Suggests FTP is not set correctly.

Originally Posted by jetnjeff

I Miss typed. TSS was 230 and 265 NP was 312 and 304 respectively.

I was speaking to the point that TSS seemed inflated for the TSS and NP were over stated for the 230 TSS ride.

Correction then would be:
The technique ride was on 10/21 NP=312 TSS=230 for 2:24 IF= .999 AP = 214.

Hammer ride was 10/17 NP=304 TSS=265 for 2:53 IF = .975 AP =230. Both days I had top tens in 1, 2 and 3 minutes for the last 17 month time frame.

I have been wonderng about the FTP too. I used the 1st hour of a 2 hour Climb In In the Smokey Mts as a test on 9/12. Got 312. Which is almost exactly what I got Last spring indoors on a trainer for an hour. My 2x20s on my usual route have only gone up a few Watts. My VO2 CPs are above the Cogan Levels. Example I got a 4 minute at 412 recently. L5 = about 333 to 377.

Originally Posted by Alex Simmons

Keep in mind that NP, when used for assessing efforts near FTP, is suggested by Andy Coggan to be within 5% which for you is ~ 15 watts.