What is the conceptual, physiological basis for NP calculation?

what is the theory? ...why, how... does that calculation = ~the physiological demands of a ride?

"1) starting at the 30 s mark, calculate a rolling 30 s average (of the preceeding time points, obviously).

2) raise all the values obtained in step #1 to the 4th power.

3) take the average of all of the values obtained in step #2.

4) take the 4th root of the value obtained in step #3."

 

Coggan's NP is a derivation of Bannister et al. TRIMP scores. The aim is provide a more accurate way to track the physiological cost of your efforts.

1.) The 30's rolling average attempts to smooth your power profile while considering the delay in physiological responses (the physiological response to a given effort is not instantaneous and many of these responses have a half life of 30's).

2.) Long before NP was established, scientific research identified a curvilinear relationship between exercise intensity and blood lactate levels. Coggan identified a 4th power relationship between power output and blood lactate response. IOW a small increase in power output correlates to a significantly higher increase in lactate or physiological cost.

3&4.) You might be familar with the term root mean square (square root of the sum of squares), which is typically used for determining the mean of wave like data. In this case we are raising power values ^4 as described in 2.) so we have to take 4th root to get the mean.

Some links that might help clarify:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769631/
http://www.physfarm.com/govss.pdf

 

thanks bmoberg337...

so basically smoothing/averaging is applied such that approximates what is observed... and the 30s comes from the generally observed half life of the physiological response... mostly... let's just smooth these point out... just wondered if there was more too it than that..

 

Quote: Originally Posted by bmoberg337 .
3&4.) You might be familar with the term root mean square (square root of the sum of squares), which is typically used for determining the mean of wave like data. In this case we are raising power values ^4 as described in 2.) so we have to take 4th root to get the mean.


The arithmetic mean is the L-1 norm; the root mean square is the L-2 norm; and the quartic root of the mean of the 4th power of a sequence is the L-4 norm. In general, the pth root of the mean of the pth power of a sequence is the Lp norm. That's where the name "normalized power" comes from.

Some links that might help clarify:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769631/
http://www.physfarm.com/govss.pdf


Or in Coggan's own words: http://anonymous.coward.free.fr/wattage/coggan.pdf

 

We have been through this before. NP was developed as a marketing tool for TrainingPeaks.com. There is no science behind it. At least acoggan (Andy Coggan) has said so.

It is easy to demonstrate for yourself that there is no substance.

Find a nice hill where you can ride for 45-60 seconds at 160% FTP up and 0% down. Ride for an hour. Your average power will be about 80%. Your NP will be 100-110%.

Compare the physiological demands of this ride to the physiological demands of riding at 100%FTP for an hour.

When I did this comparision, the 160% intervals was easy enough to do for hour after hour. Well at least as long as I can ride at 80% which is 6+ hours. While the 100% constant pace was really hard at the end of 1 hour.

 

Interestingly the Old Troll has been unable to demonstrate any of what he claims.

 

I am sure if you ask the Old Troll to do a demonstration, tell him what the demonstration needs to be, and pay him for his time, he will be willing to demonstrate.

---

I have several times offered to demonstrate for acoggan. But he has said that demonstrations do not refute his religious doctrine. There is not much sense in doing demonstrations in that situation.

---

If you would like me to do demonstrations for you, I think I have a few months before business and family affect my schedule. But you should give Old Troll first refusal.

 

If you tuned into the pitches for the WK0+ 4 software (available on trainingpeaks/youtube) then you would have heard Mr. Coggan openly admit the limitations of any mathematical modeling system. Anyone who has ever tried to make a mathematical representation of data, be it your 6th grade science experiment or stock market prices, should be aware the governing assumptions and limitations of that model.

As long as you aware of those limitations and are consistent in your use of a given modeling system, then it can provide useful for future application. The usefulness is really in the hands of the user.

 

Or even show a single file. 160% of FTP for 60 secs on 30 secs off for an hour is pretty much impossible in real life unless you've set your FTP so low as to deliberately create some data that manipulates TSS and IF to suit your (trolling) needs.

 

As I understand this you want a test that determines my FTP and then a data file of a bike ride.

The test is one day. The bike ride is another day. I suspect you want some proof of the test results and some prrof of the ride.

I can do that. Tell me what you want. I will tell you what the cost will be.

---

Just for reference 60 seconds at 160% then 30 seconds at 0% is 103% average. I said the average was about 80%. I think you have your time intervals wrong.

45/45 or 60/60 seem to give 80%.

---

acoggan claims to have a list of about a dozen people who have produced data along these lines that he seems to accept as valid.

 

acoggan (Andy Coggan) has said it is not science.

Wrapping a pitch in words of science does not make the pitch science. The pitch is still a sales promotion for TrainingPeaks.com.

---

But to the point of what Andy Coggan claims of NP, TSS and the items: He claims they are a measure of fatigue. That is provably wrong. The simplist proof is to point out that Andy Coggan's definitions of fatigue are numerous and contradictory. Pick the definition you want, Andy Coggan has already said you are wrong.

---

As for what you claim about these terms: If you publish a paper with claims and have it peer reviewed and published, I will address your claims. (Andy Coggan cannot get a peer reviewed paper on this topic.)

 

http://home.trainingpeaks.com/blog/article/normalized-power,-intensity-factor-training-stress

 

yeah, I read that... didn't really help... it doesn't explain say HOW this algorithm achieves that plainly or at all really... it just says the goal of the algorithm... thanks to bmoberg337 it's apparent that the algorithm simply smooths out the points so they more closely match physiological response... why didn't they just say that? ... honestly it sounded a little like they were 'clouding the waters to make them appear deep' a little... also doesn't go on to mention the limitations of NP in that a person with good AWC, doing many high intensity efforts can produce points can overwhelm that smoothing (NP buster) etc...

 

You used a term created to hide the fact that there is no substance to the theory. In science an "NP buster" would be proof that the theory was invalid.

---

But you can produce an "NP buster" of your own.

A lot of guys do 10-20 minute intervals for an hour ot two. They get some NP for their efforts. If they were to stack their hard intervals together at the beginning and the rest periods at the end, they should get just about the same NP.

All you need to do to produce an "NP buster" is to do intervals for a day. Remember how "fatigued" you are. A couple days later try to stack the hard parts at the beginning. Rejoice in the results.

---

I am out of pocket for the past couple days. With nothing to do, but I have my trainer and bicycle. So I go out to the local park, set up, and do 30 minutes above FTP twice a day - 60 minutes total. It is not difficult, my heart rate has not been above 170. But if were to do 60 minunte straight at FTP, the last few minutes would be hard and my heart rate would peak above 180.

Save IF, same NP, same TSS. But different in any manner I would care about - I could do 3 of these at 100% a day, but I could not do 90 minutes straight at 100%. See how easy it is to produce an "NP buster."

Now go out and do one yourself.

 

hi An old Guy... still here string the pot eh?

all of this stuff is just an approximation... even in physics, there is still no grand unified theory.. so in certain situations... like the first few nano second of time, the physical laws don't work... that doesn't mean we need to throw out all of those laws... we just need to know their limitations... newtons laws work most of the time when dealing with things on human scale, but they are just and approximation and don't work on the atomic and sub-atomic scale etc... wave mechanics do a much better job be (see above) even they are just an approximation and don't work in all situations (see above)

yes most of this stuff is not even scientifically verified.. but, if you know the limitation it seem to make sense and work a good portion of the time.

I won't respond again.. I know how you operate... so the floor is all yours... have a good one!

 

I think the Old Turd is a troll, but you should choose a different comparison than physics. What you just said doesn't hold water.

 

Post 11 of this thread seems to deal with all of your issues with this subject.

It will be good to not hear from you.

 

ok, you've caught my interest... how so?

 

I will be rude and answer for him.

In science one does experiments that determine under what conditions a theory is accurate enough to use.

With regard to NP, TSS and all the other concepts, there can be no experiment because the theory is not defined. That is, what is being measured is never defined. More properly many definitions are given by Andy Coggan.

---

In addition to other claims of measurement Andy Coggan claims that some property related to glycogen or lactate is being measured. We do know that properties of neither glycogen nor lactate are being measured. The time scales are wrong.

 

Just the part about a grand unified theory. The lack of a GUT has no impact on the validity of current theories. Those theories are theories because they've gained scientific consensus after much experimental testing. There is also no requirement for a GUT. Sure some, maybe a lot, of folks would like to have one, but there's nothing in physics right now that indicates that we need a GUT. There is an innate human desire for simplifying things, and a grand unified theory of physics would certainly beautifully simplify physics in general (by having one over arching theory, not by making things mathematically or conceptually easy to understand. If there ever is a GUT, it may not accomplish those last two bits), but the reality may be that nature cannot be explained that way. We may have to be satisfied with a patchwork of theories to cover physics...or in the case of this discussion, the fundamental forces of physics.