Power distribution chart

[msummers]

I like to determine my FT using the drop off method and guess why? It's always higher than my NP values. It seems to make no difference if I look at data for the entire season, last 28 days or 7 days or last ride, the difference is distinctive. Recently during a flat race for about an hour with lot's of siting in, my NP was 220w. The distribution chart drop off was 240-250. It doesn't really make a difference if I fine tune the values either (10w-15w 20w). Any thoughts on why these values differ?

MS

 

[acoggan]

I like to determine my FT using the drop off method and guess why? It's always higher than my NP values. It seems to make no difference if I look at data for the entire season, last 28 days or 7 days or last ride, the difference is distinctive. Recently during a flat race for about an hour with lot's of siting in, my NP was 220w. The distribution chart drop off was 240-250. It doesn't really make a difference if I fine tune the values either (10w-15w 20w). Any thoughts on why these values differ?

MS

My first question would be, have you tried to determine your functional threshold power using any of the other five ways that I have suggested? The 'drop off' criteria can often give you a reasonable estimate, but it doesn't always do so. Likewise, using normalized power from a hard ~1 race typically yields a value that is w/in ~5%, but sometimes over- or underestimates.

Anyway, assuming that your functional threshold power is really 240-250 W, the discrepancy between that and the normalized power approach could be due to the fact that you're not going "ten-tenths" the whole race ("...lots of sitting in..."). So, you may be spending significant periods of time producing power at up to your functional threshold power, much less time over it (hence the drop-off), and enough time well below it to drag down the normalized power. Make sense?

 

[whoawhoa]

I like to determine my FT using the drop off method and guess why? It's always higher than my NP values. It seems to make no difference if I look at data for the entire season, last 28 days or 7 days or last ride, the difference is distinctive. Recently during a flat race for about an hour with lot's of siting in, my NP was 220w. The distribution chart drop off was 240-250. It doesn't really make a difference if I fine tune the values either (10w-15w 20w). Any thoughts on why these values differ?

MS

"...flat race for about an hour with lot's of siting in..."

I think you answered your own question right there. The drop off method seems interesting, though.

 

[msummers]

My first question would be, have you tried to determine your functional threshold power using any of the other five ways that I have suggested? The 'drop off' criteria can often give you a reasonable estimate, but it doesn't always do so. Likewise, using normalized power from a hard ~1 race typically yields a value that is w/in ~5%, but sometimes over- or underestimates.

Anyway, assuming that your functional threshold power is really 240-250 W, the discrepancy between that and the normalized power approach could be due to the fact that you're not going "ten-tenths" the whole race ("...lots of sitting in..."). So, you may be spending significant periods of time producing power at up to your functional threshold power, much less time over it (hence the drop-off), and enough time well below it to drag down the normalized power. Make sense?

Yes it does. I have never done a 1hr TT but have performed many 20 minute intervals in the past few months and based on my observations my 20 minute NP power is 235 to 240 which I gather translates to about 223 to 228 FT (-5% factor).
I do live in a very un-flat area and my power graphs are quite lumpy. Could it be that I'm not pushing as hard as I could on the downhill portions of the session? Maybe this would support the "ten-tenths" theory? Could I then make the distinction that with enough historical data, my FT would be closer to the "drop off" method than the NP values factored out on 20 minute intervals? In your previous survey, it appears than many riders do enlist the drop off method for determining FT. I wonder if it is because, like me, they have a hard time correlating the NP interval values to the drop off and just pick the higher of the two.

Thanks
MS

 

[peterpen]

Yes it does. I have never done a 1hr TT but have performed many 20 minute intervals in the past few months and based on my observations my 20 minute NP power is 235 to 240 which I gather translates to about 223 to 228 FT (-5% factor).
...

I thought it was 95% of Pave not Pnorm?

 

[NM87710]

Maybe I'm all whack here but why use 20NP*0.95 for FT?

Semi-OT: What is the benefit of having 5 different methods(or more) to determining FT? Seems continual NP/AP/FT calculation questions(most common question in "power" discussions) implies a lot of ambiguity in determining FT. Most power users want an "easier"(not 60min steady state TT-too hard?) to deterime FT so I'm guessing that's why there are various ways to determine FT. Why not define FT as 20AvgPower from a steady state TT course-flat or up hill? Short enough for everyone to do often while being easy to understand and interpret results. Then maybe ACoggan won't spend the rest of his life answering NP/AP/FT calulation questions

 

[acoggan]

Why not define FT as 20AvgPower from a steady state TT course-flat or up hill?

Because a 20 min - or even a 30 min - effort is short enough that anaerobic capacity still plays a small, but potentially significant, role in determining how much power you can sustain. IOW, it isn't a sufficiently "pure" surrogate for power at LT, such that it can lead people to focussing on the wrong thing in their training.

 

[Woofer]

I do live in a very un-flat area and my power graphs are quite lumpy. Could it be that I'm not pushing as hard as I could on the downhill portions of the session?

Can you find a longer uphill just for testing? It's easier for me to find a longer hill than a safe place to test on rolling hills that you are describing, so YMMV.

 

[msummers]

Can you find a longer uphill just for testing? It's easier for me to find a longer hill than a safe place to test on rolling hills that you are describing, so YMMV.

Thats always been a problem here in southern Wisconsin where there are no real long hills or mountains just plenty of short steep rolling terrain (damn those receding glaciers). Steady state routes are somewhat hard to find that can last 20 minutes. In a five minute stretch you can easily go well over your FT a dozen times or more just to keep a decent momentum. Pacing can be difficult at best. Which made me think that the stochastic nature of riding this type of terrain may make it a) hard to focus on a constant steady power output, and b) cause inconsistent results from interval training data on different routes depending on the degree of how "wavey" the road is.

 

[NM87710]

IOW, it isn't a sufficiently "pure" surrogate for power at LT, such that it can lead people to focussing on the wrong thing in their training.

So does using FT=20AP*0.95 as a surrogate for power at LT potentially lead people doing the wrong training path as well?

 

[acoggan]

So does using FT=20AP*0.95 as a surrogate for power at LT potentially lead people doing the wrong training path as well?

I believe it can. That's why I don't really recommend using that approach (even though I know that Hunter likes it).

 

[WarrenG]

I believe it can. That's why I don't really recommend using that approach (even though I know that Hunter likes it).

He probably doesn't mind its *shortcomings* because they are easily accounted for when planning and evaluating someone's training. He'll base the training prescriptions on what the person does during a variety of training efforts and races. The accuracy of the number referred to as FTP is of secondary importance.

 

[acoggan]

He probably doesn't mind its *shortcomings* because they are easily accounted for when planning and evaluating someone's training.

If you truly understand the physiology of exercise and training, yes. However, not everyone does, and even those who have raced at the professional level and are now enrolled in medical school can fall into the trap of mistakenly believing that maximizing 20 min power automatically means that your LT will be maximized as well.

He'll base the training prescriptions on what the person does during a variety of training efforts and races.

He will if he listens to me :

"...as with all systems training exercise prescriptions should be individualized, in this case taking into account the power the athlete has generated in previous similar or identical workouts...the primary reference, therefore, is not to the system itself, but to the athlete's own unique (and current) ability. In this regard, the classification scheme described above should be viewed primarily as an overall framework, not a detailed plan."

http://lists.topica.com/lists/wattage/read/message.html?mid=901936066&sort=d&start=100

 

[WarrenG]

If you truly understand the physiology of exercise and training, yes. However, not everyone does, and even those who have raced at the professional level and are now enrolled in medical school can fall into the trap of mistakenly believing that maximizing 20 min power automatically means that your LT will be maximized as well.

Because of some of the previous posts by others, I was mainly adding to your point about the importance, or actually lack of importance at getting the FTP exactly right-the calculated number representing more of a starting point for suggesting training (power, HR, PE).

 

[acoggan]

I was mainly adding to your point about the importance, or actually lack of importance at getting the FTP exactly right-the calculated number representing more of a starting point for suggesting training (power, HR, PE).

So to add to your comments as well: keep in mind that functional threshold power is useful for more than just defining training levels, and these other applications require more than just a ballpark estimate.

 

[rmur17]

So to add to your comments as well: keep in mind that functional threshold power is useful for more than just defining training levels, and these other applications require more than just a ballpark estimate.

TSS and TSS++ come to mind ...

Accuracy is important but consistency/repeatability perhaps even more so. I've settled on estimating FT from shorter efforts but I have a 3-year database to back that up. AP still the gold standard but NP PB's always get me thinking ... (or testing).

rmur