BikeTechReview.com

  • Increase font size
  • Default font size
  • Decrease font size
Home Forum
Welcome, Guest
Username Password: Remember me

Rick Murphy's Normanalized Power Thread
(1 viewing) (1) Guest

TOPIC: Rick Murphy's Normanalized Power Thread

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24126

  • Ron Ruff
  • OFFLINE
  • Platinum Boarder
  • Posts: 894
  • Karma: 0
Nicko wrote:
I have a feeling that a lot of people know/understand this but think it is inconsequential, so they're quiet.


Maybe futile is a better word...

I've been trying (and succeeding) to bust GP on the trainer lately. The last one was 1min on, 1min rest (but not coasting), at low cadence-high force. First time I've ever tried it and GP beat my TT power by 10%. In fact even the AP in this test was slightly better than my TT power!

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24129

  • Jack Watts
  • OFFLINE
  • Platinum Boarder
  • Posts: 331
  • Karma: 0
Ron Ruff wrote:
Jack Watts wrote:
To me, it's really not that complicated: do as much intensity as you can until progression levels out/stops


How exactly do you do that? If you aren't showing performance improvement, then you train less?


Yep, pretty much. That, or I change the primary training stimulus.

I organize my training in 6 week blocks, focusing on one particular area (say threshold power), doing 3 workouts a week targeting certain intensities, along with 2 other workouts hitting some other intensities. I do the 'on' portions as hard as I can for the duration, and at both the individual workout level, and a more 'macro' level, when the numbers level out or decline, I cut back--figuring I need more rest.

I've retrospectively gone back to see if the gizmo metrics offered any additional insight into my training (since I'm using the software anyway), and my honest assessment is 'they don't'.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24131

  • Ron Ruff
  • OFFLINE
  • Platinum Boarder
  • Posts: 894
  • Karma: 0
Jack Watts wrote:
I've retrospectively gone back to see if the gizmo metrics offered any additional insight into my training (since I'm using the software anyway), and my honest assessment is 'they don't'.


Does the software give you any useful information?

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24132

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
Kirk wrote:
rmur wrote:
I can get it - but I know the 'steps' were long - much longer than a typical ramp test ... i.e. pseudo steady-state?

what's the applicability? Oh man, you know I can't repeat all that ex. phys. jazz ...snipped...


Please do get that protocol and data.

As for the applicability issue, I asked about that both here and at wattage years ago (more than once IIRC), and there was no response. What is your general perception of the applicability issue?


Any chance you know the protocol, have access to the data, or have created a general perception about the applicability issue?

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24133

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
Ron Ruff wrote:
Well... I wasn't confused either back in the early days... but then I had a very severe episode ~1990 where I was chronically sick, sore, and fatigued for over a year... and ever since then I sink into a similar thing (though much less severe and for not as long... usually a few days) randomly. Training load seems to have some effect... but honestly it can happen if I haven't been doing anything too. I guess I was hoping that there might be some way to track things to avoid having this occur. But if I'm honest with myself, that isn't possible.


I think a reasonable, basic principle to follow is to every week determine if you had at least one good day (regardless of the type of workout). If that didn't happen, I think one should be prepared to add some extra rest days...no matter what caused what I would call an unusual, and persistent increase in perceived exertion for a given effort. I think it is just as you have experienced, I can't say that the cause is typically related to generalized training load. I seemed to have just as many "bad periods" when training 20-30 hours per week as when doing 4-5 hours per week. Whether the cause is a virus, poor recovery, bad sleep, life events/stress, pushing things too hard in training, etc; the course of action is still the same...rest in the form of either a break in training period or a break from a certain type of training.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24134

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
Ron Ruff wrote:
Nicko wrote:
I have a feeling that a lot of people know/understand this but think it is inconsequential, so they're quiet.


Maybe futile is a better word...


I'm glad to hear that there are folks like yourselves who have the courage to bring up the topic of algorithm/math error over there. The "quiet" observation that you both allude to is pretty interesting (from a psych/sociological perspective) too. Not sure what is up with that. Maybe fear and intimidation play a role?

What does everyone else think about the "Fear" aspect?

Let me guess, the inventor/heavily invested crew over there is invoking the "chewbacca defense" in some form and not discussing the magnitudes of errors across the entire p-d spectrum?
-kraig

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24135

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
Nicko wrote:


The plot I understand, but it isn't very "intuitive". Maybe ramping the "on" peak from 100W up to 1600W over X minutes and plotting cycle "weightedP" and cycle AP against time would make it more obvious?


I think I understand what you are describing, and sounds like a good idea!

Do you think the underlying math error that is inconsistent with fundamental/first principle based physiology will ever be addressed by those that are heavily invested in promoting the math problem?

What does everyone else think about this issue?
-kraig

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24136

  • kirkalbers
  • OFFLINE
  • Expert Boarder
  • Posts: 91
  • Karma: 0
"It's pretty impressive to stand out as particularly delusional in a sport filled with delusional, self-absorbed tools." - JBV

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24137

  • Ron Ruff
  • OFFLINE
  • Platinum Boarder
  • Posts: 894
  • Karma: 0
kraig wrote:
Do you think the underlying math error that is inconsistent with fundamental/first principle based physiology will ever be addressed by those that are heavily invested in promoting the math problem?


I'd still like to see someone come up with a better algorithm... then everybody would be happy (I think). The power weighting scheme is simple and *is* fairly accurate for the conditions that were used to derive it. But unfortunately, since it lacks any physiological basis, it doesn't work so well for other conditions. I proposed using a person's max power/duration numbers in a much more complex algorithm that takes into account what is really going on. It would take somebody who understands the physiology though, which isn't me.

I know you don't believe there is any point... but many people do, and you'd probably admit that the placebo effect can work wonders... ie if you believe that tracking training load is important, then it will be. And in that case does it matter if it is done accurately or not? Maybe not so much...

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24140

  • Jack Watts
  • OFFLINE
  • Platinum Boarder
  • Posts: 331
  • Karma: 0
Ron Ruff wrote:
Jack Watts wrote:
I've retrospectively gone back to see if the gizmo metrics offered any additional insight into my training (since I'm using the software anyway), and my honest assessment is 'they don't'.


Does the software give you any useful information?


I wouldn't say it 'gives me' useful information, but that it's functionality makes it much easier for me to find the useful information which I utilize (time spent at certain power levels, trendlines of historic peak power levels, kj expenditure, etc). There's a lot of additional fluff I'd rather not be buying with the software, but to me at least, it's still worth it.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24141

  • Jack Watts
  • OFFLINE
  • Platinum Boarder
  • Posts: 331
  • Karma: 0
kirkalbers wrote:


I hadn't been over to the 'gizmo training software' list in quite a while (I'm still boycotting it...), but I had to check it out just to see if anything new has come up. Sure enough, there are always two popular topics: normonalized power and Rick Crawford My capillaries were bursting just from reading. Good times.....

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24142

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
kraig wrote:
Nicko wrote:


The plot I understand, but it isn't very "intuitive". Maybe ramping the "on" peak from 100W up to 1600W over X minutes and plotting cycle "weightedP" and cycle AP against time would make it more obvious?


I think I understand what you are describing, and sounds like a good idea!

Do you think the underlying math error that is inconsistent with fundamental/first principle based physiology will ever be addressed by those that are heavily invested in promoting the math problem?

What does everyone else think about this issue?



Is this close to what you were thinking:
[attachment=0:2iuh6btp]<!-- ia0 new_physiology.png<!-- ia0 [/attachment:2iuh6btp]
-kraig

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24145

  • tigermilk
  • OFFLINE
  • Expert Boarder
  • Posts: 107
  • Karma: 0
FYI, here's a plot of my annual best GP divided by annual best AP for various durations. Years run from Sep-Aug, and 2003 wasn't a particularly good year in terms of absolute power. For the most part, I'd say for me GP and AP tell me the same thing. Yes GP is higher, but frankly I haven't gone out with the intent of doing a 1 to 3 hour time trials. I would expect that getting those numbers closer to 1.0 would be tough (for 2+ hours), however, simply because of the caloric needs. I also think that my ratios aren't extreme since I've got very good aerobic power relative to my anaerobic/neuromuscular power. I could produce a similar chart for a teammate of mine which would likely have spikes to 1.3-1.5 for several durations. He routinely generates what I'd consider GP-buster files, and even he admits there's no way he'd be able to TT at some of the GP values he puts up.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24146

  • rmur
  • OFFLINE
  • Platinum Boarder
  • Posts: 1088
  • Karma: 0
kraig wrote:
Ron Ruff wrote:
Nicko wrote:
I have a feeling that a lot of people know/understand this but think it is inconsequential, so they're quiet.


Maybe futile is a better word...


I'm glad to hear that there are folks like yourselves who have the courage to bring up the topic of algorithm/math error over there. The "quiet" observation that you both allude to is pretty interesting (from a psych/sociological perspective) too. Not sure what is up with that. Maybe fear and intimidation play a role?

What does everyone else think about the "Fear" aspect?

Let me guess, the inventor/heavily invested crew over there is invoking the "chewbacca defense" in some form and not discussing the magnitudes of errors across the entire p-d spectrum?

do you have anything better Kraig or are you still just sniping nearly five years downstream?
GIZMO marketing specialist

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24148

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
Ron Ruff wrote:
Nicko wrote:
I have a feeling that a lot of people know/understand this but think it is inconsequential, so they're quiet.


Maybe futile is a better word...

I've been trying (and succeeding) to bust GP on the trainer lately. The last one was 1min on, 1min rest (but not coasting), at low cadence-high force. First time I've ever tried it and GP beat my TT power by 10%. In fact even the AP in this test was slightly better than my TT power!


Interesting. I think there is something to spreading out the work. One place I've noticed it is at the end of long, steady, "fill" rides. I've only tried it a couple of times, but when the perceived exertion for a given power starts to rise to uncomfortable levels, I've tried to do "spikes" to increase my average a bit. It seems to work for a little bit at least (I haven't tried it for very long since I've always come to the end of the particular ride within a half-hour of starting the spikes) in the setting of nearly constant fill rides.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24149

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
Kirk wrote:
Kirk wrote:
rmur wrote:
I can get it - but I know the 'steps' were long - much longer than a typical ramp test ... i.e. pseudo steady-state?

what's the applicability? Oh man, you know I can't repeat all that ex. phys. jazz ...snipped...


Please do get that protocol and data.

As for the applicability issue, I asked about that both here and at wattage years ago (more than once IIRC), and there was no response. What is your general perception of the applicability issue?


Any chance you know the protocol, have access to the data, or have created a general perception about the applicability issue?


What's up Rick? I suspect that you can easily formulate your perception about the applicability even if you are struggling to come up with the protocol and data.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24150

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
tigermilk wrote:
FYI, here's a plot of my annual best GP divided by annual best AP for various durations. Years run from Sep-Aug, and 2003 wasn't a particularly good year in terms of absolute power. For the most part, I'd say for me GP and AP tell me the same thing. Yes GP is higher, but frankly I haven't gone out with the intent of doing a 1 to 3 hour time trials. I would expect that getting those numbers closer to 1.0 would be tough (for 2+ hours), however, simply because of the caloric needs. I also think that my ratios aren't extreme since I've got very good aerobic power relative to my anaerobic/neuromuscular power. I could produce a similar chart for a teammate of mine which would likely have spikes to 1.3-1.5 for several durations. He routinely generates what I'd consider GP-buster files, and even he admits there's no way he'd be able to TT at some of the GP values he puts up.


My experience is that the GP curves and the real power curves don't match even when doing maximal, constant efforts out past 3 hours. What I don't know is how much higher I could push my GP for those durations because I haven't done any truly maximal variable power efforts for those durations.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24151

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
kraig wrote:
kraig wrote:
Nicko wrote:


The plot I understand, but it isn't very "intuitive". Maybe ramping the "on" peak from 100W up to 1600W over X minutes and plotting cycle "weightedP" and cycle AP against time would make it more obvious?


I think I understand what you are describing, and sounds like a good idea!

Do you think the underlying math error that is inconsistent with fundamental/first principle based physiology will ever be addressed by those that are heavily invested in promoting the math problem?

What does everyone else think about this issue?



Is this close to what you were thinking:
[attachment=0:2asno8lv]<!-- ia0 new_physiology.png<!-- ia0 [/attachment:2asno8lv]


Yeah, it's clear that the math inflates lower average powers more than higher powers. This is particularly interesting since folks can generally also vary their power with greater amplitude when the average power is lower. That's sort of a double bonus eh?

FWIW...I think that often stuff is put forth which is intended to support GP for specific power-durations, but the justification method is confounded by the issue of cumulative "on" power (at least) as we have discussed here before.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24154

  • Ron Ruff
  • OFFLINE
  • Platinum Boarder
  • Posts: 894
  • Karma: 0
Kirk wrote:
FWIW...I think that often stuff is put forth which is intended to support GP for specific power-durations, but the justification method is confounded by the issue of cumulative "on" power (at least) as we have discussed here before.


Could you briefly explain the physiological reasons for fatigue... ie the "fuel" being used at various effort levels, how long they take to replenish, etc. 1st grade level...

AndyF posted this on Wattage:
1) 20 minutes @ FTP + 20 minutes @ FTP (no break) = 54.4 TSS
2) 20 minutes @ FTP + 40 minutes @ FTP/3 + 20 minutes @ FTP = 94.6 TSS
3) 40 minutes @ FTP/3 = 7.3 TSS

I don't have the software and I'm not sure how this is computed, but I'll assume he got the numbers right. It seems that intervals with rest periods gets scored at a higher level than if the rest periods were not there! Obviously FTP/3 is such a low level of exercise that it is not causing significant stress... certainly not over a 20min time period.

If someone wished to compute training stress that is applicable to all situations, seems like a good place to start would be a person's max power-duration #s. Assign a running "depletion" score for each "fuel" depending on the intensity and duration of the effort, and a "replenish" score for the fuels that are being replenished while going easy. I guess there are waste products to consider as well... but it shouldn't be too tough to take a good stab at this and compare to actual race data.

Another thing just occured to me. One reason that short high-power spikes might give you a higher AP is because you are using a fuel that has been relatively "untapped" previously in your ride. In a max effort you wouldn't want to leave any fuel unused. This could possibly have implications for TT riding... maybe it would be faster to include harder efforts of 10-15 sec throughout the race... especially on hills.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24158

  • Kirk
  • OFFLINE
  • Administrator
  • Posts: 1102
  • Karma: 2
Ron Ruff wrote:
Kirk wrote:
FWIW...I think that often stuff is put forth which is intended to support GP for specific power-durations, but the justification method is confounded by the issue of cumulative "on" power (at least) as we have discussed here before.


Could you briefly explain the physiological reasons for fatigue... ie the "fuel" being used at various effort levels, how long they take to replenish, etc. 1st grade level...


I think we still have a lot to learn about fatigue, and I’d say it depends on the situation and what type of fatigue (within a ride, day to day, month to month). Since fatigue in the end is driven by the supercomputer which is exposed to other things/exposures besides training, the list gets pretty long with lots of unknowns, IMHO.

If it gets narrowed down to just fatigue within a ride, and maybe we look at somebody who does 5 minute VO2’s all the time, year-round, so they are conditioned to them...plus they are extremely motivated to crack their buddy who is an equal (this is a big assumption here), they are fully fueled, fully rested, hydrated, sleeping well, no viruses, not over-heated or cold, no spikes of stress in their lives, no health issues which would show up during intense exercise...etc...etc...IOW, isolation of a workout from the rest of the world...but preserve motivation....

In that setting I think if one wanted to be really basic and ignore all but two factors, those two factors would be, in general terms, depletion of “fuel supplies” and the accumulation of “bad humors” which are both perceived by the brain as “pain or discomfort”. So if this is applied to those VO2’s, there will be two things going on depending on the time course. Towards the end of individual VO2 intervals it starts to hurt. A significant component of that pain is driven by the accumulation of “bad humors”...lower pH, etc., which may directly and indirectly cause pain. As individual VO2 intervals are accumulated, successive VO2’s begin to gradually get more difficult. A significant component of this issue is driven by the depletion of “fuel supplies”...ie glycogen. Eventually, the fuel supplies are depleted enough such that the intervals can no longer be completed.

AndyF posted this on Wattage:
1) 20 minutes @ FTP + 20 minutes @ FTP (no break) = 54.4 training load gizmo
2) 20 minutes @ FTP + 40 minutes @ FTP/3 + 20 minutes @ FTP = 94.6 training load gizmo
3) 40 minutes @ FTP/3 = 7.3 training load gizmo

I don't have the software and I'm not sure how this is computed, but I'll assume he got the numbers right. It seems that intervals with rest periods gets scored at a higher level than if the rest periods were not there! Obviously FTP/3 is such a low level of exercise that it is not causing significant stress... certainly not over a 20min time period.


It seems like it...but honestly, I don’t know what the heck those numbers mean let alone the value of the differences. I'm not so sure those aren't really just kj, and when there are differences between the points and kj's, it's driven by soft-pedaling.

If someone wished to compute training stress that is applicable to all situations, seems like a good place to start would be a person's max power-duration #s. Assign a running "depletion" score for each "fuel" depending on the intensity and duration of the effort, and a "replenish" score for the fuels that are being replenished while going easy. I guess there are waste products to consider as well... but it shouldn't be too tough to take a good stab at this and compare to actual race data.


That would be an interesting, intellectual project. That aside, I still don’t see what practical benefit/solution computing a training stress number from power data would provide in the big picture.

Another thing just occured to me. One reason that short high-power spikes might give you a higher AP is because you are using a fuel that has been relatively "untapped" previously in your ride. In a max effort you wouldn't want to leave any fuel unused. This could possibly have implications for TT riding... maybe it would be faster to include harder efforts of 10-15 sec throughout the race... especially on hills.


Yeah, untapped, but also because the work is spread out over a larger muscle mass intermittently. Practically speaking though, I don’t know how purposely doing that would influence performance.

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24169

  • DMC
  • OFFLINE
  • Platinum Boarder
  • Posts: 373
  • Karma: 0
[quote="Ron Ruff":36r1mh4w][

If someone wished to compute training stress that is applicable to all situations, seems like a good place to start would be a person's max power-duration #s. Assign a running "depletion" score for each "fuel" depending on the intensity and duration of the effort, and a "replenish" score for the fuels that are being replenished while going easy. I guess there are waste products to consider as well... but it shouldn't be too tough to take a good stab at this and compare to actual race data.

Another thing just occured to me. One reason that short high-power spikes might give you a higher AP is because you are using a fuel that has been relatively "untapped" previously in your ride. In a max effort you wouldn't want to leave any fuel unused. This could possibly have implications for TT riding... maybe it would be faster to include harder efforts of 10-15 sec throughout the race... especially on hills.[/quote]

Like Sebastian Weber and STAPS?

"Q-Bert Grabsch rode the [world] Championships with an SRM on his time trial bike. You told us about a spezial pacing strategy, therefore Bert doesn want to publish his data - which is understandable. Can you tell us in general what is important in a time trial and how an SRM can help?

A-We use the so called STAPS method. It makes it possible to predict the physiological respond in the muscle due to the power output (measured with the SRM). That means if we know the power output and the athlete we can exactly predict at which power the rider can recover and how fast, how much lactate is accumulated in a minute or for example how long he can „overpower". We transfer that on a course, for example the course in Varese and can therefore predict which uphill, corner, downhill etc. has to be ridden in which how fast to gain the optimum speed."

http://srm.customer.sasg.de/index.php?o ... 61&lang=fr

If any of you manae to uncover any other links or info in Staps (english or German) please post!

DMC

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24171

  • rmur
  • OFFLINE
  • Platinum Boarder
  • Posts: 1088
  • Karma: 0
A-We use the so called STAPS method. It makes it possible to predict the physiological respond in the muscle due to the power output (measured with the SRM). That means if we know the power output and the athlete we can exactly predict at which power the rider can recover and how fast, how much lactate is accumulated in a minute or for example how long he can „overpower". We transfer that on a course, for example the course in Varese and can therefore predict which uphill, corner, downhill etc. has to be ridden in which how fast to gain the optimum speed."


hmmm ... sounds very, very much like Coggan's power algorithm .. which many of us have been using for a couple of years to suggest pacing on known variable courses. OR to retrospectively take a ride file and see just how well or poorly we pacing it (Alex Simmons POI). I've found it very useful.

But how can that possibly BE? The metric is so flawed/skewed/biased that it couldn't possibly be useful, could it?
GIZMO marketing specialist

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24172

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
Kirk wrote:
rmur wrote:
I can get it - but I know the 'steps' were long - much longer than a typical ramp test ... i.e. pseudo steady-state?

what's the applicability? Oh man, you know I can't repeat all that ex. phys. jazz ...snipped...


Please do get that protocol and data.

As for the applicability issue, I asked about that both here and at wattage years ago (more than once IIRC), and there was no response. What is your general perception of the applicability issue?

Any chance you know the protocol, have access to the data, or have created a general perception about the applicability issue?


<time passes>

kirk wrote:

What's up Rick? I suspect that you can easily formulate your perception about the applicability even if you are struggling to come up with the protocol and data.


I agree, surely Rick Murphy has formed some sort of perception about its applicability.

As for the data and protocol...Rick Murphy, surely you and/or the folks you are heavily invested with aren't censoring information regarding this request, are you?
-kraig

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24173

  • rmur
  • OFFLINE
  • Platinum Boarder
  • Posts: 1088
  • Karma: 0
Kraig,
why do you use first and last names when you hate someone? Notice I stick to first names only ...

honestly???
GIZMO marketing specialist

Re: Rick Murphy's Normanalized Power Thread 9 years, 1 month ago #24175

  • kraig
  • OFFLINE
  • Administrator
  • Posts: 3285
  • Karma: 4
rmur wrote:
Kraig,
why do you use first and last names when you hate someone? Notice I stick to first names only ...

honestly???




I don't hate you Rick Murphy - just attempting to make sure you are on your best behavior.

how are things coming with Kirk's request?
-kraig

Time to create page: 0.77 seconds

Poll

Which type of tire is more aerodynamic?