Study of Power meters.They are really same as HR monitors, but lot more expensive.

[acoggan]

I think most believe that how riders rank against each other in races is the best measure of performance. Who cares what their power is/was.

Those who wish to understand the underpinnings to that performance, so they can best go about improving it.

 

[acoggan]

Haven't they read your book?

While that is a good starting place for learning the basics (and a bit more), not all of my ideas/insights can be found in the book.

 

[FrankDay]

Those who wish to understand the underpinnings to that performance, so they can best go about improving it.

But, power is just one underpinning of the performance, it is not, as Fergie and you seem to imply, the performance itself. Focusing on this one aspect is just as likely to make people forget the other important underpinnings of race performance, IMO, as it is to enhance this one aspect.

Anyhow, again, if a PM is a direct measure of performance I simply don't understand what is so difficult in the interpretation of this data that USA cycling needs to consult you to help them interpret this data.

 

[acoggan]

But, power is just one underpinning of the performance, it is not, as Fergie and you seem to imply, the performance itself. Focusing on this one aspect is just as likely to make people forget the other important underpinnings of race performance, IMO, as it is to enhance this one aspect.

As I said, power is the ultimate measure of a cyclist's performance as the "engine" driving the bicycle forward. There are, of course, other determinants of who crosses the finish line first/soonest, such as aerodynamic drag or psychology (confidence). Fortunately, a powermeter can also be used to improve those aspects as well (as I have been pointing out for roughly a decade).

if a PM is a direct measure of performance I simply don't understand what is so difficult in the interpretation of this data that USA cycling needs to consult you to help them interpret this data.

Let me put it this way: some people look at a Rorschach diagram and only see a black blob, or perhaps a young girl or old woman. Other people, though, can look at the same diagram and see not only a black blob, young girl, AND old woman, but a whole lot more as well.

 

[acoggan]

Those who wish to understand the underpinnings to that performance, so they can best go about improving it.

Speaking of which:

http://www.biketrainingtips.com/2011/10/analysis-of-team-time-trial.html

Now just think what one could do if everyone on the TTT squad were using a powermeter, and you knew their capabilities...

 

[FrankDay]

As I said, power is the ultimate measure of a cyclist's performance as the "engine" driving the bicycle forward. There are, of course, other determinants of who crosses the finish line first/soonest, such as aerodynamic drag or psychology (confidence). Fortunately, a powermeter can also be used to improve those aspects as well (as I have been pointing out for roughly a decade).

Really. And, it does this better than other methods of improving confidence? And, you have evidence to support this contention?

Let me put it this way: some people look at a Rorschach diagram and only see a black blob, or perhaps a young girl or old woman. Other people, though, can look at the same diagram and see not only a black blob, young girl, AND old woman, but a whole lot more as well.

Look, here is the issue. Is there any scientific evidence that looking at this Rorschach diagram of power data and interpreting it in any of various ways makes a positive difference in outcome compared to using other methods? But, of course, before that can be shown it will need to be defined as to how one is to interpret this mass of data, before the concept can be tested. You are now saying, it seems, your book isn't enough. Well, what is? What does it take to make this device work for the athlete better than the alternative methods available. Certainly these two studies seem to suggest that the PM and HRM are equivalent effort feedback devices, so there must be more the PM can do. What is it? How can it be tested?

 

[acoggan]

Really. And, it does this better than other methods of improving confidence?

I never said that it did. You, however, claimed that using a powermeter is likely to distract you from determinants of performance (as defined as placing/time) other than your actual power. Aside from the fact that power is the generally the most important determinant of performance (again, defined as placing/time),* I don't think your claim is true. Indeed, there is nothing like showing somebody how much power they are "wasting" to convince them of the importance of focusing on improving their aerodynamics, or showing them that they are producing as much (or more) power as those at the highest level of the sport to convince them that they can reach that pinnacle as well.

Is there any scientific evidence that looking at this Rorschach diagram of power data and interpreting it in any of various ways makes a positive difference in outcome compared to using other methods?

Again, I have never claimed that there is, and neither has anyone else who advocates powermeter use. You, OTOH, have repeatedly tried to make the case that science supports the efficacy of your product (even though it doesn't), which is why everyone (rightfully, IMO) holds you to a different standard.

 

[acoggan]

Aside from the fact that power is the generally the most important determinant of performance (again, defined as placing/time),*

Sorry, forgot to insert my footnote...

*Evidence that power is the most important determinant of performance can be found in the fact that it distinguishes competitors of different levels. OTOH, you are as likely (or perhaps even more likely) to find people with exceptionally low drag among rank amateurs as among pro cyclists, there are certainly people racing (or even not racing) at the amateur level who can handle their bike as well as any professional, elite athletes on average are NOT different from sub-elite athletes in terms of their psychological "profile", etc., etc., etc.

 

[FrankDay]

I never said that it did. You, however, claimed that using a powermeter is likely to distract you from determinants of performance (as defined as placing/time) other than your actual power. Aside from the fact that power is the generally the most important determinant of performance (again, defined as placing/time),* I don't think your claim is true. Indeed, there is nothing like showing somebody how much power they are "wasting" to convince them of the importance of focusing on improving their aerodynamics, or showing them that they are producing as much (or more) power as those at the highest level of the sport to convince them that they can reach that pinnacle as well.

I have only said one can make the argument that focusing on one thing can distract you from other things. The one thing that can be said for certain is that there is no science to support the contention that using a PM in training or racing will make any difference to the athlete.

Again, I have never claimed that there is, and neither has anyone else who advocates powermeter use.

So, you are admitting that there is no science to support the use of a PM in training or racing. Is that correct?

You, OTOH, have repeatedly tried to make the case that science supports the efficacy of your product (even though it doesn't), which is why everyone (rightfully, IMO) holds you to a different standard.

Really? I have never said that. I tell people what our data suggests and what we expect many to see, but that isn't science. There is some science showing changes occurring that supportsome of my arguments as to why I expect these changes to occur but that is all. Science is not about "rightfully holding people to a different standard" is it? Well, maybe in the eyes of some of you here.

And, of course, I comment on the science that has been done on the product. Most of the studies have pretty much ignored our recommendations as to how to use the product. But, at least there are some studies that support partially what I say. The only study that even came close to following our instructions for best outcome (immersion training - Dixon) came up with substantial and statistically significant increases in both VO2max and power, even though it only lasted 6 weeks and occurred at a time of the year the participants expected their power to drop. Our claims haven't been proven but our claims are for a 6-9 month immersion program and for our typical user. Does immersion training work. Seems to:

From SlowTwitch interview
ST: What has happened since 2009 that you improve that much? Going 40 minutes faster is not that impressive when the previous race was in the 13-hour range, but going from 9:29 to 8:50 is quite another story.

Sam: Well, I changed my coach in April of 2009 and that really made a huge difference. We started with a long term plan and this seems to work out pretty well. The focus was merely on the run because I already was a solid cyclist. What really gave me a boost this year is the use of powercranks. Since January I am doing all my bike training with powercranks and that made me a lot stronger in the run and I also gained some watts extra power on the bike. I love to train hard so there is a lot of quality in my training. All my training is visible publicly via Trainingpeaks and Twitter so anybody can check what I am doing and watch my workout files.

So, what I say about my product does have, at least, a little scientific support for improving outcome. That is way more than can be said for the PM as regards improving athletic outcome. All this thread is doing is holding the PM to the same standard you hold me. I think the PC's win in this "scientific support" game, at least as thing stand now.

 

[FrankDay]

Sorry, forgot to insert my footnote...

*Evidence that power is the most important determinant of performance can be found in the fact that it distinguishes competitors of different levels. OTOH, you are as likely (or perhaps even more likely) to find people with exceptionally low drag among rank amateurs as among pro cyclists, there are certainly people racing (or even not racing) at the amateur level who can handle their bike as well as any professional, elite athletes on average are NOT different from sub-elite athletes in terms of their psychological "profile", etc., etc., etc.

Is that really the case? Look at the riders in the Coyle study looking at pedaling technique. The fastest rider, by far did not have the highest power. And, I think, the rider with the second highest power was only the 11th fastest overall (and in the "slow" group). In that study what set the two groups apart was mostly years of aerobic training, as I remember. And, as I remember, the fast group was about 10% faster and also about 10% more powerful. This could have only happened if the faster group was a lot more aerodynamic.

 

[FrankDay]

Is that really the case? Look at the riders in the Coyle study looking at pedaling technique. The fastest rider, by far did not have the highest power. And, I think, the rider with the second highest power was only the 11th fastest overall (and in the "slow" group). In that study what set the two groups apart was mostly years of aerobic training, as I remember. And, as I remember, the fast group was about 10% faster and also about 10% more powerful. This could have only happened if the faster group was a lot more aerodynamic.

Let me amend this statement. I was wrong as I was thinking about the pedal forces, not the power produced in the Coyle, Feltner, et. al. paper.

Here were the statistically significant differences between the very fast group and the fast group.

Years endurance training 8.8 vs 5, p<.01
40 km TT 53.9 vs 60 min, p<.01
%VO2max at LT 79.2 vs 75.3, p<.05
Vo2 at 1 hr effort 4.54 vs 4.18 L/min, p<.05
Avg pwr for 1 hr 346 vs 311, p<0.5
% type 1 muscle fiber 66.5 vs 52.9, p<.05
capillaries per mm 464 vs 377, p<.05

While power was higher in the faster group, Let's talk what the "underpinning of the performance" difference probably is here. I think one can conclude that the difference between the groups probably has nothing to do with whether they have or had power meters even though one was more powerful. The major difference between these groups, apparently allowing one to be more powerful that the other, was an extra 3 years of endurance training making one group better trained aerobically than the other, allowing it to deliver more oxygen to the muscles (look at the difference in capillarization) which allows them to do more work.

Comments?

 

[Alex Simmons/RST]

Comments?

Yes. You are boring us all to death Frank with all your twaddle.

Like your latest logical fallacy (known as the "Excluded Middle"), which is the inference that a (evidenced based) coach for instance would only ever look at power and ignore everything else that pertains to racing (or other) outcomes or performance.

 

[CoachFergie]

Yes. You are boring us all to death Frank with all your twaddle.

Like your latest logical fallacy (known as the "Excluded Middle"), which is the inference that a (evidenced based) coach for instance would only ever look at power and ignore everything else that pertains to racing (or other) outcomes or performance.

To be honest Alex I can't truly celebrate on of my guys winning all the races tonight in a Trans Tasman Challenge because he doesn't race track with a power meter

We must compare lists of logical fallacies some time.

 

[sciguy]

There is another scientific evidence of how HR monitors are just same with PM. This time from Florida, it must be world wide conspiracy against them

P.S. Are so called "scientific" guru coaches here (Alex not you) gonna accept this findings? Of course not, they just pick up things/scraps from science which gonna give them new clients and bank account get fat. I am just saying. Those people really do not care about science, do they?

Something looks a bit strange here to me. Perhaps it's due to lack of sleep last night

If you take a look at the study, one thing that I don't see mentioned or analyzed was the difference in average power for the pretreatment 20km TT and the post treatment 20km TT. I'm not talking about VO2, LT but what the athletes actually put out during their 20kmTTs.

http://www.jssm.org/vol10/n3/12/T2.htm

The average increase in power for the "power trained group" was 20 watts while for the "heart rate trained group" it was only 10 watts. Perhaps the difference wasn't statistically significant but I'm really surprised it isn't mentioned. Also, I've yet to figure out why the authors claim a "near 3-and-a-half minute improvement (7.8%) in 20km TT completion time." when the data shows less than 2 minutes for either group. (2501.4-2382.6 = 118.8s for the power trained group and 2348.8-2233.4 = 115.4s for the heart rate trained group.

Perhaps a sharper tool can clear this up for me.

 

[FrankDay]

Something looks a bit strange here to me. Perhaps it's due to lack of sleep last night

If you take a look at the study, one thing that I don't see mentioned or analyzed was the difference in average power for the pretreatment 20km TT and the post treatment 20km TT. I'm not talking about VO2, LT but what the athletes actually put out during their 20kmTTs.

http://www.jssm.org/vol10/n3/12/T2.htm

The average increase in power for the "power trained group" was 20 watts while for the "heart rate trained group" it was only 10 watts. Perhaps the difference wasn't statistically significant but I'm really surprised it isn't mentioned. Also, I've yet to figure out why the authors claim a "near 3-and-a-half minute improvement (7.8%) in 20km TT completion time." when the data shows less than 2 minutes for either group. (2501.4-2382.6 = 118.8s for the power trained group and 2348.8-2233.4 = 115.4s for the heart rate trained group.

Perhaps a sharper tool can clear this up for me.

Table 1 is somewhat different than table 2 (how to explain this?) but there are also a few conundrums in this data. http://www.jssm.org/vol10/n3/12/T1.htm By almost every metric that people tout as being important to going fast (power, power to weight, etc) the power group should have been faster than the HR group in the TT test. They were not. Aerodynamics cannot account for this difference since the testing was done on the Computrainer (unless the CT was programmed incorrectly to account for wind drag).

Or, I am missing something here also.

Anyhow, I think the answer to your question is it wasn't mentioned simply because it didn't reach statistical significance. But, in view of the inconsistencies, an email to the authors might be useful to clarify things (I don't see a way of writing a letter to the editor).

 

[oldborn]

Do not know sciguy really. But I am sure that Miami Vice guys are lurking, so maybe they can jump in and answer to those question you arise.

 

[sciguy]

Table 1 is somewhat different than table 2 (how to explain this?) but there are also a few conundrums in this data. http://www.jssm.org/vol10/n3/12/T1.htm By almost every metric that people tout as being important to going fast (power, power to weight, etc) the power group should have been faster than the HR group in the TT test. They were not. Aerodynamics cannot account for this difference since the testing was done on the Computrainer (unless the CT was programmed incorrectly to account for wind drag).

Or, I am missing something here also.

Anyhow, I think the answer to your question is it wasn't mentioned simply because it didn't reach statistical significance. But, in view of the inconsistencies, an email to the authors might be useful to clarify things (I don't see a way of writing a letter to the editor).

If one looks at http://www.jssm.org/vol10/n3/12/T1.htm then the time improvement was 4:40.8 for the power trained group and only 1:55.1 for the heart rate trained group. It would appear that the nearly 3:30 improvement the authors cite is the average improvement for the two groups combined according to this data table. My brain still isn't resolving the inconsistencies between the two tables.

 

[Alex Simmons/RST]

If one looks at http://www.jssm.org/vol10/n3/12/T1.htm then the time improvement was 4:40.8 for the power trained group and only 1:55.1 for the heart rate trained group. It would appear that the nearly 3:30 improvement the authors cite is the average improvement for the two groups combined according to this data table. My brain still isn't resolving the inconsistencies between the two tables.

Interesting.

HR group went 5% faster. Power group went 11% faster.
On average of course.

Clearing up a range of anomalies you are highlighting would be helpful.

It is of course a possibility that the HR group paced more poorly than the power group and so had a poorer second per watt ratio after the intervention.

That might not account for all the difference, but if one group paced themselves on HR (or feel) and did not use power as a guide in the testing (especially early on in the effort), then it wouldn't surprise me to see more of them start too hard and fade, perhaps badly.

The power group may have had a better idea of the power they could sustain after a training intervention involving use of power feedback and perhaps had learned to better dose their effort, or learn the way their PE varies when pacing well (whether or not they realised it).

 

[FrankDay]

Interesting.

HR group went 5% faster. Power group went 11% faster.
On average of course.

Clearing up a range of anomalies you are highlighting would be helpful.

It is of course a possibility that the HR group paced more poorly than the power group and so had a poorer second per watt ratio after the intervention.

That might not account for all the difference, but if one group paced themselves on HR (or feel) and did not use power as a guide in the testing (especially early on in the effort), then it wouldn't surprise me to see more of them start too hard and fade, perhaps badly.

The power group may have had a better idea of the power they could sustain after a training intervention involving use of power feedback and perhaps had learned to better dose their effort, or learn the way their PE varies when pacing well (whether or not they realised it).

There are so many seeming irregularities in the data it is hard to make anything of it. One issue regarding the time-trial had to do with the standard deviation of the two groups. The pre training standard deviation of the HR group was a whopping 562.5 seconds, After, both groups have similar SD's, suggesting there were one or two riders in this group where something is not right on that pre-test.

PRE 20km TT time (s) 2348.8 (314.5) 2501.4 (562.5)

POST 20km TT time (s) 2233.7 (222.5) 2220.6 (239.3)

Anyhow, it would be nice if the authors would come here and help us to better understand their study.

 

[CoachFergie]

There are so many seeming irregularities in the data it is hard to make anything of it.

--

Anyhow, it would be nice if the authors would come here and help us to better understand their study.

Yet 5mins ago it was conclusive evidence that power meters were no better than a heart rate monitor

 

[FrankDay]

Yet 5mins ago it was conclusive evidence that power meters were no better than a heart rate monitor

Huh? "conclusive evidence" Phoooeeey! Science is science and a study should be looked at just a critically whether it supports your bias or not. No single study is ever "conclusive evidence" of anything nor could it ever be. Further, all studies have deficiencies, weaknesses. Finding "discrepancies" does not invalidate a study necessarily. Depends on what it is.

Even if this study turns out to be so full of holes and worthless the fact will remain that there is still not a single study that supports the contention that a PM can be used to enhance training or racing compared to alternative methods.

 

[FrankDay]

So much for the one really good reason (at least that used to be the one really good reason) to own a PM, estimating aero drag and rolling resistance.

http://forum.slowtwitch.com/forum/Slowtwitch_Forums_C1/Triathlon_Forum_F1/Estimating_drag_without_a_power_meter_P3590389/

From R. Chung, the developer of the Chung method of using a PM to determine drag "Anyway, now that so many people have bike computers that can record speed (like the Garmin units) I thought it would be good to point out that it's possible to estimate drag using one even if you don't have a power meter. "

 

[CoachFergie]

And nor does the number on a set of scales actually change the rate of calories burnt when trying to lose weight but it does help me to decide whether to have that second beer or not.

 

[Alex Simmons/RST]

So much for the one really good reason (at least that used to be the one really good reason) to own a PM, estimating aero drag and rolling resistance.

http://forum.slowtwitch.com/forum/Slowtwitch_Forums_C1/Triathlon_Forum_F1/Estimating_drag_without_a_power_meter_P3590389/

From R. Chung, the developer of the Chung method of using a PM to determine drag "Anyway, now that so many people have bike computers that can record speed (like the Garmin units) I thought it would be good to point out that it's possible to estimate drag using one even if you don't have a power meter. "

Sure will be interesting to see. I don't post there any more.

I have run those numbers through VE (a power based method - in this case we know power to be = 0W) and get the VE lines to pretty much overlay each other (very close match on VE).

My method is fairly crude (there are far ore elegant methods which I don't know how to set up), I just set the start point at 5 metres and forced the end points to be at zero metres and solved for CdA & Crr based on the sum of the squares of the end point for each VE line = 0.

That's not quite right of course as the distance traveled in each case isn't exactly the same, so the final height would be different.

Using that I get: CdA 0.347m^2; Crr: 0.00717

which seems to be quite a high Crr.

I would need a little maths help to nail down whether or not you can actually split the pair with sufficient precision.

e.g. I have somewhat different CdA/Crr pair that seem to give good overlay of VE as well.

 

[RChung]

Using that I get: CdA 0.347m^2; Crr: 0.00717

which seems to be quite a high Crr.

I get slightly lower Crr (.0067) and slightly higher CdA (.377), but the Crr *is* high: it's my commuter bike with the heavy tires.

As usual, Frank didn't understand the moral(s) of the story: 1) varying your speed across laps or loops or test runs improves the estimation, and the wider the range of speeds the better the estimation will be; and 2) it's possible to estimate drag with field tests without using a power meter but it's much easier to do it with a power meter. This is sort of like saying that you can lose weight without using a bathroom scale but it's a lot easier to use one; or you can calculate logarithms in your head but it's a lot easier to use a calculator.

[Edit:] On reflection, that last one is not a bad analogy. The study that initiated this thread is sort of like testing whether you get the same answer for an addition or subtraction problem whether you use paper and pencil or a high-powered calculator. Frank is arguing that if you can get the same answer from paper and pencil then the calculator is useless.