Wait, so now you're saying the protocol is bad so the eyeball test wins? I thought your position was that the protocol was good.

BTW, I agree that Dave's peak efficiency isn't at 60cm/s or slower. It doesn't make sense. But you are the one gathering the data to demonstrate it, so follow the data...

---

Swimming Workout of the Day:

Favourite Swim Sets:

2020 National Masters Champion - M50-54 - 50m Butterfly

Month end is starting, so I don't have time to continue our friendly banter anymore.

Keep us posted!!

;-)

---

Swimming Workout of the Day:

Favourite Swim Sets:

2020 National Masters Champion - M50-54 - 50m Butterfly

Really? To my way of thinking the shape of the curve wil be similar whether a good day or bad. Bad days it will be somewhat lower. Very good days somewhat higher. But, the peak efficiency will still be at the same pedal speed and since that is what I am going for all I care about is the shape of the curve, not the numbers.

Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

Jason wrote: "Wait, so now you're saying the protocol is bad so the eyeball test wins? I thought your position was that the protocol was good.

BTW, I agree that Dave's peak efficiency isn't at 60cm/s or slower. It doesn't make sense. But you are the one gathering the data to demonstrate it, so follow the data…”

Ugh, I think you will find the “protocol” quite adequate for the purpose I intended, to find Dave's most efficient pedal speed. It was not a protocol designed as a research study.

It seemed to me that you were the one trying to fit the data to a linear correlation coefficient which is pretty much inappropriate because we would not expect the data to be linear based upon previous work in this area even though the coefficient was pretty good. What I pointed out was a polynomial coefficient was even better even though there wasn’t enough data to give a curve like we might anticipate. That would hurt a research study but in this case allows us to come up with a pretty good answer before Dave turns 70.

One of the problems with the real world is that rarely is the data that we have before we have to make a decision particularly good so one has to do the best they can with the limited data they do have. At least, in this case, there is some data to help Dave make this decision. Even when the data is seemingly very good the wrong decision can be made, see Challenger disaster and many others.

Everyone reading this thread can see the data we have gathered. Based upon that post what you believe Dave’s most efficient pedal speed is at the power tested. If it isn’t near 110 cm/sec please explain your reasoning.

Frank Day (

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

JasoninHalifax has expressed concerns regarding the reliability of the data prefering the mathematical solution over the eyeball test. Here is the problem of relying on mathematical solutions when trying to solve a specific problem, the math may not be relevant. The problem here, it would seem, is the daily variation in performance which is natural causing a spread that decreases the reliability of any particular solution. But, here is the result of one set of data obtained on one day. Not only does it have an almost perfect R^2 value (0.9882) the curve is what one would expect, an inverse U.






But add a few more points with a spread suddenly the R^2 makes the data look simply awful (0.26) because of the daily variation making the “best” solution not very good from a statistical sense even though each data set looks similar to the one that had a very good correlation. The only solution to such a problem is more data which is why when we include all of the data we have collected the R^ increases to 0.7.





The problem here is I was simply looking for the pedal speed that maximized Dave’s efficiency. This can easily be seen simply looking at the data but the math suggests the best solution line is unreliable because the “best solution” isn’t asking the right question. I really don’t care if the “best solution” accurately reflects Dave’s abilities because it simply cannot do so because we know there is a daily variation. The eyeball test gives us the answer we need. One only needs those good correlation coefficients if one is drawing conclusions and doing an academic study because that is what the journals demand.

This is an argument made by those trying to discredit this work by trying to confuse the issue and trying to look like some academic standard hasn’t been met. It is a false argument.

Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

Why aren’t you using actual rpm instead of prescribed rpm? Seems like another source of error...

The McDaniel paper on metabolic cost in JAP is freely available. But that's no reason to stop your thorough and carefully controlled "testing".
http://jap.physiology.org/content/93/3/823.long

Just to save you the trouble of reading the paper (or even looking at the figures) the results go like this:
1. Power output accounts for 95% of the variation in metabolic cost in the protocol including between and within subject variation.
2. Pedal speed (pedaling rate x crank length) accounted for over half of the remaining the 5%. A model including power and pedal speed accounted for 98% of the variability within and between subjects. When that two component model was applied to each individual, it accounted for 99% of the variability within each subject.
3. Neither pedaling rate nor crank length by themselves improved the model any further.
So, power matters most (no surprise) and pedal speed explains most of the rest of the variation. Crank length per se is irrelevant within the range tested of 145-195mm.
Can we be done with this thread now?

There is definitely some tilting here, and possibly even some windmills

Bio McGeek wrote "The McDaniel paper on metabolic cost in JAP is freely available. But that's no reason to stop your thorough and carefully controlled "testing".
http://jap.physiology.org/content/93/3/823.long

Just to save you the trouble of reading the paper (or even looking at the figures) the results go like this:
1. Power output accounts for 95% of the variation in metabolic cost in the protocol including between and within subject variation.
2. Pedal speed (pedaling rate x crank length) accounted for over half of the remaining the 5%. A model including power and pedal speed accounted for 98% of the variability within and between subjects. When that two component model was applied to each individual, it accounted for 99% of the variability within each subject.
3. Neither pedaling rate nor crank length by themselves improved the model any further.
So, power matters most (no surprise) and pedal speed explains most of the rest of the variation. Crank length per se is irrelevant within the range tested of 145-195mm.
Can we be done with this thread now?”

So, would you say there is something to be gained or not in the athlete optimizing pedal speed? 5% sounds fairly substantial to me in the serious athlete. Plus, that was on average so I suspect some have more to gain than others. Dave’s results from this effort seem substantial wouldn’t you say? If you agree then maybe this thread will die. If not then perhaps you can explain why not.

Also, I would take issue with this statement: "Recently, Martin et al. (33, 35) used an experimental paradigm in which both pedaling rate and cycle crank length were varied. That experimental paradigm produced several pedal speeds (one for each crank length) for any specific pedaling rate. They reported that maximal muscular power did not differ when cycling with crank lengths of 145, 170, and 195 mm, suggesting that muscular function was unaffected within that range of cycle crank lengths.” The problem is, as you know, that all your paper showed was there was no statistical difference yet it clearly suggested that there might be a difference. I would agree that if there is a difference that it doesn’t have to do with muscle function but rather I would submit that, in that range, it has to do mostly with pedaling technique efficiency, the ability to easily get over the top. Your study was simply not powerful enough to uncover the difference suggested. Further, I think this being a triathlon site, and in support of my idea, most understand that crank length certainly does affect their ability to make the cranks go around when in the aero position.


Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

I thought this thread was about crank length.
The range we tested was from 40rpm to 100rpm and 145mm to 195mm cranks. That gave us a range of pedal speeds of 0.6 to 2.0 m/s. Across that >3x range, pedal speed accounted for a little over 2.5% (not 5%) of the variation. Naturally a smaller (more reasonable range), would have a smaller effect, maybe 1% variation due to pedal speed.
EDIT: The relationship you're looking for is actually shown in Figure 2a.
Regarding what you generously call Dave's results, who knows? Dave wants to feel special with highly individual response and you help him feel special. In return he helps you get on to ST. Mutually beneficial Bromance.


Correct. I won't go down your science-denying rabbit hole where stats don't matter.

For the love of God, you cannot say this if you have any understanding of statistics at all. You can't say what a study MIGHT have said in a different scenario. You can say what the study DOES say and nothing more.

I know, I know. Eyeball test or some other convenient bullshit......

Which is extra ironic, given how much he enjoys calling everyone else "snowflakes". Well, less "ironic" and more "projection".

---

The point is, ladies and gentleman, that speed, for lack of a better word, is good. Speed is right, Speed works. Speed clarifies, cuts through, and captures the essence of the evolutionary spirit.

And... thread.

---

Tech writer/support on this here site. FIST school instructor and certified bike fitter. Formerly at Diamondback Bikes, LeMond Fitness, FSA, TiCycles, etc.
Coaching and bike fit - http://source-e.net/ Cyclocross blog - https://crosssports.net/ BJJ instruction - https://ballardbjj.com/

My improvement from going to shorter cranks was definitely NOT due to a "naturally reduced pedal speed". My natural cadence went up a few rpm, keeping pedal speed the same. When I played around with significantly shorter cranks, my cadence went up significantly, once again keeping pedal speed the same. Where my improvement came from was from being able to comfortably come over the top of pedal stroke while in a lower position. Power and HR remained the same, cadence was slightly higher, and average speed was higher due to a more aero position. Many of the people I know who have gone to shorter cranks report similar experiences.

From Cervelo:

---

"I'm thinking of a number between 1 and 10, and I don't know why!"

Being able to get over the top of the pedal stroke is what I am seeing using shorter cranks with the powercranks. I can get in aero now with them

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

JFC. 600 posts for you to realise that 150mm cranks rather than 250mm cranks might allow you to stay aero............

That's amazing. What an insight. You should publish..........

Dan should get "shorter cranks let you stay aero" up on the front page.........

Bio McGeek wrote: "I thought this thread was about crank length.
The range we tested was from 40rpm to 100rpm and 145mm to 195mm cranks. That gave us a range of pedal speeds of 0.6 to 2.0 m/s. Across that >3x range, pedal speed accounted for a little over 2.5% (not 5%) of the variation. Naturally a smaller (more reasonable range), would have a smaller effect, maybe 1% variation due to pedal speed.
EDIT: The relationship you're looking for is actually shown in Figure 2a.
Regarding what you generously call Dave's results, who knows? Dave wants to feel special with highly individual response and you help him feel special. In return he helps you get on to ST. Mutually beneficial Bromance.”

This thread is about the protocol I have been doing with Dave to, hopefully, find the optimal crank length for him for his racing (aero position). Dave has focused on crank length but if you will read my posts you will see my protocol is all about pedal speed, based upon the work you, in part, did. The crank length comes into play when we see what happens to his efficiency as he goes into the aero position with different crank lengths. That is the crank length part of the protocol and we have just started that.

Since you were involved in the paper that got me thinking about how pedal speed could and should be optimized it seems reasonable that you confirm that importance in this thread that really is about your paper. Crank length is a red herring and, I suspect, you know it. Anyhow, I have no idea how much improvement each person might see but in racing it seems that any improvement could be important so if one is serious why would one ignore it. Your own paper did find a 0.5% difference between 145 and 170 which you thought because the difference didn’t reach statistical significance and was small so could be ignored. I thought it strange that competitors would ignore any potential improvement but they have taken your recommendation as proof crank length doesn’t matter. However, as I have thought about it I suspect most of that power difference you measured was probably due to a slower pedal speed since we have found almost no crank length effect on his efficiency during our testing while in the upright position.

Anyhow, as a result of this effort (and it has been an effort, ask Dave) I have slowed Dave’s pedal speed substantially, he has now adapted to this change, and he is noticing substantial power increases in a test run he does regularly, about 10% over what he was doing last year. He is simply reporting his what he does here and soliciting comments as he typically does. The usual suspects have chimed in with the usual BS whereas is someone gets a power meter and claims a 20 watt increase everyone thinks it is something. Few here have read or understood what is going on. When I was made aware of this thread and it became clear that no one understood what was going on I offered to explain and answer questions. At least most of the posts now are about the protocol and someone who actually might understand what I am trying to do is actually posting something more than snide “waiting for the science” comments. This whole thing is about trying to use the science to improve an elite and it seems to have been effective. Thank you for doing that work even though the importance of it seemed to be lost on you and most everyone else.

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

Derekl wrote: "For the love of God, you cannot say this if you have any understanding of statistics at all. You can't say what a study MIGHT have said in a different scenario. You can say what the study DOES say and nothing more.

I know, I know. Eyeball test or some other convenient bullshit……”

Phooey! If the statistics say the P is <0.06 the editors usually prohibit THE AUTHOR from claiming anything other than what you say. However, that doesn’t prevent the author (or anyone else, like me) from making an educated guess that a more powerful design is likely (not guaranteed) to reach the “significant” threshold. What makes a good scientist is one who is both curious and observant, not one who is accepting of everything they are told.

Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

Warbird wrote: "My improvement from going to shorter cranks was definitely NOT due to a "naturally reduced pedal speed". My natural cadence went up a few rpm, keeping pedal speed the same. When I played around with significantly shorter cranks, my cadence went up significantly, once again keeping pedal speed the same. Where my improvement came from was from being able to comfortably come over the top of pedal stroke while in a lower position. Power and HR remained the same, cadence was slightly higher, and average speed was higher due to a more aero position. Many of the people I know who have gone to shorter cranks report similar experiences.”

In my experience (and I have quite a few people reporting their experience with shorter cranks to me and I am 6.2” and currently ride 135 mm cranks at a cadence of 70-80 or so when cruising) is that a few see no change and stay where they are but most like shorter cranks (sometimes much shorter) and most do not increase their cadence to compensate so their pedal speed slows. It was the experience of hearing these reports that got me to thinking about trying to develop a protocol to help people find what truly is optimal for them. Anyhow, it doesn’t matter to me if you did that as what you describe is why I think crank length also matters when one is riding in the aero position (which is what we are trying to test now). That having been said, if you haven’t done pedal speed testing then you can’t convince me that your pedal speed now is optimal.

Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

LOL. Now we're on to educated guesses based on our obvious biases.

Have you ever injured yourself in your career as a mental gymnast?

Now everyone sees why you are banned from every forum. You are so full of shit, it oozes out of this thread. The only comment you’ve made that I agree with is the fact that no one posting in this thread will have any impact on Dave’s opinion or direction he’s taking. He is so dense and determined to do the opposite of what anyone reasonable suggests that he will continue to do anything possible to be different and awkward. With 600 posts about him in this thread, he is meeting his needs as a massive narcissist and self centered attention whore.

Of course he injured himself. It's why he makes cranks and does not practice medicine

Seriously, dude. Your "protocol" has so many flaws that I can't imagine how you could consider any of these changes meaningful.

Andrewmc wrote: “JFC. 600 posts for you to realise that 150mm cranks rather than 250mm cranks might allow you to stay aero............

That's amazing. What an insight. You should publish..........

Dan should get "shorter cranks let you stay aero" up on the front page......... “

JFC. You do realize that Dave lives in the mountains and had fallen for the myth that longer cranks help you climb better because they have better leverage. He isn’t the only one to believe that. To get Dave to go shorter to help him when aero I first had to convince him it wouldn’t hurt his climbing. Dave would still be worried about this if it weren’t for his Martis testing where he not only maintained his climbing ability but improved it because of some other changes (slower cadence) made.

Not sure Dan will want to put your suggestion on the front page as part of the reason I was banned was how I defended my position on short cranks. Now lots of people think they are great but no one knows how to find the best length which is what I are trying to now.

Frank Day

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep

Lets see, if this is SO obvious, then why are SO many rider using long cranks?
If this is SO obvious, why do most bike fitters ignore fitter the athlete with shorter cranks to get a better aero position?

And why do so many read, let alone post, on a thread that they already know the correct answer? :)

---

Dave Campbell | Facebook | @DaveECampbell | h2ofun@h2ofun.net

Boom Nutrition code 19F4Y3 $5 off 24 pack box | Bionic Runner | PowerCranks | Velotron | Spruzzamist

Lions don't lose sleep worrying about the sheep