Friday, November 27, 2009

16Fishy Marketing : The Slippery Speedplay Mannequin

"I'm at the Speedplay Booth. Their leg mannequin has the biggest ass I've ever seen."
- someone at Interbike this year.

While I appreciate bike technology in all forms, I can't stand the inglorious ways by which people try to sell it to you. Readers may know that I run a section on this site titled "Marketing Mishaps". For some time now, I have been collecting events in our biking world that seem fit for it, you know - haphazardly done marketing through poor science, cheesy ad material and other nuances that just turn intelligent people off. I think I have found some new material today.

I present to you the "Slippery Speedplay Mannequin". It's all slippery (and smoke and mirrors too) if you read on.

Dynamic Test Procedure : Quickly, here's a summary of what Speedplay did to try and back up the superiority of their pedal design, the Zero. They placed the mechanized legs of an impressive mannequin (only legs) on a bike in the San Diego low speed wind tunnel. The legs were attached through cleats to the Zero pedals. The contraption was then made to pedal for a mere 5 minutes at 100 rpm in a normalized wind velocity of 30 mph. Apparently the tech guys did this twice with a 3 hole and a 4 hole mounting sole to capture any differences from mounting. Finally, they compared the results to the same procedure done to two other competitor pedals from Shimano and Look (brand was kept secret).

What To Measure? : Speedplay was interested in measuring the drag co-efficient, Cd, multiplied with frontal Area, A of the mannequin and bike system. This term appears in the equation for retarding drag force in cycling. I have written the equation for you below and what the terms mean :

It appears that even Speedplay is muddled with the terms and their definitions. In several places, they write as if they're measuring the term drag co-efficient Cd when even the computer screen in their report shows "CdA". On top of that, they take this to be a dimensionless quantity. I wonder what kind of drink the author may have had the night before?

Where Measured : San Diego Low Speed Wind Tunnel at the Air & Space Technology Center. Test section dimension of facility is 2.44 x 3.66m. Test section cross-sectional area is 8.50m^2.

Results : Going with the poorly presented results of the test, here's what I think I gleaned from it. Tabulated for you, these numbers are for one test for each item, for a duration of 5 minutes and 100 rpm cadence. Please note, all these numbers are claimed by Speedplay. The percentage differences between each line item were done by me.

Conclusions To Customers : Speedplay goes on to claim that their calculations show such drag reductions result in a time saving of 5.5 seconds/hour and 33 seconds/hour for 3 and 4 hole mounts respectively. This, they say, is equivalent to the savings gained from replacing a standard wheel with a deep section, aero front wheel.

The fishy things going on in this whole agenda are the following :

1. Time savings for whom? : While the contraption is really impressive, the 33 second savings computed by Speedplay, if correct, are for the lower legs of a "robot" riding into 30 mph headwind in a wind-tunnel specifically with 4 hole mounted cleats. C'mon, even an ape won't believe that an actual human being using this will get the same amount of claimed benefits.

2. As opposed to what? : The number of pedals in the test are small (2) and the specific brands chosen have been hidden from us. Which is understandable but again, I suppose any idiot could test the Zero pedals with a competitor's junk hardware from the past years and say that it's better. Were all brands tested the latest in market?

3. 33 seconds savings, really? : Those claimed savings seem artificially huge to me. Just for an illustration, the 2004 Nike "Swift Spin" time trial suit prototype (with tracksuit and leggings), worn by a pro cyclist, showed a time savings of 33 seconds when pedaling into 32 mph wind, over its 2003 development model. This suit is still considered by many to be state-of-the-art stuff and a production model was used in a couple of Tour de France's. To claim that switching to a small pair of pedals under a robot's legs equal the time savings of a top aerodynamic piece of apparel worn by a actual human in aero position points to something extremely fishy in the calculations. It even diminishes the many years of work Nike and other researchers conducted in trying to create a more aero suit for pro cyclists.

4. Where's the math? : Obviously, what Speedplay fails to show is the methodology behind their conclusions. How did they arrive at 33 seconds/hour time savings? Moreover, they don't investigate the variation of CdA numbers as the sample size increases by limiting themselves to one test run for each type of pedal. The worst part is, they test each pedal for 5 minutes. Hence, we don't have an indication of the level of error and range of CdA values over many trials. Now you have to ask, is this really 33 second time savings? Or something smaller?

The bottomline?

As is apparent to others, Speedplay wants you to buy their Zero Pedals which will free you of between \$130-\$330 (Chromoly-Titanium). They calculate that if you cut your legs apart, dispose of the trunk and then pedal a bike in a rare and bloody time trial of 5 minutes keeping everything tight and in-plane like a mannequin, you'll save as much time riding into a 30 mph headwind as does tossing out a 32 spoke wheel for a slick aero wheel in the front.

RELATED DISCUSSION FROM OTHERS :

Cyclocosm : Are You A Speedseeking, Torso-less Pair Of Legs?
Alien Biker : The Missing Link, The One Piece Of Gear That Will Help You Ride Like A Pro
Slow-Twitch Forums : Speedplay Claims Huge Wind Drag Reduction Over Leading Brands

* * *

1. I take it that Speedplay had this up on their website for a while. And you wonder why BSNYC didn't catch up on it.

2. The mannequin developed is very impressive. I'm a Speedplay user for the past 4 years. While I did think I was using it for speed, I attributed it to lower weight on climbs.

3. Anonymous7:52 AM

Seems like they don't even know what they're trying to prove? And is this system meant for a human or a mannequin? Muddled is not even the right word.

4. They should have just modelled this on a computer and done a fluid flow analysis. This is stupid. And wind tunnel testing is not cheap you know.

5. "And wind tunnel testing is not cheap you know."

@ bill : ...the dollar amt probably just get passed on to the customer in the end :( thank god i didn't fall for this. (and i love my keo's thank you.)

6. Anonymous9:51 AM

That was a brilliant photo at the end there. The man standing next to the dismembered mannequin as if its the solution for all our ills. I don't quite think I'm ready to cut myself open to gain the so-called "33s time savings."

7. ROB VOSSLER6:48 PM

I second Craig Wallace first comment. The fact that Speedplay has been ignorant about the critiques of these claims show that they're totally in a different world of their own. This is pretty shameful for a company to leave untouched and if anything, I expect them to atleast make a few decent changes to the writeup in the coming days.

8. ROB VOSSLER6:48 PM

I second Craig Wallace first comment. The fact that Speedplay has been ignorant about the critiques of these claims show that they're totally in a different world of their own. This is pretty shameful for a company to leave untouched and if anything, I expect them to atleast make a few decent changes to the writeup in the coming days.

9. Anonymous10:55 AM

So, can someone tell me...
If I pedal slightly toe down (which at 100rpms for me, at least, is guaranteed), effectively shielding the pedal from the wind...is the magical theoretical advantage of Speedplay negated? BTW, Richard Bryne (man shown) is the owner of Speedplay, but more interetingly, a lawyer. ...do what you want with that info...I'm certainly not implying that a lawyer could, or would manipulate data to reflect their (marketing)needs..."snick"

10. Good call on that, Ron.

Has Speedplay been given an opportunity to respond? I'd be curious what they say.

11. Fritz : Not yet, but like some commentators say its been on their site for sometime and other bloggers have picked up on it before me. Certainly they should have heard it by now.

12. Anonymous6:32 PM

Speedplay's owner is a lawyer.

Well, well...

13. Anonymous6:30 AM

I'm surprised that what is normally a quite rational blog like Cozy Beehive would start bashing speedplay pedals. You dont need to test a pedal system to recognize that the frontal area of a speedplay pedal and cleat is smaller than the frontal area of a shimano, look or time pedal. It therefore has less drag. Converting an aero benefit into an approximate time savings is a common way to make the general public grasp the overall advantage of a particular product. Is your criticism that they didnt run the test for long enough to be statistically valid? Is there something fundamentally wrong with their protocol? Or is it just that they condensed the result into a soundbyte?

14. Anon : Welcome to the blog. Look at the comment above you from another Anonymous at 10:55am. His question is pretty much valid too. How will you address that? And if surface area is king, everything on a bicycle should look like a thin stick. It isn't.

And fluid flow is complex. There is interaction with systems, subsystems, body parts etc of a human on a bicycle. You cannot put a half sized mannuquin with just legs, prove it showed some aero drag change with Speedplays compared to other pedals and declare that a rider adopting such equipment will show the exact same results. This has not been proven nor has extensive tests been conducted. In my viewpoint, according to the blog post, both the statistics and the entire logic behind the conclusions are wrong. Neither have the calculations been shown on how someone comes out with a 33 second advantage. Its just lacking in essential and important details.

Feel free to disagree with me. I take it you're not from an engineering or scientific background.

15. Anonymous1:49 AM

Its Anon again. Yes I am from a science background, but not anything related to fluid engineering. Just being honest.

I might not have made it clear in my post that the aero benefit primarily comes from the narrower, smooth spindle of the Stainless and Ti pedals. This is where I believe there's a real difference between Speedplay and other pedal systems which have a much wider axle, especially with a composite body pedal like a Look Keo.

I agree that the pedaling leg is a bit of a marketing ploy, but I bet a wind tunnel would still reach the same fundamental conclusion with Foam Dave Zabriskie over 100 minutes of testing.

My point is that the average consumer doesn't have the background in statistics to care that the testing doesn't necessarily prove that it saves exactly 33 seconds but might save between 15 and 45 seconds. I agree that they could do more testing and narrow it down, but I doubt anybody would care except bloggers and shop rats.

I'm not too fond of using time units t express an aero advantage. If I had my way, all aero benefits would be expressed as watts saved at 40km/h.

16. Uhhh...

Spindle diameter is governed by structural considerations. I don't have a set of speedplays on me but I'm fairly sure most road pedal spindles are basically the same diameter.

Cd for a cylinder is inversely proportional to dia so you won't see huge drag reductions just considering the pedal alone.

Even going to Ti, you just save weight, the member itself isn't going to get much thinner (stiffness considerations) - it's just less dense.

I don't remember speedplay cleats having a smaller frontal area than LS style pedals + cleat thats for sure - while you might induce more drag due to turbulence because of the LS caging system at that point I'd really have to see the CFD with a crank arm and foot attached for the full pedal rotation as a function of the degree the rider toes down - most riders don't maintain flat feet and that has far more profound drag effects.

Drag analysis is somewhat circumspect as well - this sort of Cd stuff measured to the second decimal pace only matters for pros who spend lots and lots of metered training time improving their aero positions (and who don't then blow seconds looking down during a TT) - things like riding in the hoods, or using water bottles and exposed shifter/brake cables blow anything on the order of 0.0X into the realm of irrelevant.

Thank you. I read every single comment.