Thanks Tom, you earned it, you named the phenomenon... also mentioned Robert, but will have to wait for a future episode to use 'Chung'ed the $hit out of it..'

Next episode is on non-linearity, which is shaping up to be super fun and interesting, I have some great stuff on both bicycle nonlinearities.. mainly aero as well as some cool stuff on computer processors and others..

Would love to hear other ideas or inputs for that or other future episodes!

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Josh, I sent you a suggestion via ST pm (Silca related though).

---

_______________________________________________

Since you've done "marginal gains" and "asymmetry" and next you're doing "nonlinearity," the obvious topics after that would be "thresholds," "loss," and "robustness."

That anecdote about passing up a space in the parking lot? I examined that problem back when I was in grad school. I envisioned it as a problem about driving up to campus and seeing a parking spot 4 blocks away. Do you take it, or do you try to get closer? If you pass it by and don't find a closer spot, it's likely to be gone by the time you turn around and head back. This turns out to be an optimization problem about risk, loss, and gain. It turns out I wasn't the first to have looked at that problem: it turns up in the literature as "The Secretary Problem." The formulation was odd: you're interviewing potential secretaries and the problem is whether to hire the one in front of you or continue searching. This never made sense to me until I learned that the original problem was called "The Marriage Problem." The irony is that my research question at the time was actually about marriage.

Thanks so much for the topics and love the feedback on the secretary/marriage problem. I honestly hadn't overlayed the marriage problem onto parking before, but I totally see it now. I would have initially thought it to be different as you are essentially able to pre map and determine the ranking of every individual spot before you enter the lot or campus, rather than evaluating them only as they are presented in random order...but in effect despite the predetermined ranking, they are presented randomly as people vacate them.. love it.

I'm not good at reading from a script so we do the podcast live to an outline so the parking story was a bit out of left field in the moment, but funny you mention the parking story as after we taped my only thought was that the best spot is the one opposite the cart return.. minimal distance to return the cart after loading your car, yet opposite side of the aisle to minimize risk of your car being run into by some stray cart rolling out of or missing the cart return!

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Listened to the second cast on the way into work this morning. You had my nerd boner rock hard. I ask that you don't dumb it down like a lot of infromation that we get tends to be.

Have you thought about also video recording for youtube? That seems to be the hot trend as well.

Another topic could be "renewal," and the example would be team pursuit. That's an interesting optimization problem as you shuffle the order of the riders and the length of the pulls. Different riders expend and recover at different rates, plus riders come in different sizes so a bigger rider doesn't get as much recovery when positioned behind a small rider. That's why you shuffle to find an optimal ordering.

Sounds like a classic differential equations problem ;-)

---

http://bikeblather.blogspot.com/

Team time trials are a pretty interesting problem in general, especially when there are hills or changes in wind speed or direction.

So it's a little worrisome that we gave you nerd wood and you now want videos ;-) But thanks for the feedback!
If that lasts more than 4 hours you might need to see a professional..

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

I look forward to listening!

I'd be curious to hear about different composite construction techniques (such as filament winding), their advantages, disadvantages, limitations, etc. I'd also like to hear about how robust different materials are (e.g. some, like me, used to really baby carbon wheels when in reality 303s are actually quite tough). Another potential topic would be the cost components of something like a wheel. For something like a 45mm deep carbon clincher, how much of the cost is tooling, how much is labor in the layup, and how much is assembly of the actual wheel?

Man, I was listening to the podcast when it got into rolling resistance thinking "I'm going to ask about track stuff" and to my surprise, you guys then launched into track stuff. Great! I'm a little surprised optimal pressure at Major Taylor was 115-20, I'd have guessed it was closer to 100. I wonder what optimal pressure somewhere like DLV is... 80? :D

---

My Blog - http://leegoocrap.blogspot.com

I'm loving the podcast so far, Josh. Keep up the good work.

I'm curious to know if vibration dampening features in frames have an effect on the speed a bike travels across gravel. I'm not sure if "rolling resistance" would be an appropriate term for this or not. Basically, will a stiff-as-a-board bike frame require more output from a rider to go the same speed as a a frame that has traditional front and rear suspension, leaf springs (lauf), a future proof shock (specialized), Zertz (specialized), a pivotless rear shock (Dogma K10), or just some really well executed seatstays and geometry that create a smooth ride?

Or, is the additional dampening purely for rider comfort without any direct impact on speed?

I know this is a little different than what you probably study at Silca, but if anyone could figure out the potential marginal gains here, I think it would be you.

I loved the second podcast on asymmetry. I have a question for you though: Going back to your testing at Roubaix on the cobbles, you guys kept lowering the pressure and the times got faster every time you did so until you broke the wheels. I'm wondering why at that point you didn't switch to larger volume tires and repeat the same test? It sounds like lower pressure wasn't explored further due to the tire size limitation (bottoming out) and not actually reaching your inflection point.

Did you not use larger tires because:

Was it because you were already using the max tire that would fit on the bike?
Already using max tire size from the sponsor at the time?
Or already conducted an aero/rolling resistance weighted analysis to determine the optimal speed when factoring the course as a whole?
Did you conduct your own analysis with larger tires just for your own research for a later date?


Thanks Josh!

-Iain

---

_______________________________________________

Awesome question Rob! What we are after here to improve speed isn't damping, but rather compliance, the Lauf fork being the best example of something that can significantly reduce impedance at a high level of efficiency. In any suspension system, it's the compliance that it most important to speed and the damping is necessary to control the spring... for small amplitude inputs an undamped spring will be faster, however, as the amount of travel increases, damping becomes necessary for a whole bunch of reasons.

My experience has been that on any surface where the roughness is greater than the safe compliance range of the tire, you will see speed benefits from some level of added compliance in the frame. Having even 10mm of travel in the bike also makes a phenomenal difference in the likelihood of pinch flatting. The problem then becomes that for a race like Roubaix, the losses to suspension on the smooth parts or when sprinting at the end of the race can be great enough to offset any gains from earlier in the race on the cobbles.

So yes, a more compliant frame definitely helps with speed over rough surfaces. The current school of thought for a race like Roubaix is that a more compliant but still rigid frame will be fastest on the whole, whereas for an event like Dirty Kanzaa you might be better off with something that has a small amount of suspension compliance in the frame and fork.

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

So the time context here is pretty fun... in 2007-2009 when we were really pushing hard to understand the problem, the belief amongst the riders was still that the 24mm tire was faster in the dry and 27 was only good in the wet because it was 'slow'. We also had to contend with the bikes of the time, none of which could really handle more than a 27 or 28mm tubular tire (so measuring ~28mm)..and sadly we haven't been that much better off as of late as I think about it!

So way back then we figured pretty quickly that the 27mm Vittoria that the team had access to was faster on the cobbles than the 24 AND safer regardless of wheel.. so at that point we realized that a good part of the solution here was going to be educating the riders and convincing them to ride the bigger tires. There is a photo on the internet of me with Bjarne and Cancellara in the LSWT in San Diego, and the whole point of that entire test was to put him on a road bike and show the benefits of the aero wheels over the traditional classics wheels and to then show that the added drag of the 27mm tire wasn't that bad, while we did field testing on Crr to show them that the 27mm tire had same or better Crr (which they were all conditioned to believe wasn't true.. wider is higher Crr was the firm belief). I'd estimate that we spent probably 20% of the Roubaix project time doing things to help with rider education and psychology about all of this as the pushback was just so strong.

So yes, we were using max tire that would fit the bike, which was also the max tire size made by the sponsor.
The whole 303 super-toroidal design came about in 2009 from this.. once we settled on the 27/28mm tire we had to design a wheel around it to minimize the aero penalty as it was somewhat significant on the 24mm wide 303 and 404 rims of the era.. and those had previously been considered to be 'wide'.

It would really only be 2012 or so before the bike manufacturers started to catch up with these changes... painful for us at Zipp, the Specialized Tarmac launched the year of the first Roubaix win (2010) could not fit 28mm tires or the 28mm wide 303 rim, even if it had a smaller tire as they had bulged the chain stays inward seeking stiffness.. so looking back it seems like this just happened, but it was really a pretty significant battle for a few years there to get this to change.

In the last 5 years the limiting factor has continued to be the frames and rim brakes... I've been involved with 7 of the last 8 Roubaix winners and all 5 of them since buying SILCA, and the challenge has been finding ways to get 30mm tires into the bikes and/or the brakes.. so you can extend dropouts and things like that, but then you need long reach brakes or offset brake bolts.. we've seen it all. Last year Peter Sagan won on a very custom bike with tires labelled '28mm' and measuring just under 30mm stuffed into some 'tuned' DA rim brakes... Ideally we get them onto disc brakes and the sponsors making 34-35mm wide rims that can be solidly aero with 32mm tires..

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Marginal Gains Podcast episode 3 just went live, it's the Q&A episode and includes a few from Slowtwitch family members, but most awesomely, includes a 4 minute call-in from THE Robert Chung!!!

HUGE Thanks to Robert and everybody who had questions, we couldn't answer them all, but are moving the format to include a topical show, then a Q&A show 2 weeks later, followed by a show 2 weeks after that.

Please keep contributing your awesome questions, thoughts, ideas, etc.

Josh

https://marginalgainspodcast.cc/ask-josh-anything-001/

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Cool concept and podcast Josh...thanks for putting it out.

I’d be interested in some marginal gains in other sports. The budgets in F1/NASCAR make for some interesting opportunities, we’ve seen skiers in speed events go from long to short to somewhat longer again, horse-racing has been refining a pretty similar model for over a century, track&field is measured in 1/1000’s of a second ( shoes??), even pinewood derby cars from cub scouts I think would all be fascinating and maybe have some parallels to bikes and tri’s.

---

" I take my gear out of my car and put my bike together. Tourists and locals are watching from sidewalk cafes. Non-racers. The emptiness of of their lives shocks me. "
(opening lines from Tim Krabbe's The Rider , 1978

Thanks for the kind words. Actually, you're one of my heroes.

Oddly, I worked on neonatal and perinatal outcomes in NICUs a couple of decades ago with the CPQCC, which has links to the current CMQCC. One of the things we were trying to do back then was figure out how to convince people to change their behavior and follow checklists.

Thanks TriDevilDog, awesome suggestions and we have some great car racing stuff in future episodes! I also love the topic of horse racing if for no other reason than the unexpected parallels to human sports in terms of negative outcomes and the massive power imbalances between management and the athletes, great ideas!!

Funny enough, we had sort of a reputation for our pinewood derby cars, and now I bring the scouts each year into SILCA and we teach them about the mill and the lathe, and also the physics behind the cars, then show them how to make killer cars, how to make an axle lathe out of a hand drill, etc. AND because of SpeedBalance I have a massive supply (see what I did there) of tungsten slugs for them to use which are just so much better than the weighted putty or zinc stuff that they sell now that you can't buy real lead weights anymore!! Really fun to do a day with the kids and then hear them talking about density, mass and potential energy!

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Robert,
I'll send you an email.. sounds so much like stuff she's working on with the group at CAPE/Stanford and also the AAP..
Doesn't surprise me that you were in front of this decades ago!!

Also, next episode is non-linearity, talking aerodynamics, computer processing and more, before going back for asymmetry Part II.

Please everybody send topics, ideas and questions!

Josh

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

As I was listening to podcast #2, I couldn't help but wonder if we're testing crr of tires (against one another) correctly. Let me preface this by saying I'm an idiot, so just remind me of such if this doesn't make any sense.

Instead of testing all tires (on drum rollers) at the same pressure, should we not test all tires at a range of pressures? To find out which tire performs best as its depressurized? Or which tire has the least amount of crr increse as the tire is depressurized? Since we want to operate on the left side of the break point, shouldn't we test for the internal tire losses at low pressures? Or can we just assume that the tire that performs best at 100 psi will also perform best at 50/60/70 psi?

Sean,
You are totally onto something and we have some future plans to do some cool stuff in this area..
yes, current testing is oversimplified, but for a couple of reasons. One, the test rigs to do what people are doing now are very expensive and to really get to the level that you are imagining will probably increase cost of test rig tenfold.. secondly, if you look at what BicycleRollingResistance does today, each single tire test probably requires 6-8 hours of work just to get a single curve showing a single load on a single surface at 6-7 tire pressures. Adding alternate surfaces, different loads, and more granular pressure changes will increase the time required to capture the data exponentially.

We saw this with wind tunnel testing, the more refined the testing, the more time and $$ required to do a single product or test, while the data coming in is coming in with diminishing returns as the directionals from part to part tend to hold between simplified testing and more complex testing, i.e. if tire a has lower Crr on a smooth drum it almost certainly has lower Crr on a rougher one.. though admittedly, with tunnel testing, we occasionally found outliers here where this broke down and the results created new opportunities to do something unique or better.

More to come..

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

You've gone too far. You're too deep in the rabbit's hole.

Do you even marginal gains, bro?

Marginal Gains Podcast Episode 4 just went live: Non-linearity, covering some fun aerodynamic stuff as well as a minor dive into computer processors

We are taping our next Q&A episode a week from today, so please be sure to get your questions in, and if possible, please call them in so we can include you on the show!

Thanks
Josh

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc