A much longer quote from Alex Hutchinson's article:
What does the carbon-fiber plate do?
The first thing to make clear is that whatever you read about exactly how the shoe works, both here and elsewhere, is speculation. No one, probably including even the designers, can really be sure exactly how the various parts are interacting to produce the observed effects. That was the message I got from Rodger Kram, the head of the University of Colorado’s Locomotion Laboratory, and his post-doctoral research associate Wouter Hoogkamer, who conducted independent testing of the new Nike shoe that is due to be published this spring.
Kram and his colleagues have conducted a series of elegant and ingenious studies in recent years that build toward the current result. Back in 2012, they showed that cushioning in shoes can improve running economy (i.e., reduce the energy needed to maintain a given pace).
Then last fall, they reported the effect of shoe weight, confirming the long-held maxim that each 100 grams (about 3.5 ounces) of shoe weight costs an extra 1 percent of energy. And crucially, as part of the same study, they tested actual running performance in a 3,000-meter race, to see whether improvements in running economy translate to real-world racing gains. The verdict: For every 1 percent gain in running economy, you really do get about 1 percent faster, at least under the conditions they tested.
So that sets the stage for their test of how much the new Vaporfly 4% improves running economy (and, by extension, marathon time). We’ll have to wait for the study to be published before discussing Kram’s and Hoogkamer’s specific results, but the general answer is: substantially. Enough to bring a sub-two-hour marathon within the bounds of possibility, if other factors like the course are tweaked to perfection. (For Kram’s and Hoogkamer’s newest research on those other factors, see this Sweat Science column.)
Is that all because the shoes have springs in them? Well, yes and no. “Virtually all modern running shoes already have springs,” Kram says. “We call them foam midsoles.”
The Oxford English Dictionary has a bunch of definitions for the word “spring”, but the one we want is “An elastic contrivance or mechanical device, usually consisting of a strip or plate of steel (or a number of these) suitably shaped or adjusted, which, when compressed, bent, coiled, or otherwise forced out of its normal shape, possesses the property of returning to it.”
Springs aren’t just coils of metal, in other words. So Adidas’s Boost foam, like the midsoles of pretty much every shoe, functions as a spring when it compresses, then springs back to push the foot onward. That’s not a technicality or a minor detail; it’s the whole point of the Boost foam, which is why Adidas boasts about the material’s “industry-leading energy return” that powered Wilson Kipsang and Dennis Kimetto to the two most recent men’s marathon world records.
The carbon-fiber plate, on the other hand, is something different. It is, certainly, a stiff plate that when “forced out of its normal shape, possesses the property of returning to it”—so it’s a spring, but in a different sense than the Spira spring. Is the springiness what makes running more efficient? That’s less clear.
Back in 2006, an Adidas-funded study at the University of Calgary showed that inserting a carbon-fiber plate into the midsole of Adistar Comp shoes—much like the new Nike shoe—could improve running economy by about 1 percent. The hypothesis, at that time, was that the plate might reduce the amount of energy wasted when the toes bend, but the researchers concluded that the final reason for the improvement was “still not fully understood.” It's perhaps worth noting that the two authors listed on the new patent of Nike's carbon-fiber plate, Jay Worobets and Geng Luo, both did their Ph.D. degrees with Darren Stefanyshyn, the Calgary researcher who was the senior author of the 2006 Adidas study. This development didn't appear out of thin air.
A South Korean study published last month found similar results, noting an improvement in running economy as long as the carbon fiber plate wasn’t so stiff that it restricted toe flexion. And as Kram points out, the role of these plates is akin to what the shanks in boots have been doing for over a century.
And there are plenty of other things going on when you stiffen your shoe. For example, as you hit the ground, the center of pressure may move toward the front of your foot more quickly, which effectively changes the length of the “lever” that is propelling you forward—a benefit that, once again, echoes the claims about Pistorius's prosthetics. In other words, there may be mechanical advantages from having a stiffened sole that have nothing to do with energy storage.
That’s the view of Nike’s designers, who see the carbon-fiber plate’s benefits as a result of stiffening the shoe rather than adding any energy return. The “parabolic” curve of the plate, rather than adding springiness, serves to minimize wasted energy at the toe joint and optimize ankle position, they say. From Kram’s perspective, too, the spring in the new shoes is primarily provided by the foam, while the carbon-fiber plate probably functions more as a lever.
Others may argue that the shoe works for different reasons, including spring-like energy return. That's a debate that perhaps will be settled with further testing. But for practical purposes, that’s less important than the fact that it does, seemingly, work—which means we need decide how it fits into the sport.