We're getting closer. Still not there.
A couple of things about the studies you linked to: The first study you mention appears to have been done by Nike themselves. Any study done by a party with a financial interest in the outcome is immediately suspect on that grounds alone. Maybe Nike was honest, and maybe they weren't, but the conflict of interest is definitely problematic. The second study you mention is a) pricey, if you link to the full article, and b) a total sample size of 14--divided into "approximately equal" groups--is hardly the kind of thing you look to for statistical significance. The second study also seems to be unconcerned about race times, only muscle soreness and fatigue. Those aren't trivial concerns, but they aren't race performance. And, muscle soreness and fatigue are very much within the realm of placebo effect.
A helpful study here might take 200 marathoners (or 10Kers, whatever you want the results to point more directly toward) and randomize them into two groups of equal size (though exactly equal isn't nearly as important here as it is when your total sample size is 14). Then they all run the marathon (or 10K), with real prize money on the line and all the other things that elevate competition. Half run in the real shoes, half run in placebo shoes (presumably some kind of commonly used racing flats). The shoes, however, look identical. If any hype is given about the shoes to any of the contestants, the same hype is given to all contestants. Then you compare results between the two groups. Muscle soreness and all might be interesting, but the key factor, presumably, would be comparison of race times.
Additionally, you would probably block your experiment for sex of the runners.
An experiment like this is costly. And it's highly unlikely you'd be able to get Nike to cooperate with you (they're already making money hand over fist and the results of such a study would be unlikely to improve their position much from what it is now), so making the shoes look alike is going to be a problem even after you buy all the shoes. Potentially, Nike could lose their total investment in VFs in such a study. Unlikely they'll sign on to provide free gear, or even gear at cost, for the purposes of such a study. And then you have to convince all these racers to lay it on the line in shoes they really know very little about.
In any case, this is how you rule out the placebo effect. I doubt we'll ever get to the point where we know on this issue.
My own hunch? The shoes make a difference, but not as big of a difference as lots of people seem to believe. There is an expectation factor here that is also part of the difference. But, that is only a hunch, nothing more.