Is the data you're talking about the data from Ian Walker's study in Bath, England? The sample size was one rider - himself. Other folks have analyzed his raw data and disputed the statistical power of the effect described. That rebuttal has also been re-rebutted. So that's three published studies about this, and three sets of headlines, but as far as I can tell, there's only ever been one actual study that collected data, and it was based on one dude in one area of one country.