Hacker News Reader

The control group could be blinded (they're unaware what is being studied).

They just get Monopoly money ;)

There are a number of study designs that don't require a null-intervention control group. Some of those would have been interesting, and might actually have yielded something closer to the true effect.