^I think that the methodology in the Kari-Grant-Tory experiments is generally somewhat sloppier than in Adam-Jamie experiments. Which is odd, because you'd think it would really be the show's writers/researchers, the people you don't see on camera, who come up with the actual experimental designs.
Yes, in the very same episode we have Adam talking about how a bigger sample size gives better results, and yet this principle seems mostly ignored on the other team.
Kari did at one point refer to the parking data as "a statistical tie" however, which usually refers to it being within a margin of error 19 times out of 20, which is a more complex statistical concept than simply comparing two means.
It is possible that they consider a 75% chance that the test reflected a real difference good enough for their purposes. It does strike me as better than "plausible," I must admit.