I like both ideas, Malc, but unfortunately I don't think either would be feasible. Option 1 would be ideal, but I think to provide a scaled-down version that is as good as the full-sized version, would take considerable effort and a lot of trial and error. This would have to be done for all bases in a group test. It would be great if the original authors could provide a super-scaled-down version, but I don't think that is likely to happen (especially due to the number recipes sourced from restaurants).
In terms of option 2, I just don't think you would be able to make the base as good as it would be were it made as per the recipe. I am willing to be proven wrong, and there are undoubtedly better cooks here than me, but I just cannot see it working.
I don't think we will ever fully circumvent the compromises of the group tests, as they stand currently. Phil's idea may have some legs though and should be considered.