You could simply test BAC at home with a specific beer, over a certain amount of time. Then drink the same beer (or a beer with the same alcohol content) when you are performing the "experiments" in a public setting. This way, you will have a very close approximation of the BAC.
-Dr. Checkevertim
Quite right Dr. Checkevertim, this would indeed be a most logical approach. However, my personal research space lacks the necessary scientific equipment to ascertain this data, and the current priority for expenditure of the limited funding backing this experiment will most likely be funneled into purchasing the lightly carbonated consumables...
How about we organise a handicap tournament?
Rating Difference Number of beers drunk by stronger player before game
0 - 99 0
100 - 199 1
200 - 299 2
300 - 399 3
etc up to a maximum of 10
This is a wonderful idea... However, I feel it is necessary to highlight that it is predicated on the assumption that there is a simple linear decrease in playing strength per beer consumed. Many of our esteemed colleagues on this thread have suggested that this may not be the case, with, for example, a single beer tending to loosen inhibitions and potentially increase playing strength. It is unclear if this hypothesis is correct... No data has been posted, but you may want to consider recalibrating your ladder if this is indeed the case...
In all "seriousness", we had the same kind of debate at the chess club a couple of months ago about giving one pawn odds in a blitz game for one minute.
There is no easy way to take pawns out in the opening without giving attacking lines, so we took positions from master games where we could take out a pawn with no obvious strategical implications.
The experimental data looks like for 5 min blitz (but remember, after the opening, so it is more likely that the whole game would be 10 min), one pawn is roughly equivalent to one minute (if the opponent starts with 4 min instead of 5), two pawns are more like 3 min handicap. And three pawns are more than 4 min, but it might just be that the positions we took are so obviously lost after stealing 3 pawns...
So, by analogy, I suspect beer vs rating is strongly nonlinear.