drifting choirs

At some point I had to laugh a bit when I realized that via dynamic tuning
you could actually have a computer-oriented system replicate a choir's
drifting away from the original pitch as the music progressed. But then it
made me wonder: is the reason for that sort of drift not simply physical
imprecision as we're apt to assume, but a result of the choir's natural
tendency to constantly adjust to the appropriate tuning. I spose that would
be hard to measure unless there's some software that takes into account the
deviations inherent in manyvoices singing in unison, averaging them out, and
coming up with a reasonable pitch average.