The original statement went for "most Christian churches", not "most Christians" (what you'd get with your "weighting"). Maybe a useless metric but the one chosen up-thread.

Going from there to implicitly weighting by membership seems rather arbitrary: I could also claim to weigh by readership (or hours of reading, or any other metric that includes actually working with the text) which might give a pretty different result, given how the idea of individually reading and interpreting the Bible is a major raison d'être of Protestantism.