I'd contribute to a google doc. We can obviously construct it so no names are involved.
I am very curious about how standardized testing factors in the process because I'm almost thinking that individual members of the committee may be lying to themselves about their own perceptions.
Let's say a Derridean applies to your program. His writing sample is very dense--lots of wordplay, perhaps some things that might be considered jumps in logic or nonstandard argument strategies, and only tangential relationship to politics and history, which would be a weakness since those are vaguely "in" right now.
I can't help but imagine that if a student with such a writing sample applies with, say, 155/149/4.5, that paper will be read entirely differently than if he applies with, say, 169/165/6.
Am I projecting too much when I imagine that in the first instance, the reader might be less merciful to what seem like errors, whereas in the latter instance the reader might be more inclined to double back on himself and think: Perhaps there is something here I missed the first time, I'll give it a second look?
I guess in short I'm saying that I think that besides being a relatively small deciding factor in themselves, I could also see standardized test scores as having kind of a larger ethos effect upon the entire application. If for example, like me, you had a difficult undergrad experience, you might rely on test scores to recast yourself as "troubled smartypants."