Kavanaugh: Should We Care About The Details?

Or should we ask ourselves to generate a generic template of basically the following form, and apply it uniformly to cases that aren't extremely unusual:

Suppose A accuses B of sexual assault; suppose there is basically no collateral evidence either way. We should conclude that ______________.

I find myself trying to piece together a story out of wispy details. I just read an assertion that Judge, the alleged witness, was a bully. I felt that move the needle toward BK did it--appreciably. I find this happening a lot in such cases. I suspect it's a big mistake. Sometimes it takes just one little thing to move the needle. This is one of the things that leads me to prefer the generic approach. Make a general judgment and, except in extraordinary cases in which the alleged details warrant overriding that judgment, stick to it.
   Well into my 30s I realized I automatically kind of believed accusations of criminality. (I told my friend Peter the public defender this, and he said: "You'd make a good judge." Damn...) So I tried to kick that habit. I really believed accusations of sexual harassment and assault. That ended as I learned more about academic feminists, and after I saw actual instances of them using bogus accusations of "hostile environment sexual harassment" (by which they meant: questioning feminism) to achieve political ends. Also after my brother was falsely accused of assault against my psychopathic father, who was the actual assailant. A clearer case of injustice there never was...but there was no way for the judge or the jury to know that. They ultimately rendered the right verdict...though it was a near thing, and not necessarily decided on the most rational grounds. I knew what had happened...but had I been on the outside, I might very well have ruled against my brother. Now I know better. To the argument But why would the accuser lie about this?, I now think I know the answer, and it's: You simply cannot imagine all the possible reasons.

   At any rate, this game of piecing together wisps of recollections and allegations...I just think it's mostly a fool's game. Which makes me think I ought to stick to the generic approach. Which is typically a good approach to reasoning anyway.
  In fact, why not:

Suppose A accuses B of something; suppose there is basically no collateral evidence either way. We should conclude that ________________.

Seems to me that We must suspend judgment may very well be the front-runner... Though, OTOH, I recently started wondering whether--independently of questions about accusations--we ought to generally believe people's assertions. The unstated idea that all assertions (and not just accusations) are, as it were, guilty until proven innocent may be a bit of creeping Cartesianism...and I've been infected by enough Peirce to now be skeptical of such things. So I may be way wrong about this.