Yes, and I think that is essentially why we're seeing these disagreements.

I've heard from lots of professors that a good conference gets a lot of "very-good-but-not-great" submissions and the job of the program committee is to pick the best among these. I wouldn't be surprised at all if minor personal preferences (which from the outside look rather random) ended up having a big say in the fate of a particular paper. Maybe some reviewers are more forgiving of poorly-written but technically strong papers, maybe some reviews consider certain fields "dead" and so are biased against them, reviewers tend to wildly different standards on how extensive an experimental analysis should be to be acceptable, ...