Well, I didn't downvote, but I think your comment misses the point. The discussion is about the requirements of a sane metaethics, specifically that you can't have both "mundane things which are still worth trying to achieve or avoid" and "sacred things that can never be compromised for the sake of mundane things". You have to treat them both quantitatively, or else never make a decision on the basis of mundane things, or else be inconsistent; and so there really has to be a point where a larger probability (or preponderance) of a "mundane" consideration outweighs a tiny probability (or preponderance) of a "sacred" one.

The precise degree to which torture is awful is kind of irrelevant to this debate. Pick another example if this one doesn't work for you.