My typical (average) percentage incorrect has been close to 40% - and discussions have gotten hot and heavry.
40% is a good number - in my much smaller contribution of questions than yours I've aimed at, and generally hoplessly missed, 33.3% incorrect, because that seems a reasonable level for teaching (which what I want to do - teaching not testing) but 40% is pretty much the same; maybe some of my miss rate has been because people think writing (writing, not copying) code to determine the answer is cheating, which is absolute rubbish for some of the questions I've posed (for example I'd like to know the references, if anyone found any documented answers, for some of my collation questions).
Sort of figured that I would mend my ways ..... but fear not, that urge will soon disappear .
Well, I'm looking forward to the rest of what's clearly a long series of questions: your rate of question generation must be many times mine.
BTW: I'm pissed off about the collation on Adventureworks :-D