Theory question: How many lines are "off"?

[BobJoeJim]

Obviously, the ideal of sports betting would be to bet the +ev side of every game that offers a +ev side, while staying away from "correct" lines where the juice makes both sides -ev. My question, which I'm not sure is quantitatively answerable, but that I do think is interesting, is: What percentage of lines are accurate enough that the juice makes both sides -ev, and what percentage of lines are profitably bettable (if you can identify the correct side to take)?

Obviously this will vary from sport to sport, so for the sake of this discussion let's focus on football sides and totals (not moneylines, let's keep it to the "50/50" bets). I'm curious to hear people's thoughts on this.

If you're curious what got me thinking about this in the first place, it was a combo of my not finding a play I liked when I analyzed the Rutgers/USF game (since I'm trying to analyze every "only game of the day"), and Assani's threads where he handicaps every NFL game. Basically, another way to put this question is: if you trust your abilities as a handicapper, and you analyze every single game each week, and are willing to take any edge, no matter how small, what percentage of the games should you probably be placing bets on? Let's assume -110 juice on all the games, again just for consistency's sake (though if you want to discuss how the percentage would increase as the juice decreased, that could also be an interesting topic. I'd like to hold it off at first, though, because it's sort of level two of this question, along with the effects of line shopping, and I'd like to focus on level one first.)