I do research in this area and it's fairly well established that when you go from something like five points to two points with ratings, you throw away tons of information. There's diminishing returns with numbers of points, but as you go lower you lose information.
The "ratings don't matter because what you want is implicit signals from peoples' actual behavior" is also disingenous because the rating behavior is a behavior that's directly tied to the stimulus in question. Not saying that indirect behavioral correlates aren't useful, only that the rating is a very powerful, direct correlate that tends to be very specific. Going back to the topic of the thread, sure, all those Facebook likes are going to be useful in predicting how much you like a candidate, but you're sure as hell going to get a lot of information by just asking them "on a scale of 1 to 5, how much do you approve of X?"
I honestly thought the 1-5 ratings were not useful since I either like or don't like movies, I am not interested in nuances. But, it's not working out as I expected.
This is far from my previous stance of "Huh, movie I've never seen before and in a different language? It's 95% so yeah, I'll give it a go."
And before someone says anything, I do vote up and down quite frequently. But can't help but notice that my suggestions were better when I had the more nuanced star system.
[1] https://i.imgur.com/MOt3XlL.png
[1.1] How I'd rate these. Cars 3: not really interested. Liked the first though. Stitches: idk, doesn't look appealing. Teeth: Classic cult film but yeah... Big Mouth: I have ZERO interest in watching this, please stop suggesting. I have downvoted this! Waterboy: I like it, but far from 95%. I'll give it like an 80.