Yep, as much as people hate it, when it comes to recommendation, up/down is all the data you need. Trying to build a system out of a 5 star system just adds unnecessary complexity, and the reality was that people used the star system differently making it even harder. up/down thumb is explicit and cleaner to work with.
The problem with the up/down system for me is not my own ability to like/dislike specific titles, but more the fact that Netflix no longer displays the average of all user votes. Sure different people used the 5 star system in different ways, and there were some who may have misused it by giving poor ratings to things they never intended on watching, but it was a great signal to me for the extremes.
Scenario: I'm considering some odd looking sci-fi movie to watch that I never heard of before. Ratings between 2-4 stars might not tell me much, but very reliably titles with only one star were terrible movies. Now Netflix happily recommends any and all sci-fi titles, saying they are a "98% match" for me! Sure by category, but when the movie is a low budget dumpster fire I no longer have that instant signaling that the previous rating system gave me.
I agree that thumbs up/thumbs down is probably better to encode appreciation than a 0-5 scale. But for building a recommendations engine a single bit of information is not enough. If I and everyone else flags shows we are not interested in viewing 'thumbs down', then all shows end up with terrible ratings. Similarly 'thumbs down' for a great movie, but seen it already thanks so stop screaming it at me. Thumbs up and Thumbs down is certainly not all the data you need, as is demonstrated by Netflix as it stands today.
