Revealed preferences vs. misaligned incentives
As far as a lack of predictive ability I think you're right. I'm more just trying to draw out a common dichotomy that comes up in certain kinds of discussions about how we ought to spend our time.
For example, some people enjoy playing video games but then occasionally feel vaguely ashamed that they didn't spend their time doing something more productive. In these cases, they might be unsure about which view to take: RPT says that the time they spent reveals their true preference for video games, but MIT says that they fell prey to the short-term incentive structure. If we want to achieve long term goals, this becomes an issue because we want to be true to our nature, and not insist on lying about our goals, but we also want to be able to avoid local maxima in the form of quick pleasure.