I think the main problem is that the criteria are too vague and one-sided.
They mostly evaluate how annoying, confusing, painful, or impractical the project is. That fits the joke of the jam, but it does not really evaluate the actual quality of the idea or the design.
For example, a completely random and messy project could score higher just because it is more painful to use, while a more original and well-designed project could score lower because it is inconvenient in a more thoughtful way.
I do not think “confusing” or “annoying” are bad criteria by themselves, but they need clearer descriptions. And I think there should also be criteria like originality, functionality, completeness, and how well the project fits the theme.
Otherwise, it becomes hard to tell whether a project is actually well designed, or just randomly uncomfortable.