From Philip Tetlock's Superforecasting [0] to Robin Hanson's Futarchy [1], there's a strong argument that predictions are entirely valueless without "skin in the game", be it in reputation score, or dollars.

[0] https://en.wikipedia.org/wiki/Superforecasting:_The_Art_and_...

[1] http://mason.gmu.edu/~rhanson/futarchy.html