> questioning the fundamental basis of this.
Empirically, you can look at https://calibration.city/ (among other such trackers) look at polymarket, filter by market midpoint and you'll see that if a market resolves in a year, and 6 months in it's at 30%, the actual event happens at (remarkably close) to 30% of the time.
Theoretically, it relies on standard market theories, like efficient markets hypothesis etc. Basically, however corn comes ot be valued correctly, much of the same mechanisms are present here