[brownbat]

This is a really important puzzle. In other terms, basically the efficient markets hypothesis vs, well, a week of GameStop prices.

At certain extremes, a stock price is clearly objective.

If the company is bankrupt and wiping all stock, that's objective. If the company does so well the shareholders demand an immense dividend or buyback, that's objective. Those are the fixed points where stock price is set to money in hand.

Are the swings in the market just irrational participants between those extremes?

We could imagine purely objective superintelligent AI dominating a market, investing only to those "true" values. Such AIs might determine p(bankruptcy) and p(payout), knowing those are the "true" outcomes of the prop bet, and set expected price as the ratio of those two probabilities.

But both those p()s are vanishingly small for most companies, and infinite precision will be impossible. Even if you were nearly omniscient about all current factors within a company, the slightest possible change in either probability could swing the ratio in dramatic ways.

(Add on to that the graveyard of companies that performed well, got fat, and failed to adapt... current performance is somewhat but not fully predictive of longevity.)

So basically, what if EMH is true, but stock pricing is a debate at an arbitrary level of precision, to make it close to meaningless in short time windows?

I'm not an expert so I'm sure professionals or academics would roll their eyes and offer something even more explanatory, but that's my best hunch at resolving this tension so far.

(The upshot of this theory is that it seems to validate strategies that help you zoom way out: low fees, broad diversified indexes, long time windows... those are the real value trades.)