When doing any sort of theoretical work in finance or economics, academics often make the assumption that markets are rational and efficient. This assumption is almost universally false. They know that markets are neither: markets may or may not be rational (we can never realistically gather enough data to know thanks to information asymmetry) and they certainly aren't 100% efficient over most timelines.

Just because the assumption is wrong doesn't mean it's not useful: without it, you can't even build a reasonable model of financial markets. But because the assumption is core to financial modeling, this means that all financial models are wrong. In fact, that's exactly how my financial modeling professor opened up the first class: by explicitly stating that all models are wrong in some way. You have to understand the ways that the model can be wrong in order to make any substantive claims about it. The Black-Shoals options pricing model assumes market liquidity; it's not very useful if you're talking about an asset that isn't very liquid.

Misunderstanding when models should and should not be applied is one of the core reasons we got into an asset bubble in the mid-2000s. Banks were pricing financial products using models whose assumptions did not hold over the long-term. They didn't understand how their models were wrong until it was too late, and it ended up losing them (but ultimately the American taxpayers) a lot of money. But because the assumption that markets are always rational and efficient is wrong, it doesn't follow that markets are never rational or efficient. Markets can be rational and efficient, but it depends on the relative timeframe and the structure of the market involved.