In order to exploit the "inefficient market hypothesis", you need to know what information is reflected in stock price and what is not. This means you need to know the market, not just the stock.

I'm not sure that's true. e.g. if you have a company that has real estate worth $10/share, and a business that's worth $10/share, and the stock trades at $15, you don't really need to know whether the market underestimates the real estate, underestimates the business, or some combination of the two.

I guess for certain kinds of dirty hedges (e.g. trading the crack spread and trading oil refinery stocks) you might need to know that.

You don't have complete information, so the market may know something about the stock that you don't. Even if you're right, the market can stay irrational longer than you can stay solvent. So you do need to know the market.

That's true. And I may know something about the stock that the market doesn't. It's pretty impractical to make your decisions by finding out that everyone else is wrong, rather than by ensuring that you're right (or as right as you can be). It doesn't scale, either -- AT&T used to have over a million shareholders, and interviewing every one of them about whether Ma Bell was going to raise their dividend next quarter sounds like a real chore.

Even if you're right, the market can stay irrational longer than you can stay solvent.

If you're leveraged, yes. If you aren't leveraged, you're going to be solvent forever. And anyway, arguing about the timing obscures the real decision -- I wouldn't discourage someone from pursuing a career because I didn't know whether or not every employer would recognize their talent after the first interview.

The market is the inefficiency, not the stock. In order to exploit these iniffeciencies, you need to decouple the fundamentals of the underlying company with the fundamentals of the market. The price at which a financial instrument trades is simply the sum of the market supply and demand. That's the big secret. If supply runs out and demand is still there, the price will go up until someone is willing to supply more stock.

That's true, but it sounds like you're talking more about speculation (guessing prices irrespective of values) rather than investing (purchasing when the price does not reflect full value).

A Random Walk and Efficient Market Theory are misunderstood and misrepresented.

People equate efficient market with perfect market, meaning the market "knows" the right price. All that efficient market theory states is that information is communicated efficiently, not that the information is correct. Other bubbles would never happen.

Proponents state that prices are only correct in the long run. Malkiel (author of A Random Walk) advocates whole market index funds simply because you cannot know what information is correct and beating the market is a fool's errand. Again the case in point being the irrational run up in tech stocks not based on fundamentals but based on the greater fool theory. Information as pricing was communicated effectively but the information is not guaranteed to be either correct or rational.

The stock market is a weighing machine not a voting machine - Benjamin Graham.

However, I do enjoy Cuban's take on this as he's been there and down that. I think he could write a great book at some point.