A lot of free-market oriented economic commentators have this stark tendency to misinterpret the Efficient Market Hypothesis as a kind of grandiose statement of laissez-faire solvency; you hear it a lot now that financial regulation is on the political agenda again, but the Efficient Market Hypothesis is one of those too-commonly misunderstood concepts.
The brunt of this misunderstanding can sometimes be traced to the fact that the Efficient Market Hypothesis is often summed up as, "individual investors cannot perform better than the market as a whole," as if this means that markets see perfectly into all the financial possibilities of an economy and select the optimal choices. In reality, it's a kind of tautology: the market chooses what capital projects do and don't receive funding, and thus decide generally which succeed or fail.
The Efficient Market Hypothesis in many senses is valid, but unremarkable. We could just as easily formulate an "Efficient Vehicle Hypothesis:" let's say Billy has to choose between a red car and a blue truck to drive from New York to Seattle and after contemplation, decides to take the red car. Several weeks later, after arriving at the Pacific Coast, he says to himself, "Boy, I'm glad I choose the red car! If I'd chosen the blue truck I'd still be back in New York because it hasn't moved anywhere!" It needs no elaboration that Billy's statement is patently absurd; the red car made it to Seattle because of his choosing it.
The fact is that financial markets are presented various investment opportunities, most of which would be economically viable and thus with significant enough capital, nearly any project can produce returns, even disbarring Ponzi finance. For various informational reasons markets choose one investment opportunity over another, but their situation is not unlike Billy's.
Billy can't know if taking the blue truck would have been faster, or if it would have stalled or crashed, or if it would have won him new friends or anything else. In the same why, there's absolutely no way we can call market outcomes "optimal" because we cannot compare them with their opportunity costs whatsoever. They are "efficient" in an economic sense, but every possible economic equilibrium is as well by definition. So unfortunately the "E" in the EMH functions as a vehicle for smuggling in value judgments for a lot of commentators.