Efficient markets mean different things to different people. There are at least three different forms of the efficient market hypothesis. Also, you haven't defined what you mean by "capitalizing" or making money in the market. So, this question is pretty much a Rorschach test. Even in a "strong EMH" market, there are two ways to make money: the risk free rate (the time value of money) and the risk premium (the insurance you are paid in order to compensate you for the chance of losing your investment). These are the returns a buy-and-hold investor expects to capture. In a "strong EMH" market, by definition, nobody can expect to earn excess returns (above and beyond investment returns). If you choose some other definition of efficient markets then you might reach a different conclusion. Martin
If one is assuming that market efficiency refers to the strong version of efficent market hypothesis, the only answer to this poll would be yes. Efficient market hypothesis basically states that all information about a company is already completely priced into the current stock value, and that the future price of the stock is impossible to predict (random), so it is therefore impossible to predict future price moves in the market (and make money as a result of these moves). So my take on this poll is that the markets must indeed be inefficient in order to make money from them. I have only read one academic article about the markets that I have agreed with fully thus far, which basically suggested that our return expectations are based on a weighted moving average of past returns (this was used as an explanation for the stock bubble a few years back). So basically, if a stock moved up 20% yesterday, 10% the day before, 5% down the previous day, etc., the hypothesis is that one would place the most weight on the most recent return when forming expectations for tomorrow's return.
Delays in the availability and dissemination of information can lead to short-term inefficiencies. Much of the current charting and analytic software is burdened with such delays, using one-minute-old data.