From wikipedia, I think these are illuminating towards the subject:
Arguments concerning the validity of the hypothesis
Some observers dispute the notion that markets behave consistently with the efficient market hypothesis, especially in its stronger forms. Some economists, mathematicians and market practitioners cannot believe that man-made markets are strong-form efficient when there are prima facie reasons for inefficiency including the slow diffusion of information, the relatively great power of some market participants (e.g. financial institutions), and the existence of apparently sophisticated professional investors. The way that markets react to surprising news is perhaps the most visible flaw in the efficient market hypothesis. For example, news events such as surprise interest rate changes from central banks are not instantaneously taken account of in stock prices, but rather cause sustained movement of prices over periods from hours to months.
Only a privileged few may have prior knowledge of laws about to be enacted, new pricing controls set by pseudo-government agencies such as the Federal Reserve banks, and judicial decisions that effect a wide range of economic parties. The public must treat these as random variables, but actors on such inside information can correct the market, but usually in discrete manner to avoid detection.
Another observed discrepancy between the theory and real markets is that at market extremes what fundamentalists might consider irrational behaviour is the norm: in the late stages of a bull market, the market is driven by buyers who take little notice of underlying value. Towards the end of a crash, markets go into free fall as participants extricate themselves from positions regardless of the unusually good value that their positions represent. This is indicated by the large differences in the valuation of stocks compared to fundamentals (such as forward price to earnings ratios) in bull markets compared to bear markets. A theorist might say that rational (and hence, presumably, powerful) participants should always immediately take advantage of the artificially high or artificially low prices caused by the irrational participants by taking opposing positions, but this is observably not, in general, enough to prevent bubbles and crashes developing. It may be inferred that many rational participants are aware of the irrationality of the market at extremes and are willing to allow irrational participants to drive the market as far as they will, and only take advantage of the prices when they have more than merely fundamental reasons that the market will return towards fair value. Behavioural finance explains that when entering positions market participants are not driven primarily by whether prices are cheap or expensive, but by whether they expect them to rise or fall. To ignore this can be hazardous: Alan Greenspan warned of "irrational exuberance" in the markets in 1996, but some traders who sold short new economy stocks that seemed to be greatly overpriced around this time had to accept serious losses as prices reached even more extraordinary levels. As John Maynard Keynes succinctly commented, "Markets can remain irrational longer than you can remain solvent."[citation needed]
The efficient market hypothesis was introduced in the late 1960s. Prior to that, the prevailing view was that markets were inefficient. Inefficiency was commonly believed to exist e.g. in the United States and United Kingdom stock markets. However, earlier work by Kendall (1953) suggested that changes in UK stock market prices were random. Later work by Brealey and Dryden, and also by Cunningham found that there were no significant dependences in price changes suggesting that the UK stock market was weak-form efficient.
Further to this evidence that the UK stock market is weak form efficient, other studies of capital markets have pointed toward them being semi strong-form efficient. Studies by Firth (1976, 1979 and 1980) in the United Kingdom have compared the share prices existing after a takeover announcement with the bid offer. Firth found that the share prices were fully and instantaneously adjusted to their correct levels, thus concluding that the UK stock market was semi strong-form efficient. The market's ability to efficiently respond to a short term and widely publicized event such as a takeover announcement cannot necessarily be taken as indicative of a market efficient at pricing regarding more long term and amorphous factors however.
Other empirical evidence in support of the EMH comes from studies showing that the return of market averages exceeds the return of actively managed mutual funds. Thus, to the extent that markets are inefficient, the benefits realized by seizing upon the inefficiencies are outweighed by the internal fund costs involved in finding them, acting upon them, advertising etc. These findings gave inspiration to the formation of passively managed index funds.[1]
It may be that professional and other market participants who have discovered reliable trading rules or stratagems see no reason to divulge them to academic researchers. It might be that there is an information gap between the academics who study the markets and the professionals who work in them. Some observers point to seemingly inefficient features of the markets that can be exploited e.g seasonal tendencies and divergent returns to assets with various characteristics. E.g. factor analysis and studies of returns to different types of investment strategies suggest that some types of stocks may outperform the market long-term (e.g in the UK, the USA and Japan).
I particularly agree with this following section:
Skeptics of EMH argue that there exists a small number of investors who have outperformed the market over long periods of time, in a way which is difficult to attribute luck, including Peter Lynch, Warren Buffett, George Soros, and Bill Miller. These investors' strategies are to a large extent based on identifying markets where prices do not accurately reflect the available information, in direct contradiction to the efficient market hypothesis which explicitly implies that no such opportunities exist. Among the skeptics is Warren Buffett who has argued that the EMH is not correct, on one occasion wryly saying "I'd be a bum on the street with a tin cup if the markets were always efficient" and on another saying "The professors who taught Efficient Market Theory said that someone throwing darts at the stock tables could select stock portfolio having prospects just as good as one selected by the brightest, most hard-working securities analyst. Observing correctly that the market was frequently efficient, they went on to conclude incorrectly that it was always efficient."
Adherents to a stronger form of the EMH argue that the hypothesis does not preclude - indeed it predicts - the existence of unusually successful investors or funds occurring through chance. They also argue that presentation of anecdotal evidence of star-performers to cast doubt on the hypothesis is rife with survivorship bias.
However, importantly, in 1962 Warren Buffett wrote: "I present this data to indicate the Dow as an investment competitor is no pushover, and the great bulk of investment funds in the country are going to have difficulty in bettering, or... even matching, its performance. Our portfolio is very different from that of the Dow. Our method of operation is substantially different from that of mutual funds." [2]