I completely disagree. It is one thing if this fund's algorithm is based entirely off of data extracted from Twitter, that would be an utter mistake. However, if the algorithm is combining fundamental financial data, analyst's outlooks/forward looking statements, current trends/movements in the market, AND Twitter/media data then I think this could be quite successful. A huge reason why our capital markets are not perfect is because of emotion in the market (case in point Ford's strong run for the last year fueled by optimism and relatively strong performance and the pullback since qtr 4 earnings were released, and a thousand or more other examples). Effective investment algorithms must attempt to be comprehensive in taking into account factors that effect share prices and the public's sentiment is certainly one of those factors.
The positive trading sessions for companies with quality Super Bowl advertisements following the big game is a perfect example of this point. Check at this article:
http://www.kiplinger.com/columns/picks/archive/does-the-supe...
A 1% lift captured in ten days on a hundred million dollars is what annualized return? I'll let you do the math, but I'd take it.