In the stock market, there are countless strategies for making a buck. Some investors like to focus on the fundamentals of the companies they invest in — poring over financial statements to figure out which firms are over- or under-valued. Others invest based on trends or macroeconomic events, like whether the Fed is raising or lowering interest rates.
These may be effective approaches, but the greatest trading strategy of all — were it possible — would be to simply learn how much a particular asset will, in the near future, be valued by everybody else in the market. After all, all the hard data in the world cannot compel a seller or buyer to give you the price you want.
That’s why, at the end of the day, stock markets are about mass psychology as much as anything else. And with the proliferation of the internet, it has never been easier to tap into moods and feelings of the masses. This is what researchers Tobias Preis, Helen Moat, and Eugene Stanely had in mind when they set out to prove that you can make money in the stock market just by following what people are searching for on Google.
(MORE: How Does One Fake Tweet Cause a Stock Market Crash?)
In a study published yesterday in the journal Nature, these researchers showed that from 2004 through 2011, by making trades purely based on the prevalance of specific search terms, they could earn outsized returns. The most lucrative search term these researchers found was, unsurprisingly, “debt.” The researchers found that if they had sold a Dow Jones Industrial Index fund during times when the search term “debt” spiked, and consistently did this over the 7-year-period between 2004 and 2011, they would have earned a healthy 326% return. By contrast, had they simply bought a broad stock market index fund in 2004 and held it until 2011, they would have earned just 16%.
Some of the other terms that would have yielded hefty returns were a little less intuitive, like “color,” “stocks,” and, oddly enough, “restaurant.” The fact that these terms also yielded strong results shows why this paper is less revelatory than it may seem at first blush. That people were inquiring about debt in the years leading up to what was basically a crisis of too much debt isn’t all that surprising. Of course, you would have had to have known in 2004 that the term “debt” was one that you should track in order for this trade to have worked out. And plenty of investors did know that debt was a big problem, and made a ton of money off of that knowledge, even without the help of Google.
Furthermore, traders are already taking advantage of similar strategies to the one outlined in this paper. Some high frequency trading firms are employing algorithims that comb through news sources to identify combinations of terms that may bode well or ill for stock returns. In fact, some speculated that these sorts of programs were partially responsible for Tuesday’s mini flash crash, presumably instigated by a fake tweet on the Associated Press’ Twitter feed, which was momentarily hacked.
The big take-away from this study is that there’s much that market participants can glean from massive amounts of information flowing around the Internet each day. And rest assured, brilliant minds on Wall Street and across the world are figuring out how to do this as we speak. At the end of the day, however, the stock market is just one big game of trying to predict the future. And no amount of information in the present will make that an exact science.
(MORE: High Frequency Trading: Wall Street’s Doomsday Machine?)