Those trending topics on the left-hand sidebar of your Twitter account may be more than aggregations of crass sexual stereotypes or the latest false rumor of a celebrity death. According to a new study out of the University of California, aside from giving you a glimpse of humanity’s reality TV-poisoned id, Twitter trends may be predictive of stock market performance.

Vagelis Hristidis, an associate professor at the university, along with one of his graduate students and three members of Yahoo! Research Barcelona, have created a trading strategy model that examined the correlation between Twitter topic activity and stock market prices and trade volume. Focusing on the volume of tweets regarding a topics associated with companies that are publicly traded, Hristidis and his research team developed a model based on Twitter data and compared it with several other models used to predict stock market performance, including the Dow Jones Industrial Average.

While they expected that the trade volume would correlate with the number of tweets, they discovered that the number of trades “is slightly more correlated with the number of what they call ‘connected components.'”

For example, it’d pollute the integrity of the data if the research team tracked all tweets containing the phrase “apple.” Aside from being the preeminent tech company in the world, you may remember that “apple” started out as a fruit. The team needed to separate tweets about the fruit from tweets about the tech company, their products, and tweets about the company’s financial performance, so the fruity tweets were excluded from the data.

To test the different models, the research team simulated investments between March 1 and June 30 of last year and compared the performance of the different strategies. Within that window, the Dow Jones Industrial Average fell 4.2% whereas the Twitter model only lost an average of 2.4%, suggesting that the Twitter model may outperform the Dow Jones.

While its tempting to tease out causality from the correlation, Hristidis acknowledges that, should the market turn bullish, the Twitter-turned-trading strategy model might not produce the same results. Speaking with Forbes about the relationship between Twitter conversations and a stock’s price, Hristidis said, “To be honest, I’m not sure why we found this correlation” but it could be that people tend to dwell on a bad piece of news whereas good news is not as fun to meditate on. And lord knows people love to complain on the internet.

The relationship between Twitter conversations and stock price’s became particularly striking with an example from earlier this week that includes two odd components: Republican presidential candidate Mitt Romney’s spokesman, Eric Fehrstrom, and the beloved magnet doodler of children of multiple generations, the Etch a Sketch.

Recall that on March 21 while speaking on CNN, Fehrstrom, whether tongue-in-cheek or not, compared his boss to an Etch a Sketch. Subsequently, people caught wind of this gaffe and related their thoughts via Twitter, boosting the frequency of posts including the phrase “etch a sketch” on Wednesday after Fehstrom’s comment.

What’s more, on the day following the gaffe, the stock value of Etch a Sketch’s manufacturer, Ohio, skyrocketed 200%.

So was this Team Hristidis’ model at work again? It’s hard to say since the quality of the tweets would be expected to not have been related to the stock value of Etch a Sketch’s manufacturer, Ohio Art Company. “That’s a great point,” Hristidis told WPN when asked about Ohio’s stock value, “but if many of the tweets look non-financial, we would have to refine our rules or drop the stock from our group of studied stocks.”

So the take-away from all of this speculation: Twitter may not be the new yardstick by which Wall Street speculators gauge the performance of a company’s stocks, but it could certainly add to the inventory of tools investors consider in their trading strategy.

Got any meditations or dismissals you’d like to add to this story? Feel free to share them below in the comments.