Cash in on the Wisdom of the Crowd
There’s value in knowing what other investors are thinking.
That’s been a key principle of investing for generations.
Long ago, investors used all sorts of indicators to try to get a sense for the feelings of the crowd. They’d look at esoteric gauges of sentiment like put and call option volumes and more direct measures like sentiment surveys.
The most famous sentiment survey is the one that’s been published by the American Association of Individual Investors for the last 31 years.
The problem is that those sentiment indicators are only partially correct. Market indicators like the put/call ratio are a level removed from what investors are actually thinking.
They may have some general correspondence with sentiment, but they’re a little like assessing heart health by looking at a patient’s dental records — why do that when you can use an EKG machine?
There’s another problem.
Sentiment for individual names can be wildly different from what investors feel about the rest of the market.
Do you think Apple has the same investor sentiment as shares of Tesla or General Electric right now? Of course not. But the sentiment gauges most investors have access to don’t differentiate between them. They can’t.
Social media technologies have changed that. Suddenly, investors have minute-by-minute opinions of thousands — or even potentially millions — of market participants.
In the last decade, Twitter has also become a sort of watering hole for traders in particular. It’s the place where people go to talk about their latest buys. To celebrate their wins and lament their losers. Very often these days, Twitter is where news is reported first. All of those things make it an incredibly powerful place to measure sentiment.
Most importantly, it’s usable sentiment: direct feelings about a particular stock.
Last week, I told you about SERINA — the Sentiment Recognition and Interpretation Algorithm that my colleague and I have been working on for the last couple of years.
Every second, around 6,000 tweets are posted to Twitter. Obviously, a tiny fraction of those are about stocks. And an even smaller fraction are useful.
The system we’ve built is a direct feed from Twitter’s servers — but it’s designed to only capture meaningful tweets. The other 99.9% of Twitter noise is ignored.
The result is charts like these:
The chart above shows a sentiment spike in shares of beaten-down pot stock Aurora Cannabis (NYSE: ACB) last Friday.
Since then, shares have absolutely exploded higher, surging more than 99% in the last three trading sessions.
Of course, not all of the signals we see are quite as crazy.
(The typical smaller, more consistent winners are arguably even better over the long run!)
But it’s a good reminder that sentiment can be an incredibly powerful predictor of stock prices, especially in uncertain times.