“Buy the rumor, sell the news”. It’s a proverb we’ve all heard. But as far as proverbs go, they only get us as far as conventional wisdom. To be a winner in the markets today often takes something else; something unconventional. The problem of news in trading is a relevant one. The holy grail is getting early information. For the vast majority of traders, though, getting news as early as the inner circles do is all but impossible. With just one phone call from the 51st floor, someone will beat you to it, legally or not.
However, that’s not even the main problem with trading on news. The bigger issue is that markets sometimes seem to react to news stories, and sometimes not. I’m a trader who cut my teeth the hard way. While I was in college, I began trading with money that I scraped from wherever I could, or was given by my family. Eventually, I was able to trade my way through traveling in Europe.
I lost a lot before I learned how to make that happen. When I was starting out, I would buy a stock on a good earnings report, only to see the price continue edging down regardless. Meanwhile, I’d see another company announce that it won a legal battle, and its share price would do a victory walk.
There didn’t seem to be any rhyme or reason to how the markets would absorb news. I eventually had to train myself to pay less attention to the news and shift my focus towards hardcore technical and fundamental analysis. Counting on the market reacting to news proved to be too fickle. Many traders feel the same way, yet most of us wouldn’t dare trade without a news feed going. After all, one unexpected story can still interrupt a perfectly good trend.
Cut to a couple of years later. I had a solid trading strategy. After hiking the Austrian Alps by day, I would trade in the evening from a Tyrolean-style lodge with a balcony overlooking the mountains. One evening, I felt my phone buzz in my pocket, and a news alert popped up. “Positive clinical trial results for Eli Lilly Migraine Pill”, it said. The stock was rising, eventually 1.3% for the day. At this point, I saw an opportunity. Not to buy Eli Lilly shares, but to answer a question once and for all.
Why do stocks sometimes seem to respond to news, and sometimes not? The market took this particular story to heart, but I’ve seen stories of a similar nature elicit no apparent reaction. The easy explanation is that sometimes the news is already priced in. Yet, I felt there was more to it. I decided to answer the question with a means I truly believed in: data science.
After analyzing years of data on several securities, I came to a conclusion. Market prices aren’t driven solely by news, but they go in and out of news-driven periods. When they are in fact news-driven, you can use headlines to your advantage. During non-news-driven periods, you can identify other “background factors” moving the price. This became one premise of my 2018 paper published in the peer-reviewed scientific publication, the Journal of Big Data.
There is a clear cycle to when prices are news-driven, and when they aren’t. Knowing where we are in that cycle can allow you to capitalize on news-initiated trends. There are a few pieces that go into understanding news and price: sentiment, timing, and historical relationship. My aim is to take years of nearly obsessive research and hand you a few takeaways right here. Bear in mind, these observations are based not only on my own research but also on extensive reading of the body of research literature on the topic.
The first element is sentiment. Many studies, including a 2019 study by Vanstone et. al in the journal “Applied Intelligence”, show that there is a correlation between news sentiment and prices. However, knowing this alone is not enough to get ahead as a trader. Timing is also critical. A 2012 study published in the Accounting Review Journal by G. Mujtaba Mian et al. proved something interesting. When sentiment is already negative about a stock, any more negative news has a strong chance of bringing the price down even further.
It’s not just about current sentiment, but how news stories resonate with that sentiment. It’s the timing of articles with respect to the existing sentiment that’s important. Lastly, the historical relationship between news sentiment and price has to be broken down. How often did a negative article cause a dip in the price? How often did a positive article cause a rise in price? To quantify this over a defined period tells us how responsive a stock is to the news.
Sentiment, timing, and historical correlations can demystify the new-price relationship. I’ve used them in my own research to figure out when to let the news influence my trading decisions, alongside the technicals. It’s also been useful for longer-term investments, namely determining if the time is right to sell a stock I’ve been holding simply because of a negative story that came out. This has taken lots of uncertainty out of trading, and boosted my ROI. However, it takes considerable time to do this research oneself. That’s why I created TripleHint.
TripleHint’s online software allows you to see exactly how responsive any stock is to news. It allows you to see whether today’s publications are confirming the market’s sentiment biases or not. It will organize current news for by sentiment. It will also tell you how likely the price of any stock is to move based on any given article. The analysis timeframes are adjustable to suit your trading style.
TripleHint launches this fall and will also support currencies and commodities. I’d like to personally invite you to check out the software demo here: https://www.triplehint.com/preorder/
About the author
Lucas Jacaruso has traded stocks, commodities, and currencies for most of his adult life and believes today is the most exciting time ever for data science and predictive analytics.
"Technology shouldn't replace human decision making in trading or otherwise, but it can complement our intuition with better information."
Lucas attended the University of Southern California and the University of Graz, Austria. In 2018 Lucas had his original trend-prediction research published in the peer-reviewed Journal of Big Data. His paper explored a novel approach to forecasting, applied to blue chip stock prices and humanitarian issues.