Get The Full Series in PDF
Get the entire 10-part series on Charlie Munger in PDF. Save it to your desktop, read it on your tablet, or email to your colleagues.
Does News Sentiment Data Help Discretionary Traders?
I read it many years ago when I first got into finance and recently I reread it but with a different objective in mind. Now I work in fintech and I wanted to see if I could gain any fresh insights as to how discretionary traders might use our news sentiment data, scanned from millions of news articles online.
I have already covered possible applications in macro markets here. In this article, I wanted to confine myself to cases drawn from the second half of the book, in which Schwager interviews traders of ‘Mainly Equities’.
1. Not a Trade to Be Xeroxed
The Market Wizard William O’Neil highlights the error of relying too heavily on one proven metric such as the P/E ratio. He relates how when he was younger he “bought Northrop at four times earnings” thinking it could only go higher but then having to watch “in disbelief as the stock eventually declined to two times earnings.”
In another anecdote he relates how someone once advocated shorting Xerox because it was overpriced.
“I still remember in 1962 when an investor barged into my friend’s brokerage office, declaring in a loud voice that Xerox was drastically overpriced because it was selling at fifty times earnings. He went short at $88. Xerox eventually went to $1,300, adjusting for stock split.”
Could news sentiment data could have helped avoid these mistakes? Perhaps, according to research by Empirical Research Partners (ERP). In one study about value stocks, ERP showed that adding a news sentiment overlay could help screen for higher success candidates from a basket of overpriced prospects - and vice versa for undervalued.
The study showed stocks with a high P/E ratio tended to have a higher chance of going down if they were also accompanied by a period of negative news sentiment.
The study recommended using sentiment as a timing tool to ‘pull the trigger’, whether going short or long.
2. A Good Basis For a Trade
Whether and when prices will break out of a long-term basing pattern has been an age old question for traders.
Buy too soon and the price could go back down, stopping you out; wait too long and you could miss the train higher.
“The idea is to buy when there is the least probability of a loss.” Says the same Wizard as above, William O’Neil.
Could sentiment data help traders time such a breakout?
It seems likely: research by Empirical Research Partners (ERP) shows that sentiment precedes momentum, suggesting a breakout in sentiment would precede a breakout in price, and traders looking to trade the breakout from a basing pattern should wait for a breakout in sentiment first.
3. The Confident Contrarian
Market Wizard Michael Steinhardt uses a concept called ‘variant perception’ to describe his trading style. This is where he develops perceptions that are at variance with the general market view, and trades them until he feels they are no longer relevant. The question is, could news sentiment add any value to his trading processes?
It seems possible. One drawback of his style is that he tends to enter trades too early. Below is an example to illustrate the point.
“We have been short Genetech for a year and a half. There was a period of months and months when we lost a lot of money in that position. But I stayed short because I continued to have a variant perception about the future of their drug TPA.. It is our perception that, in a year or two, TPA will be a minor drug that will be supplanted by more effective drugs that also cost substantially less.”
In the above example, he could have harnessed news sentiment by waiting for sentiment to dip into negative territory before shorting the stock.
Unfortunately, despite being able to analyze his mistakes Steinhardt seems resistant to change.
“There is a general view you shouldn’t short a stock until it has already peaked and started down - that you shouldn’t go short until the stock is already reflecting problems that are evident for all to see. In some sense, I can understand that. Maybe that is a superficially safer way to short stocks and you can sleep more comfortably using that approach. However, I have never done it that way. My attitude has always been that to make money in the markets you have to be willing to get in the way of danger.”
Tough talk, but does it make good trading sense?
4. A 1980s ESG Controversy
Environmental, Social, and Governance (ESG) investing is not just a 21st-century fad, it also crops up in Market Wizards, only in this case with a tobacco company.
Here, Wizard Micheal Steinhardt talks about how he has shorted a tobacco company undergoing a class action for contributing to a smoker’s death.
“I went short a tobacco stock about a month ago. My reasoning was that if the plaintiffs won the case the stocks would go down a lot, but if the plaintiffs lost the stock wouldn’t go up by much…”
What is interesting is that this reflects a phenomenon discovered by researchers at Monash University in Australia in a recent ESG study using RavenPack news sentiment data.
The researchers found that most stocks tend to experience market overreactions after the release of negative ESG news about them. After the overreaction, the majority then tend to drift higher over the medium-term.
The advice is to trade the drift higher, yet Steinhardt’s strategy is to trade the overreaction, so he is potentially missing out.
Later in the interview, he says he is planning to cover his short “on the news”.
“So once the news is out, the game is over for you?” Asks Schwager.
“Right. That was the only reason I was short.”
5. Exploding Butterflies, an Optional Extra?
The news is the main cause of volatility in markets and some traders like Market Wizard Tony Saliba, like to trade volatility regardless of the direction. They do this using complex options trades. Given the close relationship with the news it would seem our data would have an obvious use-case in options trading.
Saliba uses two main types of options trades. The first is called the Butterfly, and it makes money when the market is flat. The second - his own invention - he has dubbed an Explosion and makes money when the volatility increases.
The question is, how could news sentiment data help in these sorts of trades, if at all?
The answer is in a research white paper called, “Abnormal Media Attention Impacts Stock Returns,” which shows how our news analytics, specifically the data measuring the volume of news about stocks, could be used to actually predict volatility, enabling the anticipation of changing market conditions.
“We demonstrate that our measure of abnormal media attention (i.e. Event Buzz) assists in the selection of stocks that are likely to experience strong price movements, resulting in an improvement of average per-trade returns.” Says our Chief Data Scientist, Peter Hafez.
What goes for stocks also goes for other asset classes too, suggesting Event Buzz may have applications for options traders in many different markets.