For years, we have tracked the performance of a “Billionaire’s Club” of the world’s best investors. Through quarterly filings, we are able to see when these investors buy and sell and compare their results to how they might fare using TradeStops.
What we have found is that, in most cases, the performance of these top investors could be improved, sometimes dramatically improved, by using TradeStops to manage their portfolios. Our software helps make the best even better!
In the past, we only had internal data to support this conclusion (that our software can help make decent investors good, and good investors great). But now a groundbreaking research paper backs up our analysis in a different way.
This new academic research confirms that professional investors are good at buying stocks — but their selling decisions are often worse than random! In other words: All Wall Street pros might be better off with TradeStops. Let’s examine why.
In a nod to Daniel Kahneman, the research paper is titled “Selling Fast and Slow: Heuristics and Trading Performance of Institutional Investors.”
(A heuristic, by the way, is just a mental rule of thumb, like grabbing a coat if it looks like rain.)
The academics behind the study come from places like the Booth School of Business at the University of Chicago, the Department of Social and Decision Sciences at Carnegie Mellon University, and the Sloan School of Management at the Massachusetts Institute of Technology (MIT).
The unique thing about the study is who it was conducted on.
There was already a great deal of research on the topic of investor habits and biases. But almost all of it is based on the habits and patterns of the small investor, with data pulled from the major retail investment brokerages.
For this study, the researchers looked to the institutional “big leagues” — professional money managers with an average portfolio size of $573 million!
The study took data from 783 institutional portfolios over a 16-year period, from 2000 to 2016, for a total of more than 4.4 million buy and sell transactions.
The study did another unique thing in creating “counterfactual” results, to test whether a buying or selling pattern was better than random.
They did this by comparing the results of each transaction with a hypothetical “random” alternative transaction, taken from another position that was not traded in the portfolio that day.
What the study found was that, on balance, professional investors do in fact show skill in their decision-making process for when to buy stocks. The stock buys on average were better than random and beat their performance benchmarks.
The sell decisions, though, were significantly worse than random — to the point where throwing a dart or flipping a coin would have been a better strategy in terms of net annual performance.
Interestingly, the sells based on information from earnings reports were performance-enhancing. It was the stock sales not associated with news that were worse than random (where a coin flip would have been better). You’ll see the reason why shortly.
The researchers found that the stock sales were worse than random because of a common heuristic, or mental rule of thumb. This rule of thumb has been observed in small investors, but now we know professional investors fall for it, too.
The rule of thumb in question — which again, made results worse than random — was a habit of selling stocks as a result of “extreme event” levels of price movement.
There were two conditions where professional investors were more likely to sell a stock in a manner that would hurt their total performance later:
- If the stock was down a lot — a big loser — there was strong temptation to sell.
- If the stock was up a lot — a big winner — there was strong temptation, too.
The researchers called this a “U-pattern”: Stocks that had fallen by a large amount were sold more often at one tip of the “U,” and stocks that had risen by a large amount at the other. The bottom of the U was stocks in a normal range.
Why would this hurt performance? For two reasons.
First, because if a stock has already fallen a great deal, the loss already exists, and selling too late just increases the odds of missing a rebound. The time to sell a poor investment is before it morphs into a disaster, not after.
And second, because if a stock has strongly risen, the forces that created a powerful uptrend in that stock are likely to drive the trend further. Selling something just because it is up creates a habit of missing future trend gains — sometimes incredibly spectacular trend gains, the kind that make a big difference over time.
There are so many additional things that make this study fascinating. For example, they discovered this “extreme event” U-pattern bias was only present in the sell decisions, not the buys. That is why the buying performance was fine.
When it came to removing stocks from the portfolio, professional money managers were thus using rules of thumb and the influence of past price movement in ways that hurt them. When it came to new buys, they didn’t do that.
Based on data, the researchers theorized that, for the professional money manager, the buying process and the selling process are different animals in a psychology sense. This makes sense considering the different scenarios for buying versus selling:
- When professional money managers are buying a stock, they tend to be paying close attention and using their well-honed analytical skills. All of their analysis and judgment is brought to bear on the decision to buy something.
- When those same managers are selling a stock, they may be raising cash in the portfolio because they want to buy something else, or trimming back the portfolio on the whole. They aren’t paying as much attention!
The “not paying attention” theory is backed up by the better performance of stock sells related to earnings reports.
The worse-than-random result didn’t apply to stock sells related to earnings developments. When a professional money manager sells because of an earnings report, they are applying judgment to the stock itself. They are responding to data, not selling for some unrelated reason, and once again (drum roll please) paying more attention.
It’s the “selling for some unrelated reason” that winds up being the killer. The money manager pays less attention, and then goes with a built-in cognitive bias to kick out stocks that are either up a lot or down a lot, and this hurts them.
The researchers did all kinds of analysis on this extreme event U-pattern and the ways that it hurt.
They confirmed that the U-pattern persisted across all kinds of buckets and timeframes. They showed that the more aggressively a money manager displayed the U-pattern, the worse their relative performance tended to be. They also showed that, for the track record of an individual money manager over time, performance when the U-pattern was frequent was worse than when it was infrequent.
And they showed that money managers tended to use the pattern more often in tough market periods, demonstrating a tendency to rely too much on rules of thumb under stress. (The greater the stress, the more tempting it becomes to “go with your gut” and the harder it is to think.)
What can we take away from this study? A few important things.
First, the whale-sized money management pros have been taken down a peg. There now exists academic proof that the biggest Wall Street pros, the ones who run hundreds of millions, make the same kind of mistakes small investors do.
Second, the Efficient Market Hypothesis (EMH) takes a serious hit from these findings. If the professionals show irrational tendencies, too, it is impossible for markets to be “perfect” or to perfectly self-correct, because the perfectly rational investment professional is a myth.
Third, the habit of selling “extreme event” U-pattern stocks in a portfolio — kicking out a name just because it is up a lot or down a lot — is a tempting thing to do, for small and large investors alike, that hurts performance on balance.
And fourth, TradeStops can directly safeguard against this detrimental tendency, by helping investors remain detached and rational in their sell decisions.
That is because, if you use risk levels and trailing stop points as determined by the TradeStops software, you won’t have “extreme event” stocks on the downside — because you will have sold poorly performing positions before they fell too far.
And, if you maintain the discipline of sticking with solid upside trends as identified by healthy green zone positions in your TradeStops portfolio, you will have greater psychological fortitude in staying with those trends for years.
Richard Smith, Ph.D.
CEO and Founder of TradeSmith