Some traders sabotage their trading through biases and information overload, explains Dr. Gary Dayton, who cites recent research proving that more data does not mean better results.
There are a lot of different things we can do to improve our trading, but there are also things we can do to sabotage our trading as well. One of those things is confirmation bias.
We’re going to talk to Dr. Gary Dayton today about that term and what it means to us. So, Dr. Dayton, what is confirmation bias?
Confirmation bias comes out of behavioral finance. It’s essentially taking information to confirm what we already know. Now, it feels good because it validates what we already know, but it might not necessarily be the best thing to do.
In a classic sense in trading, confirmation bias would occur, let’s say for example we use the MACD to signal our trades. So, we get a trade signal. That’s all fine, but before we take that trade, we say, “Well, let’s just check the Stochastic or the RSI,” things that would kind of support the MACD perhaps, but really don’t add to the quality level of our decision making.
There was a really interesting study done on horse-race betters—paramutuel betters—that will help illustrate this point. They took a group of professional horse race betters and they gave them five pieces of relevant information. I’m not a horse better, but the age of the horse, the record, the track conditions, those sorts of things; and they made a betting decision on that. That was recorded.
Then, in the next trial, they doubled the amount of information and they gave them ten pieces of information. Same thing, they made their bets.
They doubled it again on the third trial to 20 pieces. Then, on the fourth trial, we now have eight times as much information as we had in the first trial, 40 pieces of information.
Well, they also in this study measured confidence, and confidence, along with the amount of information, confidence soared. But when they looked at performance, it was flat. They had the same outcomes, the same performance, on 40 pieces of information as they had on five pieces of information.
So, the message is pretty clear: when we are adding information to what we already know, and these professional paramutuel betters were able to do it with five pieces, they didn’t need eight times as much, but their confidence did soar, and so they become potentially unsoundly overconfident.
In trading, that can lead to increased leverage, overtrading, and other erratic trading behaviors.
What we really want to do is not look to add to information that we already know, but to look for disconfirming information.
How would we do that in trading?
Let’s say we’re in a rally, and we want to assess the strength of that rally. Well, it’s going to come on a pullback, and if that pullback shows light volume—no real supply coming into that market—we’re disconfirming the presence of supply, and therefore, confirming the strength of the rally.
So, that’s a better way to approach it.
You mentioned it leading to too much confidence with too much information. I can also imagine with traders that too many indicators could lead to confusion and indecision.
Sure. We call that “analysis paralysis.” If we have a number of clocks all chiming off, which one is right? It’s the same kind of problem that can arise.
Related Reading:
