People overestimate the probabilities of unlikely events. People overweight unlikely events in their decisions.
-Daniel Kahneman, Thinking Fast and Slow
I think about unlikely events all the time.
Coincidentally, I often get emails asking me the eternal question: "What would happen if...?"
What would happen if my broker goes bankrupt?
What would happen if the GBP has another flash crash?
What would happen if we have have another meltdown like 2008?
There's nothing wrong with worrying about those things. There's nothing wrong with being prepared.
Think about the quote at the top of the post. Daniel Kahneman's research into the nature of our own minds is some of the most brilliant research ever done. He tore down many of the misconceptions we carry around with us on a daily basis.
And, according to his research, we overweight the probabilities of unlikely events while making strong decisions on unlikely things.
In other words, we do what I've already done many times: we throw out a good possibility based on the idea that an unlikely event will ruin us.
For example, let's take my new Fair Value trading system that I'm now trading. It's my favorite way to trade (a counter-trend methodology with a taste of value investing), and it's also the method that's made me the biggest amount of money.
According to my research, and using a small trade size, the maximum drawdown for the entire portfolio came out to about $3,120 (again, small trade size).
So, just like clockwork, I started imagining trading this portfolio on a $6,000 account and went higher from there.
Hmm, a $3,120 max drawdown on a $6,000 account seems a little risky. Maybe $8,000 would be better. Or even $10,000...
This is where Thinking Fast and Slow comes in.
That $3,120 drawdown happened exactly once in almost 14 years. Once. I think that qualifies as an "unlikely event." Yet, I strongly weighted my decisions on that one unlikely event--just like Kahneman said.
So it got me thinking.
What if I tried to overcome my bias toward overweighting unlikely events? What would that look like?
Again using the Fair Value portfolio (and a small trade size), here's a list of the portfolio drawdowns greater than $1,000 since February, 2003:
You can see that the biggest drawdown occurred in 2008. There were also big ones in 2009 and 2015, and a medium one in 2011. But that's it.
From 2003-2016, drawdown reached the $2,900 mark only three times. In other words, if we committed to aggressively trading this system for just one year, there was a 78.5% chance you would not get a drawdown over $2,900.
Let's take it a step further.
What if we decided not to put weight on unlikely events and really wanted to go for it? What if we wanted to use the percentages and try a one-year experiment in an attempt to generate monster returns?
Here's one thing we could hypothetically do.
Looking at the drawdown amounts, we can see that the other three drawdown amounts are less than $1,500. Only four times in fourteen years did the max drawdown exceed $1,500.
So what would happen if we traded the portfolio on just a $1,600 account?
Well, in 2008, 2009, 2011, and 2015, we would have lost our whole hypothetical, experimental account. That's not fun.
But what about the other years?
If you take those four years out of the equation, we would not have margin-called in any of the other years from 2003-2015.
In fact, if we decided to do a one-year experiment and traded on a $1,500 account, here are the hypothetical returns:
In summary, if we hypothetically decided tomorrow to try a one-year experiment with this portfolio on a $1,500 account, there is 71.4% chance that we could make an average of 50%.
If we went to Vegas, what table game could we play that would give us a 71% chance of turning $1,000 into $1,500?
And how many times would we be willing to play that game if we lost the first time?
This is the type of thing Kahneman was talking about. I see the $3,120 drawdown and immediately run away from that unlikely event as fast as I can.
When we break it down, though, a 71% chance to win 50% is not a bad bet. And don't forget, with this type of "100%-style" system, we would be winning every series of trades in the years we didn't bust.
One last thing.
If we decided to not cut it so close, we could do a one-year experiment on a hypothetical $2,500 account. That takes one bad year of max drawdown off the table.
If we did that, we'd have a 78.5% chance to make an average of 30%. Would we play that game? How many times?
This is the type of thing I'm thinking about at the start of 2017.
Make no mistake: I'm not at all recommending throwing money around willy-nilly. Being safe is never a bad choice. This type of experiment may never be a good idea for your situation.
But it's time for me to at least consider what Kahneman said. Maybe we can think about it together.