The Science Of Staying In The Game

Regardless of the circumstances, when we win, life is perfect and we are the greatest.

For example, let's hypothetically say an Ohio-based major college football coach is actually a flesh-eating ogre. As long as he wins, nobody cares.

Same thing for trading. It doesn't matter if our trading system is too risky, and we are are extremely undisciplined. If we go on a winning streak, everything is peachy.

The problems arise when we lose.

Ninety percent of the emails I get come after the robots lose a trade, and it was the same when I coached tennis. I only heard from the parents after their kid lost.

I understand, though. Losing is scary. Losing is depressing. Losing makes you feel like the sun is never going to rise, and you're being chased by a T-Rex while riding in a rickety jeep.

But dealing with the losing is the real key to success. If you can meet with triumph and disaster and treat those two imposters just the same, you can make lots and lots of money. If you give up when you lose, you get nothing.

So what's a good way to handle losing?

Let's take a look at one of my robots (the USDJPY).

The USDJPY is a daytrading robot that trades a few times per week. It has a nice sample size of trades (over 1,200) and has a nice win percentage (about 79%).

Most of the time with a robot like this, you'll be celebrating a winning trade.

But not always.

After a nice winning streak in September, this robot finally took a loss last week (October, 2016). That loss wiped out a lot of the gains I had made while I was winning.

I had two options:

  1. Find the nearest high bridge and take a running, head-first leap.
  2. Look a the data and deal with it.

I chose option #2 (as far as you know).

One thing I like about this robot is that is has multiple ways to stop out of a losing trade. It has a hard "emergency" stop, and also a "situational" stop.

An emergency stop is a large stoploss that is the last line of defense when things go badly.

A situational stop is one where time just runs out on a trade. It hasn't hit the target and it hasn't stopped out. It's just time for the trade to end because it hasn't done a whole lot.

With my robot, most of the losses are situational and easier to deal with. However, the loss last week wasn't situational. It was the whole enchilada.

So a good question to ask is: How often does this happen? Or, how many times over the past thirteen years has it hit the emergency stop?

If we look at the data, we see that there have been 267 losing trades in the past thirteen and a half years. Of those 267 losers, 55 have been big emergency stop-outs (or close to it).

Here's the breakdown of how many big losses this robot has taken each year from 2003-September, 2016 (keep in mind that this robot has posted no losing years, no matter how many big losses it took):

  • 2003 - 6 big losses
  • 2004 - 5 big losses
  • 2005 - 7 big losses
  • 2006 - 4 big losses
  • 2007 - 6 big losses
  • 2008 - 5 big losses
  • 2009 - 2 big losses
  • 2010 - 3 big losses
  • 2011 - 5 big losses
  • 2012 - 2 big losses
  • 2013 - 1 big loss
  • 2014 - 4 big losses
  • 2015 - 4 big losses
  • 2016 - 0 big losses

That's an average of 4 big stop-outs per year. And, yes, that 2016 number is correct: Through September of this year, this robot had taken zero big losses at the full stop-out amount.

Instead of bridge-jumping, I probably should have anticipated this loss, since I hadn't had one all year!

But let's go further and look at the streaks.

Over the past thirteen years, this robot has had losing streaks. How many and how long were they?

The longest losing streak this robot has ever had is 4. It's lost 4 trades in a row just one time.

The next longest losing streak is 3, and it's done that six times (this occurs about once every other year).

Furthermore, it's lost 2 trades in a row thirty-seven times (this happens about 3 times a year), and it's lost 1 trade in a row one hundred and seventy-one times (a little more than once a month).

That data gives me a nice picture of what kind of losing I can expect.

On the flip side, it's had a ton of winning streaks. Some highlights are:

  • This robot has won 18 trades in a row three times (once every 4.5 years)
  • This robot has won 12 trades in a row five times (once every 3 years)
  • And this robot has won 7 trades in a row fourteen times (more than once every year)

If you know what to expect from your trading system, it's much easier to handle the losing.

Or as the great philosopher/Roadhouse bouncer Dalton once said, "Those who go looking for trouble are no match for those who are ready for it."

So make sure you know your trading system intimately. Understanding exactly how much losing you can expect will give you the strength to handle the bad times.

And stay in the game.