Mathematically Making Money (Guaranteed)

Mathematically Making Money (Guaranteed)

May 3, 2017

There’s a trading firm out there that makes money every day.

Did you know that?

It’s true.

Virtu is a quasi-high frequency trading (HFT) firm that disclosed not long ago that it only had one losing day out of a thousand. You read that right.

It made money 999 days out of 1,000 (actually, it did better than that).

Most people’s reaction to hearing that has been to cry Madoff. A lot of people think that if someone or something makes money every day, it has to be a ponzi scheme built on lies. Or, at the very least, a firm that makes money every day has to be doing something extremely shady, if not outright illegal.

But, according to UC Santa Cruz professor Gregory Laughlin, the process of making money every day doesn’t have to be illegal at all. In fact, for Virtu, it’s inevitable and quite explainable.

Laughlin cites the Law of Large Numbers as the reason for Virtu’s success. He did a deep dive into the company’s business model and learned that Virtu makes millions of trades every day. And they don’t have super-secret trading methods that game the system, they just trade a lot because they act as a market-maker. Even though they have an incredible record of making money on a daily basis, they only win about 51% of the time. Nothing extraordinary about that.

It all comes down to the sheer number of trades the company takes. It’s like they take decades of trades in a few hours. They go through the same ups and downs we do, except they do it in hyper drive.

Think about it: How long would it take for us to make two million trades? If one of our systems trades 500 times a year (10 trades a week), it would take us about 4,000 years to get to two million. If we could let 4,000 years of trading play out, do you think we could be profitable? Of course we could.

And isn’t that a valuable lesson? If we’re sure we have an edge, then it’s empowering to know that all we have to do is trade the crap out of that system to make it to the finish line a winner.

To put it numerically, if our system can win at least 51% of the time with a 1:1 risk to reward, we are guaranteed to win– if we take enough trades.

That works for other winning percentages, too.

To repeat, we know that if our system wins 51% of the time with a 1:1 risk to reward, we have an edge and we’re guaranteed to succeed. Here’s a few more winning scenarios:

  • A 60% win percentage with a 1 : 1.2 ratio (risking $120 to win $100)
  • A 75% win percentage with a 1 : 2.1 ratio (risking $210 to win $100)
  • A 90% win percentage with a 1 : 8 ratio (risking $800 to win $100)

Regardless of our risk to reward ratio, the math tells us that we will be profitable in the long run if we stick to our system.

Are there any real-life example of a business sticking to a small edge and watching it play out over a large amount of examples? Guess how this was paid for?

This is the exact methodology used by casinos. They have a small edge and all they need is for people to keep playing. They know that after, say, two million bets, they’re going to be really profitable. It’s science.

So how does this affect our trading?

Using this way of thinking, all we have to do is look at the numbers for one of our trading systems, and then let the Law of Large Numbers take us to the promised land.

For example, I did a little research on my new robot (The Heron). It’s a daytrading system that takes over 2,200 trades over a 13-year period.

I started from the very first trade and tried to see if the Virtu way of thinking worked in that system. Here’s what I found.

Over every series of 235 trades, the robot was profitable. Put another way, if I started trading it back in 2003 and analyzed my profit and loss after every 235 trades, I would have been profitable every time, sometimes wildly so.

Of course, we can’t guarantee ourselves a daily profit without trading thousands and/or million times every day.

But it we simply take our personal timeline and stretch it out a bit, we can have the exact same success, just on a different scale. Over a month or several months or maybe a year (however long it takes to get the certain number of trades we need), we can know that we’ll make money.

The only way we wouldn’t be profitable after a certain period of time is if we quit.

As long as we have a tested system with a mathematical edge, success is guaranteed.

Thank you, Virtu, for making that so clear.

Now all we have to do is hang around long enough to see the math pull us through.