SCOTT WELSH TRADING BLOG

The Holy Grail of Backtesting?

The Holy Grail of Backtesting?

Sept. 11th, 2019

The difference between our current life and a financially-independent life resides in our trading system.

If the system we purchased (or made ourselves) performs like we want it to, everything could change.

But our trading system has to work.

And how do we know it’s going to work?

Backtesting.

Backtesting allows us to research our system to see if it’s profitable and also to understand its expectation. If historically our system produces 20% per year on average, then we can make plans, adjust trade sizes, and save enough money to change our lives.

Of course, some people don’t believe in backtesting. That’s fine. This conversation isn’t for you.

But my question is: if we don’t believe in backtesting, what do we believe in? Should we just trade our life’s savings on a hunch? An unverified tip from someone else? Blind faith in an Index Fund?

If trading without testing is fine for you, great. But none of those options are good enough for me.

Which leads us back to backtesting. If backtesting is the key, how do we do it? What should we trust?

Now it gets tricky, because there are a lot of ways to backtest. First, we have to decide what we’re testing for.

Some people test for maximum profit. That’s what we want, right? So that’s what they’re looking for in their testing.

Some test for the least amount of drawdown possible. Rule #1 of trading? Don’t lose money. That also makes sense.

Some test for Profit Factor. That ratio is the amount of net profit over the amount of net loss. That’s a good ratio and also a good idea.

Some test for Return On Account (ROA). This is the basically the amount of profit you’d get after your worst drawdown. This is another good thing to look at.

Some test for other things.

So, which one is best? Um…

It depends?

Arguments can be made that any of those are superior to any of the others. And those arguments won’t get us anywhere.

They all might work and they all might fail.

So, that’s not the Holy Grail we’re looking for.

What is?

Walk-forward optimizing.

Walk-forward optimizing is the most advanced testing feature the general public can use. It’s very complex and very powerful.

And some of the best algo traders in the world swear by it.

Why is it so great?

Because the biggest mistake we make is not using enough data. We don’t go back in time far enough and then our system fails as soon as we go live because we didn’t have enough trades and didn’t see enough market cycles. (A very famous trader says he needs to see at least 1,800 trades before he even considers a backtest).

But even worse, we don’t have any idea of what it would do in the live market ahead of time. We test our data and then start trading.

Walk-forward optimizing is supposed to fix all that.

It gives us a super-advanced way to use super-advanced software to give us the most realistic expectation of what our system can do.

In our next email, we’ll describe exactly what it is and see it in action in a walk-forward test.

Talk to you soon.

 

Get the robots here.

To get on the email list, go here.

To see my weekly YouTube video, go here.

 

Disclaimer:
It should not be assumed that the methods, techniques, or indicators presented in these products will be profitable or that they will not result in losses. Past results are not necessarily indicative of future results. Examples presented on these sites are for educational purposes only. These set-ups are not solicitations of any order to buy or sell. The authors, the publisher, and all affiliates assume no responsibility for your trading results. There is a high degree of risk in trading.

HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN INHERENT LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING. ALSO, SINCE THE TRADES HAVE NOT ACTUALLY BEEN EXECUTED, THE RESULTS MAY HAVE UNDER- OR OVER-COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEY ARE DESIGNED WITH THE BENEFIT OF HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN.