#2: Early backtesting to build a trading strategy

Tradebot 3000
Mar 1, 2018 · 3 min read

Getting some early testing done really helps (and it’s fun, honest)

I wasn’t looking for a ‘system’

When I started learning about trading, I instinctively disliked the idea of a ‘system’. For me it meant the “do this and make millions!” approach that tried to bypass actually understanding what you’re doing in the markets.

Thanks a lot to the Two Blokes Trading podcast, I started to appreciate that people had a systematic approach to spotting and taking trades. Although I still leaned more towards the idea of understanding versus any kind of automation, I realised that this would be important to testing and verifying what I was doing.

Also, I’ve spent a lot of ‘junk miles’ in front of live charts, taking the odd demo trade but not really learning much very quickly.

‘Pre-backtesting’ to build the strategy

Another big Two Blokes theme is backtesting, a concept that I hadn’t picked up in my other research. It sounded quite dull, but they pushed it so hard that I had to take notice!

Basically this means taking your strategy and running it through past charts to see how it performs. Thanks to the Trading View bar replay feature, this is actually really fun (if you like that sort of thing).

The problem was that I didn’t have anything to backtest as such, so I decided to do some very quick sort-of-backtesting to try to define one.

Testing price action strategy on AUSUSD daily

Taking the principles from the Urban Forex Mastering Price Action course and a little from Learn to Trade the Market (mentioned in my first blog) I picked AUDUSD and rewound the chart to the start of 2015.

Thanks to a bit of research leading me to Hugh Kimura’s free resources on Trading Heroes, I started putting some of my basic principles down on paper using his template.

I decided to start with a system that:

  • trades the daily chart (since this will fit my lifestyle when I go live)
  • follows the momentum from the controlling force in the market (buyer/seller)
  • looks to trade breaches through key levels in the direction of the ‘trend’
  • chooses a 2:1 or 2.5:1 risk/reward ratio based on market opportunity, but never less that that
  • places limit orders for entry and doesn’t move stop loss
AUDUSD, Daily, 2015

Results and lessons

As you can see from the chart, I didn’t make many trades. I actually entered 7 trades over the course of this chart, and the test took me about 2 hours.

The overall result was +560 pips, with 4 winning trades, 2 losing and 1 which ran away without hitting my entry.

The main thing was how useful this was in entering a lot of trades over the course of an evening and quickly finding out some early strengths and weaknesses. I would recommend this to anyone.

Some quick lessons:

  • I tend to enter one candle too early, or with weak confirmation signals
  • My system doesn’t lead to many trades (this is fine and good to know)
  • I sometimes forget to fully fade the opposing momentum to check my decision (and I thought I was very patient)
  • The approach was profitable in this period (but this market was trending strongly in one direction)

Early days but faster learning

I’d stress that this is a very small sample, and not much to go on. Next step is to continue through 2016-present, and see what my strategy looks like then.

So if you’re like me and confused with where to start, I would recommend putting some skeleton principles down on paper and applying them to some charts. This isn’t true backtesting, but it helped me to see the principles that I needed to nail down in my plan.

You’ll soon start to edit the plan and add more criteria, much faster than doing it in a real-time with a demo account.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade