Zigzag Circles Strategy Test 1st April → 6th June@LonesomeTheBlue has recently published an interesting script LTB Zigzag Circles . I got curious about its performance, so I tinkered out a strategy and tested it out on GBPUSD.
I will use ZigZag Circles with a period of 14 on a 30m timeframe. Furthermore, I will utilize round levels to place my entries, exits, and stops.
When there is a bullish signal:
► a limit entry is placed at the nearest-below x000, x200, x500, or x800 level.
► a stop-loss is placed halfway through to another such level below.
► a take-profit will be at the nearest x000, x200, x500, or x800 level to the upside.
When there is a bearish signal:
► a limit entry is placed at the nearest-above x000, x200, x500, or x800 level.
► a stop-loss is placed halfway through to another such level above.
► a take-profit will be at the nearest x00, x200, x500, or x800 level to the downside
(check the first picture for clarification, round levels are shown in gray)
#1 A first entry signal at 17:00
Look at the limit order. A limit is right before 1.38200, the exit before 1.38500, the stop-loss between 1.38000 and 1.38200.
#1 A first order is closed in profit in roughly 12 hours
#2 A second position opened and closed according to rules
#Not All Entries Trigger
Obviously, I didn't leave the entry orders hanging out there forever.
#3 A first loser
This is how a lost trade looked like.
#5, #6 A little problematic area
This is a problem you might encounter with LTB Zigzag Circles. Signals point in both directions... I wouldn't take the long one as the signal is far from my level, but during the backtest, I follow the rules. Hence a win and a loss.
#8, #9, #10, #11 a series of losses in a consolidation area
When the price keeps on contracting, there might be several false signals in a row. Even though you may still see some respect for the x500 zone, the bounces are small and not enough to reach the profits. In real trading, the entries could be identified as unconvincing and avoided.
#14, #15 Two winners in a row
Although the stop-loss was dangerously close to where the price once was for the first one, this is how the winners in a row might look like.
#22 Another problematic situation
Lastly, I wanted to put attention to this imaginary trade. The very point of using limit orders in this test is to make sure I have enough time to get the signal and evaluate its quality. I do believe, there wouldn't be enough time to open a position from a level designated by rules. So I put it on the nearest one below. I know I wouldn't want to take that trade after I was notified at the candle's close. If you would, count it as a loss.
Conclusions
37 trades were open since the beginning of April, out of which 44% were profitable with an accumulated profit of 1141 ticks .
When backtesting, I like to subtract 3 pips from every trade to give it a better feel of the real world. You make mistakes and you also pay spreads and fees. If I subtract 3 pips from every single trade, the strategy will still be profitable, but barely. It's 91 ticks only. Of course, the real spread is smaller than that, so this is a good result also.
Lastly, I saw the price to continue in the direction of the trade far exceeding its original take profit. With advanced trade management (which can not be backtested easily), profits would be higher.
Remember, that 37 trades is a narrow sample size. It is enough to give you an idea of what parts of the strategy to improve. But for accurate statistics, you should test your strategy 180+ times.
