Aaaand the numbers are in! Now that a full trading quarter has gone, let’s take a look at the backtesting results of the HLHB Trend-Catcher System since I tweaked its parameters earlier this year.
Read all about my HLHB Trend Catcher System if this is your first time hearing about it!
As I’ve mentioned before, I’ve decided to stick to my second tweak in my forward tests for the next couple of weeks. Basically, I’m adding the ADX indicator as a filter on top of the RSI and applying all of them on the 4-hour time frame instead of last year’s 1-hour charts.
Here’s what came up after crunching the numbers in Q2 2017:
Take note that we are risking 0.5% per position. It would probably be better if the risk is bumped up to 1.0% per position, but I’m keeping it at 0.5% for now for easier comparison against my other backtest results.
First of all, you can see that there are waaay fewer valid signals in Q2 compared to Q1. One explanation for this is that, unlike in Q1 2017 when major central banks hinted of their biases for the foreseeable future, Q2’s economic themes were a lot less trend-friendly.
Brexit concerns, for example, were tempered by better-than-expected U.K. data. Ditto for the ECB’s tapering-is-not-tightening rhetoric, which saw strong reports even as the region’s banking crisis loomed. And let’s not forget the Fed’s undeniably hawkish bias that was countered by the Russia-Trump press convfefe!Meanwhile, EUR/USD and USD/JPY’s win rates have improved though EUR/USD’s average wins and losses have stepped back. And can we talk about GBP/USD’s three valid signals that turned out to be fakeouts? WHAT. IS. UP. WITH. THAT?!?!
I reviewed my charts and found two possible explanations for the HLHB System’s poor performance on both EUR/USD and GBP/USD.
One is that some of their strongest trades started after consolidations. This increased the chances of the RSI already hanging out below/under 50, so the signals were invalidated for the lack of an RSI crossover.
Another thing I’ve noticed is that Q2 2017 was just plain unfriendly for trend strategies that have (relatively) tight trailing stops.
If you check EUR/USD and GBP/USD’s numbers, you’ll see that their biggest losses were due to big, single candlesticks that happened to be at the wrong place at the wrong time. Not only did this stop out existing trades, but it also meant that ADX was often still below 25 when the signal candle popped up.
Overall, it looks like only USD/JPY’s strong and steady trends yielded serious pips for this trend-catcher. And because it STILL shows the same problems that we spotted in Q1 2017, I’m starting to think if this 4-hour gig is more trouble than it’s worth.
Are you spotting any tweaks I can make to the current system? Or should I get back to the good ol’ 1-hour charts?