The search continues! Ready for the backtesting results of my latest HLHB tweak?
For those who are just tuning in, you should know that I’m on a mission to make my Huck Loves Her Bucks (HLHB) mechanical trading system a bit more sustainable by finding a way to get fewer signals while getting the same fabulous results.
Here are links if you want to catch up:
HLHB System’s original parameters
HLHB System tweak #1 backtest results
HLHB System tweak #2: new rules
In case you missed it, tweak #2 basically makes the following changes:
- Use H4 instead of H1 and adjust profit targets (from 200 to 400) and trailing stop (from 50 to 150)
- Add the ADX (14) indicator applied to the close and buy/sell only if it hits 25 and above
- Closing an open position due to a new signal is only applicable to VALID signals
So, without further ado, here are the backtesting results from Q3 2015 – Q4 2016:
As you can see above, adding the ADX did the job of reducing further the number of valid signals. Oh, and take note that I’m comparing an 18-month backtest vs. the 12-month performance of the previous parameters.
What’s even more exciting is that the win rates of tweak #2 are generally better than in tweak #1. And, though average wins have failed to maintain their “I’m twice the size of you” advantage over average losses, the average wins of tweak #2 are still higher than in tweak #1.
One thing I’ve noticed is that there are only a handful of valid signals (read: average of 2-3) per month. Heck, there are even months when ALL of the signals are invalidated by either the RSI or the ADX filters!
I’ve also found that adding the ADX doesn’t guarantee that you’ll catch all the juiciest trends. In fact, tweak #2 still saw a lot of valid fakeouts and strong, shoulda-been-profitable signals that were invalidated.
Last but definitely not the least, the figures above are still shy of the HLHB’s % profits from 2016. EUR/USD’s 1.62% gain is a far cry from the 5.88% uptick on the original parameters, while GBP/USD’s 8.10% increase is better than the 6.34% gain in tweak #1 but is still short of the 13.19% gain it saw in 2016.
For now, it looks like I’m getting closer to refining my HLHB into a more sustainable mechanical system. The search isn’t quite over for me though! I’m still on the lookout for tweaks that might help catch trends on the 4-hour time frame. Specifically, I’m looking for signals that would catch more of the big trends that tweak #2 missed. Maybe just using the ADX and scrapping the RSI filter altogether? What do you think?
Got any ideas on what I can change? Maybe something that could help out with NOT invalidating good signals?
See also:
My Q4 2016 Forex Trade Review and Reflections
HLHB System’s 2016 Performance Summary
Read the risk disclosure!
This content is strictly for informational purposes only and does not constitute as investment advice. Trading any financial market involves risk. Please read our Risk Disclosure to make sure you understand the risks involved.