I’ve looked at and backtested to high-heaven a range of strategies. Nothing against mean-reversion or trend-following, but being consistent in those areas is harder than in momentum for my style of trading. Mean-reverting into the trend is great, but it relies on support/resistance levels working, or candlestick patterns that aren’t statistically proven anywhere. Momentum has its own issues - the signals can come late, for instance - but overall, if measured well, it can give much more consistency. My typical backtesting method now involves 20+ pairs over 5+ years, so the results of major concepts are usually quite conclusive, whether the equity curve goes up smoothly, is more of a 50-50/low edge wavy curve, relies on a few lucky hits or even does amazing until a catastrophic loss. Obviously I prefer strats that do the former. Smooth growth with low DD. Momentum is fitting that bill this last 6 months.
I’ve tried to measure momentum in a bunch of ways, but my favourite to date is measuring movement across a basket of pairs using the same base currency and measuring that against the quote currency. So if the base is up and the quote is down, across a handful of pairs each, it’s a good starting point for a trade. Then you can check if you’re with the trend, at a good support or resistance area, check your favourite indicators for confluence etc. but to my mind the odds are well in your favour if you’ve confirmed that the dollar is rising and cable is falling before you enter, for instance.
So I recently wrote an indi that gives signals over 3 types of momentum measurement, in varying levels of confidence, giving an early-warning effect. I have a bot that was very much backtested over 20+ pairs running since April (has taken no trades yet) and will be semi-manually trading using the indicator. Next up to build is a panel to set up my SL/TP/exits dynamically so I can enter in one click and let it manage the rest. It’s early days, but after 30 months of testing hundreds of permutations, I’m honing in on what works for me. I might bring back some mean reversion stuff if it gives me another ‘early warning’ to the momentum trades, but that’s all in good time.
For info, I’m also trying to build my trading capital, so will be eventually selling the momentum EA (trading bot for MT5) I’ve built once the forward testing definitely matches the backtesting. No reason it shouldn’t, as it’s a very simple setup based on newbars and just joining the strongest moves, but needs proving over the next year at least for now. If it helps, I can bundle in the manual trading indicators and/or panel, but that’s a different kettle of fish. I need to start talking to people looking to do this, have capital ready to trade but don’t necessarily want to spend 2+ years getting to square 1. To do that I’ll be sharing what I’ve done so far (I’ve got a huge plaintext file journal) and hopefully be able to then prove that I’ve reached a stable place to work from.
PS - You can join the Discord group and discuss this at the link below. Talk soon!