Quite a while since I updated this thread. The main conclusion is positive but not as expected.
The orthodox thinking on long-term trends says that the majority of days will have closes that follow the trend - higher in uptrends, lower in downtrends. And if price is not trending, there will not be a majority of upward or downward moves, these will be random, 50% up and 50% down - regardless of the EMA slope directions.
In fact, almost as soon as the trial began, market behaviour flipped between orthodox and abnormal, so that it is statistically much better to sell when the EMA’s are sloping upwards and buy when they are sloping downwards. Coming back later with some figures.
Thanks for the update @tommor . " “It’s better to have tried and failed than to live life wondering what would’ve happened if I had tried .” - Alfred Lord Tennyson.
OK, some weird demo results but which might be of interest and value.
While the majority of sessions produce positive returns, keep trading using with-trend positions as defined. As soon as the majority of sessions using this rule become losers, reverse the trade directions. Using the reverse rule on the 36 days since the flip-over occurred, this has produced 27/36 winning days (75%). On average, 12 trades were taken per day. The trade win rate was 277/446 (62%). I had expected about 50% in both cases. Weird.
Counter-trend daily positions continue to make profit, i.e. taking the opposite trade direction to the slope of the 20 and 50EMA’s. I’m trading the same 15 pairs (from the 28 majors minus all AUD and NZD pairs). The daily performances turned counter-trend on 02/09.
These are some performance data from 03/09 onwards -
sessions traded = 52
profitable sessions = 40 (77%)
trades taken = 663
winning trades = 409 (62%)
You say the MA’s need to be sloping a certain direction to tell you whether to buy or sell…
But sloping over what timeframe? Are you just eyeballing it? You obviously can’t take the slope of a point, so can you define how many periods back you use to determine whether it’s up or down?
I do in fact take the slope direction from the D1 chart at the prior day’s close. If its flat, I go back to the day before and take the slope direction at that close. I did say this was dumb though didn’t I…
There needs to be a really good reason within the terms of the strategy to not use a SL. These trades are minimum stake per pip and are closed after max. 9 hours. As they are baskets of usually about 14 trades there is also some degree of hedging each day.
I only excluded the AUD and NZD pairs so as to reduce the margin requirement. The system demands a trade on up to 15 simultaneous positions per day, but I have no reason t believe involving the 13 Australasian pairs also would lead to a different result.
tommor your strategy works. 100% However what kind of Forex trader are you?.Have you taken the babypips forex profile test. Some traders are scalpers, day traders, swing traders, ect… If you are are Swing Trader and have the patients to wait 2 to 4 weeks awesome !!!
$200 Account 1:1500 Leverage 28 Pairs US No Stop Loss Trading same direction on Day chart I took to $438 Lot size .01 in two weeks. However I am Scalper and do not have the Patients. It was a Demo.
Trade your profile , do lots of Demo;s and if you are a Day trader Go in direction of the Weekly/Day Time Frame. Demo first wait for the Reversals on lower time frame happens once a week and then open on another Demo and you will see what I mean.
I am almost fully a long-term trader, probably swing trader describe sit best. But I have always thought I needed a short-term strategy, as mechanical as possible, to use my margin allowance and utilise some of my spare time. Its not been as successful as I had hoped but the game isn’t over yet.
there’s always the possibility that this could prove also to be viable and sensible on shorter timeframes than daily?
i don’t know whether it will interest you, but some Ichimoku systems definitely have an edge, and are pretty much mechanical, for the very small minority of traders who have the patience to take only the very best, very highest-probability entries (which can also easily be defined mechanically), and use settings appropriate to what you’re doing (which, again, very few people do)
Thank you @tommor for this idea, and to everyone who has commented. It seems, from my limited experience, like there could be a number of ways to build and refine this into something pretty interesting.
My trading so far has mainly been on stocks, and mostly buy and hold for weeks/months or even years (so arguably more investing than trading). I have dabbled a bit with spread betting on indices and forex but without any real structure or discipline.
I’d like to use this as a learning exercise to try doing things properly: start with backtesting, refining, demo trading and hoprefully making it all the way to a profitable, live strategy. I was a programmer some years ago so I also like the idea of having a go at automating the model on MT4 eventually - but that’s getting a bit ahead of myself for now.
I’m pretty comfortable working with large amounts of data but i’m struggling to actually get hold of the data I need…
I have found this link as a good source for historical OHLC data for all the major pairs Forex and Currency Rates - Investing.com UK but I can’t seem to find anywhere that can give me the London open/close data which the strategy is based on. Does anyone know any sources for this? Is it a case of findin hourly data?
I’d really appreciate any pointers. I’m sure I’ll have more questions. Happy to share my progress and results if others are interested.