Not pushing the thread off of the rails AT ALL. Any and all discussions are really welcome.
Well it’s not a strategy to be honest. You know: buying something and if it goes against you then you just hold onto it hoping. On that disgusting horrible trade of mine: yes Gold did rebound and months later had I been able to hold onto the huge hole that I’d dug for myself I would have cleared out with some serious money. BUT: you never know where the bottom is. There is an OLD saying in the stock market: in a bear market you run out of money long before you run out of ideas!!! LOL!!! And Gold was REAL bearish at the time. Point is: it was sheer coeincidence that Gold turned the very next day. There was really no telling if it could go down further. And I’d reached my pain threshold at the time and wasn’t prepared to lose that capital that I had in the account. And Silver is a very good example of this. Silver went down at some point and I always thought that one could buy a metal and if you just held onto it for long enough the price would rise. That ain’t true. Barring one occasion: Silver spiked at one point and then went down again and just stayed there for months and months and months and months. And if you were sitting on a loss that’s tough to wait out for sure. Not to mention the fact that you could be using that money somewhere else to make money. And remember you’re paying interest every single day. The interest that I paid on that Gold trade was in the thousands and that makes the loss even worse obviously i.e. every day you are in the trade your capital is bleeding away slowly but surely.
So there’s some things from the horses mouth for you to digest!!! LOL!!!
GBPUSD test results, optimum parameters are TP 100, SMA 140 to 170. Profit over 1 year 23.3%. So there’s no generic high performance settings, very unfortunately, which means that they might shift over time. What works today might not work tomorrow, or next year.
I have also tested USDJPY and the results are indeterminate and profits not very large - best is about 4% from a mish-mash of parameters, no clear optimum.
I actually cannot believe that I have wasted an entire week watching this lot go absolutely nowhere. I’m seeing tankers on fire, Trump continuing to try bully the rest of the world, and who knows what else. And the only thing that appears to have moved with any conviction is oil. And tomorrow is Friday.
Oh well. Let me take a break. Take nice leisurely stroll in the cold to the shops. Life is bliss.
For the sake of interest: in these tests that you’re doing are you able to tell how many times your stops are being hit??? And on what basis are you calculating where your stops are???
No stops. I have found that stops kill profits. All I use is the emergency stop provided when price closes on the wrong side of the SMA. The SMA value varies as per the results I have published.
I could look at how many times the emergency stop is being hit, but that would require the stratey tester to use the “visual” option and test one setup at a time, which is VERY time consuming. I can do that for the most profitable settings and let you know.
My settings are: Capital $3000, lot size 0.1 (to allow the EA to do 0.01, 0.02 etc.), no stops, no trailing stops, no move to B/E, enter when RSI has closed beyond 75/25 for two days etc as per the book, but no exit when RSI(2) crosses 70/30 - the exit is by TP. TP and SMA as per the published results.
I understand that the EA can use lot sizes that are not neatly divisble by sub-lots of 0.01, 0.02 etc, and that the EA will round the resulting sub-lot sizes UP. I’ve been so busy with testing that I haven’t got around to checking that, but will do.
OK, for the best settings that I tested there were 46 individual trades (made up of 1s, 2s, 3s and some 4s). Price crossed the SMA160 four times in the year, causing 9 individual losses (worth $134.47). There were 37 individual wins worth $849.28. Total profit off $3000 was the difference: $714.81, or 23.8%.
Here’s the whole year visually presented. I watched it all play out and I can assure you that the EA is working exactly as intended.
Having ascertained that a TP of 100 and SMA of 160 is the optimum for GBPUSD over the last year, giving a profit of $711, I went back to the “by the book” approach, closing trades using the RSI crossing over the opposite 30/70 level. With SMA 160 I got a profit of $189 and with SMA 200 I got a loss of 19 US cents, call it break-even.
So it does seem that this optimization process I’m going though, while seemingly not providing a generic magic number that works across the board, can tweak what is already a sound system and make it pretty impressive. Of course, as always, historical results, no matter how accurately modelled, are no indicator of future performance.
Hey listen. As is said before: I know it’s not for my benefit but I do thank you. If nothing else: your findings are pretty much the same as anybody else that’s embarked on testing this concept i.e. it’s pretty robust. And nice work.
For what it’s worth there’s only two things that you need to be aware of (one of which you’re already aware of but I will mention for the sake of completeness):
Demo and historical data will yield different results unfortunately. There’s and entire list of reasons why unfortunately. Exactly HOW much of a difference though with THIS system is debatable though i.e. the reliance on stops for stop losses is negligible so for the most part stop hunting and slippage is not something we really need to be concerned with.
The other elephant in the room UNFORTUNATELY is the possible pitfall of curve fitting a system i.e. by tweaking parameters specific to an instrument and time period. I sure do hope it works for you and I mean that most sincerely. I myself prefer those TradeStation tests that were done for a variety of reasons (not least of which it suits me to take them as gospel!!! LOL!!!) i.e. they were performed using years and years worth of data to begin with. And in all of those tests: this system only lost the plot for one particular year.
Anyways. I’m pleased with your findings anyway ESPECIALLY seeing as it has been a goal of yours to prove the system on spot FOREX. It is my sincere hope it serves you well.
Hi Dale, thanks for your comments, yes there’s always these elephants in the room. Which is why this lark can be so interesting and soul-destroying all at once!
Re forex, I began testing on forex pairs, being what I know and feel most comfortable with. However, I also have a demo account with various forex pairs on it AND three indices - DAX30, S&P and DJ30. That’s running the system as published. But I’ll get onto testing the indices too. My problem is that I have no sense of the appropriate lot size to use, so I’ll just wing it and see what happens. Yes, I know, I know, and I won’t do this at home!
I see the DAX value is currently around 12500. What are those things? Pips? Euros? What might an equivalent TP be to match 100 pips of GBPUSD? See how little I know about that stuff?!!
On a test run I made 84.25 on one of the trades. What are those numbers? How much value do they represent?
Just realised I made an error with the testing - I left the “Exit when RSI goes over 70/30” setting switched on, even though I was supposedly testing specific TPs. Sorry! Deleting all the results that I had placed below here.
This is the dog that’s been biting me! Spinning a chart back in time and running my strategy bar/candle at a time, on the Daily, and things are looking great! Then I go down the rabbit hole and start “refining” on the 4H and then to the 1H then I take a quick look at the noise in the 30M and I am further refining and creating exception rules and checking the moon phase and tide charts! It becomes a total CF! Like holding fridge magnets and my forex compass in the same hand. Grrr.
So I’ve learned I can be a curve fitter and am now sticking to entry on the daily using my 4H simply as a “heads-up”, No longer trying to anticipate tops and bottoms and simply looking to take a chunk - that’s a fancy forex term - out of the move. DONEbOOm while pyramiding/adding to my position.
I tested 60 combinations with TP ranging from 90 to 420 points/pips and SMA from 170 to 290. The top 5 results here suggest that I was right earlier about the optimum SMA - hovering between 170 and 200 (so maybe I should look below 170 too), and the TP fairly high, for someone accustomed to forex numbers, but maybe not to you lot!
