Topics created by richard96816

Thanks. I tried that, but will try again.

I've been searching the whole system for the file without luck. No new files on the system after running it. No error messages. 😞

Thanks.

Thanks. Unfortunately, that doesn't seem to include all information. Like Magic Numbers.

Not sure what else is missing.

if your strategy heavily depends on ticks or calculation based on ticks (or not closed current candle) - opening and closing trades, than it is not possible to test with good accuracy in tester

if your strategy depends on repainted signals, it is not possible to test with good accuracy in tester

if your strategy (opening and closing trades) depends on history candles (first closed candle, or second, or indicators based on that, and so on) than you can test it in tester

.... and so on ... it all depends on that strategy
.... and also for demo/live, if you was profitable first month, the second one you can be lossy (this is true for all strategies) ... forex is more lossy than a casino (99% of all traders lose their money - if anyone claims the opposite, than is working for some organization (books, forex training, signals, EAs based on shits, broker itself) or is not experienced enough)

@fxdreema Thanks for the response. I worked around it already.

Learning to be much more thorough in my testing.

Thanks.

@miro1360 Thanks very much for the response. Your example looks interesting. Need more time to absorb.

Thanks!

No. I don't even look at those that seriously. I know that often people want to check pretty much all of them and I don't understand that. For me it's enough to check 2-3 of these, to play with them for a while and then uncheck them. For something more serious then use the Constants.

@miro1360 @richard96816

Usually people looking for these approaches are using (or considering using) multiple entries on losing positions (i.e. averaging), hoping price will bounce back and allow them to exit at a tiny profit.

This does happen most of the time, so it's in fact very easy to get a streak of 20-30 days without a single losing trade... The problem is when one single move doesn't come back to your average price... then all your money is gone in one trade..... and you can be 100% sure that this WILL HAPPEN at some point, and it won't take very long. It's just a more painful way of dying 🙂

There is no magic, the multiple tiny profits happen because the trader is taking infinite risk (the whole account) in exchange for a minimum profit. So, for each single trade, the chance of exiting on tiny profit is much higher than blowing up the account... However, you just have to trade long enough to see your account blow up in that one bad trade.

MT4 was the winner and I think it still is... depends on what platform the brokers provide. MT5 started somewhere around 2010, but MQL5 was different and complicated for most people, so MT5 never got more popular than MT4. More than that, in MT5 we were only allowed to have 1 position per symbol. But not very long time ago they added "Use hedge in trading" option, so MT5 is now almost the same as MT4 in the way trades work. Also around that time they stopped developing MT4 anymore.

I'm not sure that those real ticks are really real. You know that in MT4 you can import history data from somewhere else, and you can import real ticks... but when you do that, you are using many gigabytes of data. Where are those gigabytes in MT5?

Okay, never mind. I figured out how to show variables in the tester log.
fxD can be a treasure hunt as times ... 🙂

I put this line into a Custom MQL code block and connected it after my Buy and Sell blocks ...

Print("***** T/P: " + v::TakeProfit + " S/L: " + v::StopLoss + " *****");

Stuff inside quotes is just text. Variables are prefixed with 'v::'. Constants would be prefixed with 'c::'

My code was occasionally setting stop loss to zero. 😞 Made a ton of funny money.

Documentation is like writing a computer program. If you don't cover every detail completely and perfectly it will probably fail at some point.

(Yes, writing documentation kind of sucks. Been there, done that. 😞 )

It's a necessary evil.

Thanks for all you do!

Thanks for your response. I appreciate that there's power within the tool. I'm just used to having defaults more carefully chosen. So you can whip up a simple EA and touch the least number of options. And fall prey to fewer unexpected side-effects.
fxD is a power saw without the blade guard. Even experienced carpenters make mistakes from time to time.

When the defaults are well chosen the operator needs to keep fewer issues in mind and can focus better on their current strategy.

Having an EA unexpectedly reach outside itself and modify other running EAs is troubling. Every simple fxD example EA I've seen online makes this mistake. That's not good.

The simple case should be automatically self-contained, not touching other EAs. To touch other EAs should require consciously making that choice.

fxD makes 'badly behaved' EAs by default. 😞

Isn't there a custom indicator to do that already? I know that in the past every time I searched for an indicator to do something as basic as this, I only found crap, but this was years ago.

I found this topic meanwhile: https://www.fxdayjob.com/candlestick-pattern-indicator-mt4