Interesting topic for many: what's new in MetaTrader 4 and MQL4 - big changes on the way - page 34
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
What are you guys talking about?!
What spread history, if they still can't even figure out the normal history. I just downloaded the EURUSD M30 chart from the official MetaTrader (NordFX):
2008, September. On the left are the daily bars and on the right the hourly bars. All this stuff is on the M30. The Expert Advisor cannot be sure what data it is dealing with at the current moment, and you are talking about the history of some spreads. Things are no better in the exchange segment. There are no normal futures glues even for real accounts. Hanging position after expiration has been hanging on my demo account for half a year now, I can't close it. Because of all this and much more, with all due respect to MQ and their work, I just can't use MT tester at all. And some spreads have nothing to do with it at all.
Is the flipper a Bid or an Ask?
Let's disassociate ourselves for a moment from the notion of selling and buying (this is just the direction of one party's operation).
Let's introduce the concepts of a trader and a client (this will dot the i's and cross the t's).
If you trade with limits, you are a trader, and your orders will be displayed in the DOM.
If you trade with stops or market, you are the client (you kind of look at the current "offer" and agree to it, "offer" in the sense of what bids are currently available).
So client's orders are not shown in the market as they are executed right away. They are executed with a small difference (provider's commission), but the Last price shows the client's price, while the trader's price is slightly different for the worse.
Let's disassociate ourselves for a moment from the notion of selling and buying (this is just the direction of one party's operation).
Let's introduce the concepts of a trader and a client (this will dot the i's and cross the t's).
If you trade with limits, you are a trader, and your orders will be displayed in the DOM.
If you trade stops or market, you are the client (you kind of look at what is the "offer" and agree to it, "offer" in the sense of what are the current orders).
So client's orders in the market are not visible in the stack because they are executed immediately. They are executed with a small difference (provider's commission), but the price displayed by the client is the client's price, while the trader's price is slightly different for the worse.
Let's disassociate ourselves for a moment from the notion of selling and buying (this is just the direction of one party's operation).
Let's introduce the concepts of a trader and a client (this will dot the i's and cross the t's).
If you trade with limits, you are a trader, and your orders will be displayed in the DOM.
If you trade stops or market, you are the client (you kind of look at what is the "offer" and agree to it, "offer" in the sense of what are the current orders).
So client's orders in the market are not visible in the stack because they are executed immediately. They are executed with a small difference (provider's commission), but the price displayed by the client is the client's price, while the trader's price is slightly different for the worse.
That's a nice twist. Yes there are bids in the cup and I know that those prices at the top are market aski for me and those at the bottom are limit aski. And vice versa. 2 different orders, 2 different prices, and the output is one flipper. I don't really care - I explained it to the man, and you try to explain it to me.)
No, it's not. It is just the opposite. If without Ask the TS is profitable - then the probability that the TS will continue to be profitable - will be higher than if with Ask - there is a profit, and without Ask - there is no profit. At least, the stability and robustness of such TS will be out of the question...
Bullshit! The accuracy of testing is the closest result to the one that was in real trading (if there was a trade), or could have been (if there was no trade).
When you write a TS and run it on a real account, then you just need for this TS to behave exactly as it did in the tester. Because if it will behave differently (even more profitable) on the real - then it's not your TS, it's a chance.
What the hell do we need a tester, if it cannot even come close to a real trading robot? With such tools even to study the market is self-restriction and mockery.
Well, in theory, yes, you could probably come up with such a STABLE and profitable TS. Although, I seriously doubt about the stability of tick systems - slippages and requotes will do their dirty work. I tend not to use data of timeframes below H1 precisely because of its extreme instability.
Well, the one-minute and above spread is kind of there, use it as you need.
I mean, I don't mind that "it would be nice to have information... "But I don't see much sense in it, and I don't think developers will bother.
What tick spread? Do you even understand that there shouldn't be any spread in the tester? And the 99% of ticks are unnecessary.
You must have only HighBid and LowAsk. No ticks or spreads. Stop talking nonsense.
Why the hell do we need a tester if it can't even get close to the real thing? I've noticed a pattern in the market that has disappeared in the past.
Test it in the real world, it's the surest way to do it.
ZS: I've noticed that the patterns disappear as soon as I notice them.)
Bullshit! Accuracy testing is the closest result to what was in real time (if there was a trade) or could have been (if there was no trade).
When you write a TS and run it on a real account, then you just need for this TS to behave exactly as it did in the tester. Because if it will behave differently (even more profitable) on a real - then it's not your TS, but the chance.
What the hell do we need a tester, if it cannot even come close to a real trading robot? With such tools even to study the market is self-restriction and mockery.
What you are writing here is already a plus for this toolkit.) The real and the tester will never coincide. You can set a lag on the spread and simulate different outcomes of your trading strategy. Tell me which toolkit allows you to do it without any limitations? And why doesn't your broker give you this information, if that's what you want, since the tools of this mocking product allow you to do so (MT5)?
test it out in the real world, it's the surest way
ZS: I've noticed that patterns disappear as soon as I notice them )