Welcome to Forex-TSD!, one of the largest Forex forums worldwide, where you will be able to find the most complete and reliable Forex information imaginable.
From the list below, select the forum that you want to visit and register to post, as many times you want. It’s absolutely free. Click here for registering on Forex-TSD.
Exclusive Forum
The Exclusive Forum is the only paid section. Once you subscribe, you will get free access to real cutting-edge Trading Systems (automated and not), Indicators, Signals, Articles, etc., that will help and guide you, in ways that you could only imagine, with your Forex trading.
Elite Section
Get access to private discussions, specialized support, indicators and trading systems reported every week.
Advanced Elite Section
For professional traders, trading system developers and any other member who may need to use and/or convert, the most cutting-edge exclusive indicators and trading systems for MT4 and MT5.
The grail_1.mq4 by Sergey Kovalyovis a good example how things happen:
If you backtest it, you will see, that before 10/01/2006 it will gain high profit, after 10/01/2006 the algorithm looses. This is at every currency!
So what happend in 10/2006?
YES, Sergey published his article and the ea in 9/05/2006!
So two interpretations are possible:
a) Some big players without brookers limitation use the grail system with big profit and this changes market.
b) Brookers were hit by the grail system and implemented filters to avoid it making profit.
I think it is both of it. If it is only b), then the grail system should still work at non-mt4-brokers. I will check this at ...
Hi BigBoppa, I think 2nd answer is true on MT4 platform. So for this reason I would like to test my EA on ECN like MBTrading for example.
2 problems:
1) MBTNavigator isn't hedge capability so I need 2 account to trade long and short orders in the same moments.
2) I need third part software to make automatic trading. I like NinjaTrader 6.5 that use C# language (similar mql).
Now I'm working to translate my EA and then I'll start forward test again.
Do you have same skills to translate EA from mql to C# ?
Thanks
The grail_1.mq4 by Sergey Kovalyovis a good example how things happen:
If you backtest it, you will see, that before 10/01/2006 it will gain high profit, after 10/01/2006 the algorithm looses. This is at every currency!
So what happend in 10/2006?
YES, Sergey published his article and the ea in 9/05/2006!
So two interpretations are possible:
a) Some big players without brookers limitation use the grail system with big profit and this changes market.
b) Brookers were hit by the grail system and implemented filters to avoid it making profit.
I think it is both of it. If it is only b), then the grail system should still work at non-mt4-brokers. I will check this at ...
Finally someone else seems has picked up on this - I have posted this very date several times but most think I have just had one brewskie too many!
But this has nothing to do with any particular system. I have three EA that I developed, each using different entry techniques, and they all show the same fall-off at this date. In addition, this is evident in almost every other 'winning' EA - the reality is that most EA's are no better than random enteries, meaning when commissions are taken into consideration, they are out-and-out losers. Finally, check this recent post of mine:
- these are the results of an always-in-the-market martingaler, which will close all orders if the account d/d exceeds 5%. What it clearly shows is that after Sept-Oct 2006, 'abnormal' volatility of some sort suddenly kicked in. Abnormal in the sense that EA's that were extremely profitable continuously in the preceeding 7 years are no longer...
The real mystery is, what caused this, and why is this change so pronnounced and abrupt - one would expect a gradual build-up, but this does not happen...
Hi BigBoppa, I think 2nd answer is true on MT4 platform. So for this reason I would like to test my EA on ECN like MBTrading for example.
2 problems:
1) MBTNavigator isn't hedge capability so I need 2 account to trade long and short orders in the same moments.
2) I need third part software to make automatic trading. I like NinjaTrader 6.5 that use C# language (similar mql).
Now I'm working to translate my EA and then I'll start forward test again.
Do you have same skills to translate EA from mql to C# ?
Thanks
Bolla
Yes, I agree, to a certain extent we are at the mercy of crap MT price data - this could indeed be the reason. But with regard to my above post about the martingaler, I find it hard to see how ECN data would significantly alter this result.
What also seems clear is that all ECN data differs as well. For example, Interactivebrokers' feed is sourced from at least 6 ECN's, on a 'best price basis'. While this results in huge available liquidity, it also is an extremely effective way of filtering out all the spikes we are familiar with from the MT brokers. So running a scalping EA such as 'Lucky' with this type of no-restrictions feed is doomed to failure, as practically no trades are triggered (I know - I've tried...)
Hi,
I'm testing few scalpers based on Championship 2007 EA's.
This is MetaQuotes demo, standard setting.
I thing that broker will use filter for these EA's.
Bongo
Finally someone else seems has picked up on this - I have posted this very date several times but most think I have just had one brewskie too many!
But this has nothing to do with any particular system. I have three EA that I developed, each using different entry techniques, and they all show the same fall-off at this date. In addition, this is evident in almost every other 'winning' EA - the reality is that most EA's are no better than random enteries, meaning when commissions are taken into consideration, they are out-and-out losers. Finally, check this recent post of mine:
- these are the results of an always-in-the-market martingaler, which will close all orders if the account d/d exceeds 5%. What it clearly shows is that after Sept-Oct 2006, 'abnormal' volatility of some sort suddenly kicked in. Abnormal in the sense that EA's that were extremely profitable continuously in the preceeding 7 years are no longer...
The real mystery is, what caused this, and why is this change so pronnounced and abrupt - one would expect a gradual build-up, but this does not happen...
I also have a system that worked until October 2006, but unfortunately discovered it somewhere in the middle of 2007 What I found strange about it, however, is that it was only profitable when run on the historical data from the MT History Center, but not with Alpari M1 data, for instance. I thought at first that in October MetaQuotes Corp. changed the source of their historical data, and with the new data my EA didn't perform so well. Later, I remember coming across a post, explaining that MetaQuotes used combined data from several brokers. Assuming that they do that indeed, historical data cannot be the reason for the bad performance of the EA... can it?
I also have a system that worked until October 2006, but unfortunately discovered it somewhere in the middle of 2007 What I found strange about it, however, is that it was only profitable when run on the historical data from the MT History Center, but not with Alpari M1 data, for instance. I thought at first that in October MetaQuotes Corp. changed the source of their historical data, and with the new data my EA didn't perform so well. Later, I remember coming across a post, explaining that MetaQuotes used combined data from several brokers. Assuming that they do that indeed, historical data cannot be the reason for the bad performance of the EA... can it?
Hi, I appreciate the response. Where can I lay my hands on Alpari M1 data - or are this via the 'History Center' in the usual way??? I have installed several MT 'flavours' already (IBFX, WCH,and Aleccohfx) and all their price data produce identical backtests for me. Also, I was under the impression that since Metaquotes removed the 'Recalculate' option, it was no longer possible to use third-party data - is this true?
I also remember reading that Metaquotes' data was derived from various sources. The only way I could see the results we witness is if the majority of the data is somehow poorly 'fabricated' (something blatantly exploitable, in the M1 interpolated tick data), up until Sept-Oct 2006, and then MT downloads and stitches on something akin to 'real' price data up to the present date. Just look at Metatraders 1999 EURCHF data and it is clear that this could not possibly from a 'real' price feed...
EDIT: - Just installed the Alpari MT and did a backtest. As can be seen from the results, the results are virtually identical - therefore, so is the price data...
Do you have same skills to translate EA from mql to C# ?
Thanks
Bolla
Hi gbolla,
unfortunately i do not have C# knowledge (even if it´s similar to mql)
I will code it at another "Trading-Language" for other brokers.
The ea has a quite simple algorithm, so it would be the easiest way to code it new from start, not to recode it.
What I found strange about it, however, is that it was only profitable when run on the historical data from the MT History Center, but not with Alpari M1 data, for instance. I thought at first that in October MetaQuotes Corp. changed the source of their historical data, and with the new data my EA didn't perform so well. Later, I remember coming across a post, explaining that MetaQuotes used combined data from several brokers. Assuming that they do that indeed, historical data cannot be the reason for the bad performance of the EA... can it?
Linear Mode
