Quote:
Originally Posted by hotsawce  Ken, I understand your status, but rather than writing off the importance of straight through brokers and their spreads, have you ever actually taken the time to compare? There are pages here of us discussing different brokers, comparing statements and trades, and ultimately coming to the conclusion that even small differences in spread affect how the robot works.
The evidence conclusively shows Megadroid will behave better (execute more trades) with a straight through broker with a low spread. Please, go back a few pages and look at some of the statements. Nearly everyone who has used an ECN broker has been making a killing. Why do you think Megadroid consistently makes a killing on www.forex-robots.com? It's on Alpari UK, which doesnt have a dealing desk and is STP.
Honestly, though, I really don't care if you refuse to believe this little differences are significant.  My setup slays, and I've got Blackdog to toy with when I want a little more. |
hotsawce, The constant attitude is quite annoying. I feel like I made the mistake of questioning the proverbial sacred cow.
I have read all of these postings, and I never said there was anything wrong with straight through proccessing. I currently have accounts at FXCM (straight through, no dealing desk, 5 decimal pricing, and very variable spreads), and FXDD (4 decimal pricing, stable spreads, I'm not really sure about their dealing desk, but its really of little concern to me). Last night, 3/16, MegaDroid took the same trade on both accounts. It was a set 2 Aggressive trade. On the FXDD accounts it closed with 6 pips profit. On the FXCM account it closed with 3.8 pips profit. The only real difference being that on FXCM the spread fluxuates constantly, whereas FXDD's spread is almost constant at 2 pips.
There is no doubt that MegaDroid can and will take more trades with a tighter spread. It uses a range calculation to determine trading potential, and wont take a trade if the combined profit target plus spread cant be easily captured. On the other hand it seems to have much better results with a stable spread that doesnt fluxuate wildly during its entry and exit proccesses. The problem being, a lot of people use MegaDroid, if they all take a similar trade, at a similar time, this will be reflected directly in the spread, if you are using straight through proccessing. All those little orders coming through at once will be detected, and the spread will automaticaly widen before they can all be proccessed. However, the advantage of a few more trades will generally outweigh this.
There was never much question about this, so I dont know what your case is. The real question was: If this is done through a commision based broker, (I believe ATC is commision based, but I could be wrong), what are you paying in commisions? Its usually a pip, or more. And how is this affecting the end results, not just the number of trades, but the overall profits? My concern is that commisions are not factored into the trade ahead of time, but are deducted after the fact, thereby reducing the actual profits. In MegaDroids case a lot of its claimed profits are really breakeven trades with just 1 pip profit, in all cases these would get wiped out and probably go negative after commisions. MegaDroid does have enough 10 and 15 pip winners to make up for this, (megadroid.com has a posting of what I believe is PipBusters 50K account, with after trade commisions, and it shows great profits, but I also believe its a demo account which is not quite accurate to slippage and spread changes), on a robot with smaller profit targets and more frequent losses (FAP-T), this could really be a problem.
I really dont understand what complaints or problems you might have with my posting. It was mostly for the benefit of a new trader who had asked me specificaly about this EA, and all I recomended was staying objective, comparing the comments on a few different forums, and studying the backtesting results, which do show that the current great run we have been having is not the norm. In past years, the backtests show more losses, larger losses, and consecutive losses. This has to be figured in to your overall plan if you intend to survive this game long term.