Brokers and Strategy Tester Accuracy?
Does that mean that both tests were accurate for that particular broker? Meaning, both brokers handle their spreads differently and are extra sensitive on some than others. So in a sense both are correct?
There are places on internet offering live-quotes of the largest brokers live account spreads. Compare those to your test results, especially in the winning case.
That's a really good idea. I'll do that just to see the consistency of the brokers to the platform. But I was also interested in the way certain platforms execute their trades with scalping. When having the SAME exact historical data uploaded, why are there major differences in the results. I was going to demo the profitable broker; but it's one of the ones that does not allow tp's less than 10 pips. So I can only back test it for that one. But the other broker matches its back tested data with its demo trading, so I think the problem may be within the platform itself......maybe
This is what I refer to as the Metallic Dragon in the room. New new guy comes in and his jaws drop to the floor because of what he's seeing but all the other traders around him is paying no attention to it. I was the very same way. First the Data... Sucks. Second the spread's between brokers would kill any scalper EA in terms or make it look like a Holy-Grail. Third, real life scalper programs and broker conspiracy don't mix. The best test for your scalper would be on a Real Account for as small testing money as possible. The back_test is only good for seeing if it's a working program. All in my Humble Opinions.
I thought of that too (with the spreads), which is why I made the tp 9 instead of something like 2. That way it would work out fine if the spread increased 7 pip larger than expected. But I guess you might be right, I am a newbie to this.
You also should understand that Mt4 back-tester Does Not simulate Real Ticks. Rather it uses 1-Minute Data by default. So if the prices went down 10_pips before going up 100_pips. Or vice-versa goes up 10_pips before going down 100_pips within a Minute. In either case, and depending on how much stop_losses your scalper is allowing, if say 100 pips stop_loss, you wouldnt know what hit you first, the stop or the tp. And if you're running a scalper with tp of about 9 and a sl of over 100 that's just more concerning than any other performance matrix you're trying to generate. All in my Humble Opinions.
I just want to thank you for taking the time out to respond to me. You really are helping me and making me see things more clearly. This is the only way I'm able to learn because I cannot seem to find the answers over the internet.
But I just have one more quick question in response to what you just have posted. In the EA that I was using, there were no stop loss. Instead, I was intending on using a margin call as a stop (while dividing up my total funds). Anyway, I made sure in all the minute data, there was NO TIME in which a margin call would have happened. Even if I took the highest point within a month and subtracted the lowest point, the pip difference still wouldn't cause a margin call because the lot size was too low. Also, there could only be a max of ten trades open at a time.
After further analysis, I noticed that many trades would open within a minute time frame as the ticks go up and down, but then as soon as it reaches its target profit, it would close the trade (which can happen within the next few minutes, many times within the next few hours, and every now and then within the next few days). But the frequency of trades are so much that it makes up for all the open trades (max 10).
I'm just trying to understand what the major flaw of the EA is because I don't think tick data would make THAT much of a huge difference. None of the trades were closed within the same minute; and according to the historical data, there were no times in which a margin call could have happened.
If trades are opened, and tps are reached in later time frames, and it's through a ECN metatrader broker, I'm not too sure where the inaccuracy rests.
I'm not trying to find the easy way out or holy grail (although it would be really nice), but I'm just trying to understand the coding better and why one broker may proves extremely profitable and another doesn't.
Thanks
Well, I got that same idea (to use my entire account as the stop_loss) about 4 months ago when I started. My thinking was similar to yours at that time. Also the strategy I was happy about at the time was a no-stop-loss strategy. I taught forget about what all these people are saying, I'm gonna put in no more than I can afford to lose. All I have to do is make sure that my probability of winning is more than the losses. Simple example is if I'm gonna deposit 500$$ to a broker. My no limit strategy would have a stop.loss of 400$$/Pips=1$. All I have to do is win enough to cover the $400 before my first big loss. So if my take.profit is 10$ then I need 41 wins to every loss to stay ahead. I suck at math, so I think in these terms.Who cares if I get a margin call on the last 100$ as I'm wiling to risk 400$ anyways :). I don't know what the win % off the top of my head you need to achieve this, but it's pretty unrealistic.
Well, if you plan to never trade again after you lose that 500$ stake than this is Not such a bad idea. This is the reason I made a point about it depending on your stop.loss and about stop.losses over 10x times the take profit. Sooner or later you'll realize that this strategy is sitting on losses to compensate for the wins. A series of 2 or 3 blown margin calls could really depress someone. I'm not saying this is not a good way to make money as there are die-hard believers in these types of systems. I just think if you spend the time, you can find different alternatives. Now if we're no longer talking about a Scalper looking for 12 or less take profit. then we have a different story. Or on the other hand the Scalper have a 12-points take profit and about 48 points or less stop.loss. Risk to Reward ratios is 1 name of the game. Then you have something you can better live with. All in my Humble Opinions.
Well, I got that same idea (to use my entire account as the stop_loss) about 4 months ago when I started. My thinking was similar to yours at that time. Also the strategy I was happy about at the time was a no-stop-loss strategy. I taught forget about what all these people are saying, I'm gonna put in no more than I can afford to lose. All I have to do is make sure that my probability of winning is more than the losses. Simple example is if I'm gonna deposit 500$$ to a broker. My no limit strategy would have a stop.loss of 400$$/Pips=1$. All I have to do is win enough to cover the $400 before my first big loss. Who cares if I get a margin call on the last 100$ as I'm wiling to risk 400$ anyways :).
Well, if you plan to never trade again after you lose that 500$ stake than this is Not such a bad idea. This is the reason I made a point about it depending on your stop.loss and about stop.losses over 10x times the take profit. Sooner or later you'll realize that this strategy is sitting on losses to compensate for the wins. A series of 2 or 3 blown margin calls could really depress someone. I'm not saying this is not a good way to make money as there are die-hard believers in these types of systems. I just think if you spend the time, you can find different alternatives. Now if we're no longer talking about a Scalper looking for 12 or less take profit. then we have a different story. Or on the other hand the Scalper have a 12-points take profit and about 48 points or less stop.loss. Risk to Reward ratios is 1 name of the game. Then you have something you can better live with. All in my Humble Opinions.
Right after I got your last response, I had put in some modest parameters:
Tp = 20 pips
Trailing Stop = 10 pips
This way it wouldn't be classified as a typical scalper and has a much better risk:reward ratio.
However, I still come within the same problem. I had an initial deposit of $10,000 for both brokers. One broker had a profit of $74,118; while the other had a net loss of -$9,799 (which is expected). Basically, one had a profit of over 700% within a month, while the other had a net loss of nearly 100%.
Since now I had incorporated a larger target profit and a stop loss on equal 5 digit MT4 platforms with identical data, would you say that the problem probably lies within the EA or the broker?
Thanks again for all your help.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Anyway, I was testing my scalping strategy between many different brokers to basically check for consistency on MT4. However, I noticed something between different brokers. When I deleted the entire currency history, and uploaded identical currency data, I started the EA.
It keeps on turning out that both brokers provided DRASTIC back tested results. Basically, one broker received little profit and ended up with a loss at the end of the month as with many scalping EA's. However the other broker ended up with a 300% profit.
I tested random months with the "desirable" broker, and the EA keeps on showing enormous returns with very a small drawdown. Similarly, the opposite happens with the other brokers.
I know pip prices differ slightly among different brokers; but I don't think trading results would be so different. Although BOTH are 5 digit MT4 trading platform brokers, should their results be very similar? Or is it just hard to predict scalping EAs on backtested data? Maybe some brokers altered the strategy tester itself? (ohh yeah, both had a 90% modeling quality and same time frame)
Thanks!