Чтобы пользоваться чатом, пожалуйста, войдите в систему
Обратно к контактам

Fast Forex Millions back and forward testing by Remus

Jan 21, 2012 at 09:21
9,728 Просмотры
120 Replies
Участник с Jul 24, 2011   91 комментариев
Jan 22, 2012 at 16:45

MDP has hard coded setup change in its mql file. which that made me stay away of using it.( this means MDP is a semiautomated EA if trader can manage to do it ) code for the new EA should be decompiled and checked for the same issue. hard coded setup change can also be done in DLL. to understand that the DLL function must be checked if any date/time data is been used.

fyi
walker


Not everyone knows that stuff. EA programing... I know I don't
ForexScam
forex_trader_32776
Участник с Mar 28, 2011   1008 комментариев
Jan 22, 2012 at 16:55
@Remus, Thanks for this tid bit. If logic is based on M15 it shouldn't take long to figure out - If you can share the Template file to Walker and I , we'd appreciate it (PM).

@Walker - Your right about the BT manipulation - A coder can easily block out bad trades to spice up the BT outcome. For example FX voodoo did such that - This ea was decompiled and sure enough there were bad months that were blocked out during a back test to prevent the BT from showing the true story if the EA were to run 24/7. - Others can read more here - https://www.pipinvestment.com/scams/metatrader-backtesting-scam

Anyways, I'm still interested in the outcome of any commercial Bot. Keep us posted Remus =)

I also noticed that on the FFM website, after clicking off, you can get access to a pdf that vaguely states the logic. Its not much to go off and plotting a few trades on the M15 will display more than what the pdf stated.

@Remus, did you make any or optimize the atr setting within the FFM or run the BT at default?
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 17:09 (отредактировано Jan 22, 2012 at 17:23)

   walker36 posted:
   Hi Remus

after BT finished just click 'Open Chart' ( right side of the strategy tester ) this will open the chart and will show up the trades taken on chart.
then right click on the chart and save as template. then any time when you apply this template to chart it will bring the results.


Thanks, I did not know about that, but remember seeing that option, so I'll use it next time!
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 17:14

   walker36 posted:
   'But I will disagree with 'demo is no proof' in this particular case. Considering the EA enters trades at normal times (as opposed to MDP which trades only on volatility spikes, for example), a demo and a live account would have almost the same exact fills.'

if EA average TP > 15-20 pips then demo results will be the same as real account unless slippage and connection issues on real.


Thanks Steve for adding to my comment. Larger TP is essential to what I said, indeed. To be even more clear, for anyone still not believing this, just consider 2 EAs, ones expected to make 10 pips each trade, and one expected to make 1.5 pips each trade (that is the Pips Expectancy, btw, reported by myfxbook in the stats). If your broker slips you 0.5 pips on average, the first EA will lose 5% of profitability, and the second will lose 33%. For higher slippage they'll lose more, with the second going into negative territory fast. FFM is shown to have a 22 Pips Expectancy on the sales page, but I got around 8 on my 2011 backtesting. Even at that value the demo and live should be roughly the same. If not, a broker change is necessary.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 17:20 (отредактировано Jan 22, 2012 at 17:21)

   Alec posted:
   If I understand it well:

The best one can do with a 'decent spread', is a monthly 3% profit.

EUR/USD with fixed 0.01 lots. Right?

Yes, Alec, that's the conclusion you can draw from 2011 BT. And if you look at the very low DD (2.95%), that's actually very good. If you use 0.02, you'll get double each month, for double the DD. Of course, it's just a BT :)

I like my testing with fixed lots to see how the performance is throughout the testing period (otherwise a recent great performance could hide the very bad period that happened 1 year ago, for example). And I used 0.01 for a $1,000 account, since I really see no point in those backtests with million dollar accounts, trading 1000 lots each time - 99% of the traders are far, far away from that, and even most of the brokers cannot handle 1000 lots per trade.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 17:31

   ForexScam posted:
   @Remus, Thanks for this tid bit. If logic is based on M15 it shouldn't take long to figure out - If you can share the Template file to Walker and I , we'd appreciate it (PM).

@Walker - Your right about the BT manipulation - A coder can easily block out bad trades to spice up the BT outcome. For example FX voodoo did such that - This ea was decompiled and sure enough there were bad months that were blocked out during a back test to prevent the BT from showing the true story if the EA were to run 24/7. - Others can read more here - https://www.pipinvestment.com/scams/metatrader-backtesting-scam

Anyways, I'm still interested in the outcome of any commercial Bot. Keep us posted Remus =)

I also noticed that on the FFM website, after clicking off, you can get access to a pdf that vaguely states the logic. Its not much to go off and plotting a few trades on the M15 will display more than what the pdf stated.

@Remus, did you make any or optimize the atr setting within the FFM or run the BT at default?

Yes, I'll do that later today!

The 'revealing the strategy' document (Secret Report.pdf) you mentioned is close to a joke, imo. I'll let everyone to decide its value, but I don't think I would have released a report that seems to have no practical value (for the user).

About optimization: none done, nor planned. I attempted to do it once, MT4 crashed, and I won't try again anytime soon, since there are other areas I need to focus on. The developer said the 'out of the box' settings are optimized, so now I just want to see how good that corresponding performance is!
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Jul 24, 2011   91 комментариев
Jan 22, 2012 at 19:36

   NextLevelForex posted:
   
Yes, Alec, that's the conclusion you can draw from 2011 BT. And if you look at the very low DD (2.95%), that's actually very good. If you use 0.02, you'll get double each month, for double the DD. Of course, it's just a BT :)

I like my testing with fixed lots to see how the performance is throughout the testing period (otherwise a recent great performance could hide the very bad period that happened 1 year ago, for example).

I agree, you did good, that was the best choice for analysis purposes.

Now I'll wait for further developments, from you or from the experts Steve and Will :)

And by the end of the week, I may buy the robot too.
Участник с Jul 24, 2011   91 комментариев
Jan 22, 2012 at 20:42 (отредактировано Jan 22, 2012 at 20:44)
Apparently, Million Dollar Pips ceased to work out at the end of October, so there was a need for their team to come up with something else. MDP was quite profitable for about 3 months -- in very strict conditions.

It remains to be seen for how long FFM will be making pips.

On this *really* early stage, from the 5 accounts I'm watching, there are 2 winners to 3 losers. Being that one of the losers is the only real money account (Birt's).
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 20:47 (отредактировано Jan 22, 2012 at 20:48)

   Alec posted:
   Apparently, Million Dollar Pips ceased to work out at the end of October, so there was a need for their team to come up with something else. MDP was quite profitable for about 3 months -- in very strict conditions.

It remains to be seen for how long FFM will be making pips.

On this *really* early stage, from the 5 accounts I'm watching, there are 2 winners to 3 losers. Being that one of the losers is the only real money account (Birt's).

FFM should not suffer at all from the problems MDP had, if it is indeed trading as the vendor says (and so far seems to be true). To recap, the biggest MDP problem is getting good fills, which is only getting worse as more people are trading it.

And it's too early to say anything about forward performance yet, so I would draw any conclusions. Once we have 100 trades, then we can talk (and the good thing is that the EA seems to be trade as often as advertised, so in very few weeks we could have those 100 trades).

On a related note, I'll do my best to start forward testing on a real money account today.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
walker36
forex_trader_36599
Участник с May 13, 2011   1341 комментариев
Jan 22, 2012 at 20:57
there should be something attractive about William s products. MDP has no vendors live account but Beta tested with on real with many traders. and now Remus, you are going to put it on Beta test for the creator. ( I call that beta test due vendor has not tested on real account if did I did not see ) my advise is to first test it on demo. demo <> live will not be different as generally accepted here at the thread.

but at the end it is your money😄


   NextLevelForex posted:
   

   Alec posted:
   Apparently, Million Dollar Pips ceased to work out at the end of October, so there was a need for their team to come up with something else. MDP was quite profitable for about 3 months -- in very strict conditions.

It remains to be seen for how long FFM will be making pips.

On this *really* early stage, from the 5 accounts I'm watching, there are 2 winners to 3 losers. Being that one of the losers is the only real money account (Birt's).

FFM should not suffer at all from the problems MDP had, if it is indeed trading as the vendor says (and so far seems to be true). To recap, the biggest MDP problem is getting good fills, which is only getting worse as more people are trading it.

And it's too early to say anything about forward performance yet, so I would draw any conclusions. Once we have 100 trades, then we can talk (and the good thing is that the EA seems to be trade as often as advertised, so in very few weeks we could have those 100 trades).

On a related note, I'll do my best to start forward testing on a real money account today.
Участник с Jul 24, 2011   91 комментариев
Jan 22, 2012 at 21:19 (отредактировано Jan 22, 2012 at 21:22)
Good, Remus. When (if) I buy the EA, i may start trading immediately on a real money account -- starting with very small lots, of course. After all, we only have 1 month to decide if the robot is worthy or not.

This leads me to something that is nagging me: Their change of 'store', from Clickbank to Click2Sell. Clickbank, I know they were good for returning the money when needed. They work well like that, policy wise and whatnot.

Now if I lose money with the robot, I will certainly want that 30 days 100% Money Back Guarantee. Anyone has any experience or knowledge of how Clik2Sell deals with these matters?

With Clickbank it was very practical, straightforward, and fast. What if they moved to this 'store' because the previous one *really* gave the money back? Or am I being paranoid?
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 21:29

   walker36 posted:
   there should be something attractive about William s products. MDP has no vendors live account but Beta tested with on real with many traders. and now Remus, you are going to put it on Beta test for the creator. ( I call that beta test due vendor has not tested on real account if did I did not see ) my advise is to first test it on demo. demo <> live will not be different as generally accepted here at the thread.

but at the end it is your money😄


Very well said, Steve ;)

Here are the 2 main things that attracted me to FFM:

- the first commercial EA I ran across, that claims Pips Expectancy higher than 3 (they have those stats on the sales page, and I was very happy to see them; they said 22.9 for EURUSD, while I got 8.5 for my “Decent broker 2011 test” - still good), which indeed would make it a lot less broker dependent than most all other EAs in my portfolio

- if the future results are only half of what is advertised, I don't need to pay someone up to 40% as performance fee. Although I did not give up on that front, and even have some good leads so far, you probably saw that I get a lot of poor systems on that thread (and lately an angry guy who's mad that I showed his martingale to the public)

I'm also selling this EA, so that in case people want to buy through my link, I will get something in return for the extra effort to make my findings public. But that's a distant point 3, so I’m not adding it to the above list. I have lots of complains about FFM (top one being the lack of real money account forward testing by vendor, even for only 2 weeks), but if the overall results look good, those complaints won't matter much ;)
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 21:44 (отредактировано Jan 22, 2012 at 21:46)

   Alec posted:
   Good, Remus. When (if) I buy the EA, i may start trading immediately on a real money account -- starting with very small lots, of course. After all, we only have 1 month to decide if the robot is worthy or not.

This leads me to something that is nagging me: Their change of 'store', from Clickbank to Click2Sell. Clickbank, I know they were good for returning the money when needed. They work well like that, policy wise and whatnot.

Now if I lose money with the robot, I will certainly want that 30 days 100% Money Back Guarantee. Anyone has any experience or knowledge of how Clik2Sell deals with these matters?

With Clickbank it was very practical, straightforward, and fast. What if they moved to this 'store' because the previous one *really* gave the money back? Or am I being paranoid?

You are not being paranoid at all Alec! Withing an industry where SCAM is the primary way of doing business, I think it's more than ok to have all these concerns. Here is my take on this.

Towards the end of last year Clickbank enforced rules that limit ability of vendors to hype their product (https://www.clickbank.com/product_requirements.html, see bullet point I.d; also see Will's blog post on that topic: https://www.pipinvestment.com/software/more-transparency-towards-clickbank-vendors)

Just couple of days ago Plimus cut access to all Forex EA vendors (very radical and un-business-like measure, if you ask me).

So my GUESS is that FFM team had a better option with going with Click2Sell (I did not read Click2Sell's policies, maybe they allow hyping up the same way Clickbank used to). I personally did not have experience with Click2Sell so far, but I know I can call up my credit card company and get a pretty prompt refund if FFM team of Click2Sell starts playing games with me on the 'refund' front.

If you want to buy, I will kindly ask you do it through my link. At the same time I would advice waiting until more more information (from me and other testers) surfaces.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
walker36
forex_trader_36599
Участник с May 13, 2011   1341 комментариев
Jan 22, 2012 at 21:50 (отредактировано Jan 22, 2012 at 21:52)
I will write in general again. not offensive just informative from my point of view:

I had coded and tested many EA s also traded manually.

1- if trade is open more than some hours lets say 2-3 then the trade is open to future news and market effects and most of the time those effects are against the trade. better to have trades as short as possible.
2- when any EA BT ed ( self coded, commercial no difference ) or what ever we see on the chart is happened that way. there is no guarantee of that that will happen same way.
3- ( after all those tests and codings ) eachtick coded EA s are the most dangerous ones regarding to the bar open ones. due the ticks should be the same to catch the past performance which is impossible.

basicly what I am saying is eachtick coded EA s are more likely 'putting a stretch body wear on to elephant', at the end it does split with the first move.

for me bar open EA s are more realistic for the future tests.

that comment can be wrong/wright but my personal conclusion about any EA.

walker
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 22:20
Breaking news! Please listen to this!

I installed FFM on a real money account; the market has been opened for 19 whole minutes, and the account did not double in value! Worse, the EA has not traded yet! What do I do now?...

Joke aside, I was able to setup a real money account with FFM, and will post link later today.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
walker36
forex_trader_36599
Участник с May 13, 2011   1341 комментариев
Jan 22, 2012 at 22:24
am I reading correct, you did not double in 19 minutes!!! EA is useless😂
Участник с Apr 25, 2011   262 комментариев
Jan 22, 2012 at 22:37

   walker36 posted:
   am I reading correct, you did not double in 19 minutes!!! EA is useless😂

Yes, you are reading it correctly! And I checked again, now the market's been open for 35 minutes, but the account it's in the same state! Why is this happening to me!!? No trades, no doubling, no tripping! Maybe I should call my broker? :D:D:D

I did not see any posts with this attitude on this thread, which is actually absolutely great, and felt the need to 'fill in the gap' :)

Anyway, I will take a break now, go outside and get some fresh air (we still have snow), and post more info later, or maybe tomorrow.
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
Участник с Jul 24, 2011   91 комментариев
Jan 22, 2012 at 22:48
Eh eh eh
ForexScam
forex_trader_32776
Участник с Mar 28, 2011   1008 комментариев
Jan 23, 2012 at 05:40
I came across two other accounts with open trades. I compared my generic logic to the opened trades and my broker's data wouldn't have opened a position on AUDUSD (long) while one of the demo accounts did and the other one did not.

Therefore each broker will surely have different results with this EA.

https://www.myfxbook.com/members/jsshear/fast-forex-millions-demo/227987

https://www.myfxbook.com/members/birt/fast-forex-millions/229091
Участник с Apr 25, 2011   262 комментариев
Jan 23, 2012 at 07:27
Here is my real money account, trading with 1 microlot at FXDD:

https://www.myfxbook.com/members/NextLevelForex/fast-forex-millions-fxdd-real/106578

My theory is that the demo and real accounts will have super similar trades, since the price feed is the same and the fills are not going to be a problem for FFM (i.e. slippage will be minimal). So far we have only one trade closed in the new account, and that matches exactly the corresponding trade in the demo account:

https://www.myfxbook.com/members/NextLevelForex/fast-forex-millions-fxdd/228516
He who thinks he can, and he who thinks he cannot, are both right. ~ Henry Ford
You must be connected to Myfxbook in order to leave a comment
*Коммерческое использование и спам не допускаются и могут привести к аннулированию аккаунта.
Совет: Размещенные изображения или ссылки на Youtube автоматически вставляются в ваше сообщение!
Совет: введите знак @ для автоматического заполнения имени пользователя, участвующего в этом обсуждении.