Phantom 6 System | Page 55 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Jag's Phantom 6 System 295 replies

Phantom 6 - Manual System 8 replies

Phantom Entry on ODL Japan. 2 replies

Tradeview CAD/JPY "Phantom" spike 0 replies

Phantom high, phantom low 0 replies

Attachments: Phantom 6 System
Exit Attachments

Phantom 6 System

  • #1,081
  • Mar 18, 2011 1:26pm Mar 18, 2011 1:26pm
Hi All, and in particular Roundrock,

I am reporting on the DD EA Basket Closures for the week ending 18th March 2011


OOTB settings on EA Phantom6_01DD
100:1 Leveraged Account.
Running only during London and NY sessions.

I think that my old copy of the Phantom6_01DD At 15 mins may have become corrupted. I thus downloaded a new copy form FF

1st Basket: MMTU at 1% @15 Min trade additions
First trade entered at 08.45 GMT on 15th Mar.
Basket closed at 12.42 GMT on 15th Mar.
Trades: 12
Largest Floating DD –0.84%



2nd Basket: - MMTU at 1% @ 1hr trade additions
First trade entered at 11.35 GMT on 16th Mar.
Basket closed at 21.10 GMT on 17th Mar.
Trades: 43
Largest Floating DD –5.34%


EA off the chart on FRI 18 Mar



2nd Test
OOTB settings on EA Phantom6_01DD @1hr trade additions
500:1 Leveraged Account. MMTU at 1%
Running only during London and NY sessions.

1st Basket:
First trades entered at 16.01 GMT on 10th Mar.
Basket Closed at 15.42 GMT on 14th Mar.
Basket Trades = 6
Largest Floating DD = -3.42%

2nd Basket:
First trades entered at 08.17 GMT on 15th Mar.
Basket Closed at 12.43 GMT on 15th Mar.
Basket Trades = 4
Largest Floating DD = -1.76%


3rd Basket:
First trades entered at 19.00 GMT on 16th Mar.
Basket Closed at 06.59 GMT on 18th Mar.
Basket Trades = 10
Largest Floating DD = -5.94%


EA off the chart on Fri 18 Mar after 3rd basket closure

Nicastro
  • #1,082
  • Mar 18, 2011 1:32pm Mar 18, 2011 1:32pm
Hi fellas,

I finally got a version of Phantom6_01R with the addition af a Buy EURCAD trade and a sell USDCAD trade that compiles as bug free.

The two new pairs are shown as Phatoms but they are not trading.

There is obviously some small thing that I cannot see that is preventing these two pairs from trading.

Can an experienced eye be cast over the attached EA so that it will now trade.

Yours in desperation................

G Nicastro
Attached File(s)
File Type: mq4 Phantom6_01xR -withTH_GN.mq4   66 KB | 282 downloads
  • #1,083
  • Mar 19, 2011 8:42am Mar 19, 2011 8:42am
Hey all,

had not much time recently to follow the thread, just wanted to report a bit.

I am still alive on all accounts, the EA survived the EURJPY dip very good. DD was never over 20% ... for me that is very much acceptable. But I took down lotsizes again to 0.01 to avoid more.

I see you guys talking about the CL strategy a lot, when to cut the loosers. What I personally did was to trail the winners. The loosers are kept alive via the DD, but I did not want to turn a winner into red piips again, so I set a trailing stop to 25-35 pips depending on the pairs. This explains the sharp upward spikes in the balance curves.

I had full closures on thursday and friday this week, they took a long time of around 2-3 weeks each. But hey ... it could have been way worse at all ^^

It all comes down to a few clear things in my eyes :

1. Basket trading combined with sharp MM and good account size allows you to withstand a lot.
2. The adding of trades is nice with this EA because of the diversified strategy of 6 out of 14 pairs, so roughly adds trades only after certain turnaround points.
3. I will experiment with other EAs for the entries, but the basket theory shall weigh them out.

One account has been rising from 500 euros to 2000 in around 7 months of active trading since may 2010, that makes a healthy 300% and never gave me too much danger imho.

So, keep rolling cheers xxl205
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #1,084
  • Mar 19, 2011 9:33am Mar 19, 2011 9:33am
thanks for updates, man!

i watch your records everyday. last 2 weeks have been a huge turnaround for you. (simultaneously, my tiny LIVE account was margin called again ) it is great to see all your accounts finally flying out of deep water again, but it does take human intervention according to your experiences. i need to learn that for sure.

Quoting xxl205
Hey all,

had not much time recently to follow the thread, just wanted to report a bit.

I am still alive on all accounts, the EA survived the EURJPY dip very good. DD was never over 20% ... for me that is very much acceptable. But I took down lotsizes again to 0.01 to avoid more.
  • #1,085
  • Mar 19, 2011 11:44am Mar 19, 2011 11:44am
Quoting charvo
thanks for updates, man!

i watch your records everyday. last 2 weeks have been a huge turnaround for you. (simultaneously, my tiny LIVE account was margin called again ) it is great to see all your accounts finally flying out of deep water again, but it does take human intervention according to your experiences. i need to learn that for sure.
Absolutely. I also was close to a heart attack sometimes, but it has been working out like any other recovery that for example Steve Hopwood implements in his EAs (FLTR, Beast etc).

So if Phantom buys its pairs, and the trend goes against you on the chosen one, you simply carry those with you. If it then decides after a certain time again, you get a better medium price ... thats the whole magic. You have to buffer this with your account size / DD. All other strategies use some kind of martingale to recover, but as Phantom keeps lotsizes the same, you maybe adjust them manually ... if you know WHEN to do it ;-)

I had even thought of something else ... why not let Phantom buy the LOOSERS of each hour ? because right now the trend often ends after maybe 3-5 hours, and then turns against you. Thus I was not cutting loosers but I trailed the winners to keep DD lower. If we'd buy the biggest LOOSERS of each hour with a pipstep adjustment, wouldn't this turn out faster then ? Just my cents on that ... what do you guys think ? That would get us closer to the turnaround points of each pair ... but I have no clue how this would work out, and I cannot modify an EA in that way ... *help* ;-)
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #1,086
  • Mar 20, 2011 12:54am Mar 20, 2011 12:54am
Quoting xxl205
I had even thought of something else ... why not let Phantom buy the LOOSERS of each hour ?
One big advantage of testing EAs on Live accounts is, you re constantly trying to survive and innovate. Ideas are flying thick and fast! Good luck to you. Your contribution to 2a development is appreciated.

Please keep posting..
"Put all your efforts in becoming effortless"
  • #1,087
  • Mar 20, 2011 3:46am Mar 20, 2011 3:46am
Hi,

I want to share one of my last tests, the number 23.
Trying XTB broker (one of the best, in my opinion) I needed to understand when and how the margin call worked, so I used (and still using!) MMTU=10.00 and PP=5.00 (yes, ten and five).

It started on 2011/03/03, at 15:00:01 with 100.000 EUR balance, now its P/L is +86,118.80 (floating P/L -17,551.22).

Actually, the lowest margin is 30.98%, the Largest Floating Loss is -95,698.42 and Max Floating drawdown is -74.52%.

With these numbers it's easy to understand that it got more than one margin call (on March, the 13th and 14th) with highest risk to blow up. But.... It's still alive! Very alive.....!

That's why I say this Phantom is amazing.

Live stats: http://www.mt4i.com/users/achab-test-23/stats.

Good Sunday.
/achab
Attached File(s)
File Type: txt test 23.set.txt   < 1 KB | 240 downloads
  • #1,088
  • Mar 20, 2011 4:08am Mar 20, 2011 4:08am
Quoting IamTHAT
Today I closed 2a when profit was around 7% . I had set 10% Profit Pct.
May be we should trail the basket to take advantage of the momentum.
I also thought about this new feature to add, as well as one that gives you automatically the possibility to close the positive position if the cross is 60-70% near to the PP but with less than 5 (or 10) minutes to the 00 time (before to open a new position, I mean).

It's better to add the half of your PP in one hour and start again instead of to wait more hours to get the double, or less, imho.

/achab
  • #1,089
  • Mar 20, 2011 4:16am Mar 20, 2011 4:16am
Hi guys,

I'd like to propose doing a poll about how our MMTU and PP sets.

My opinion starts from the relationship (ratio) between them, that should be >= 2.0 (such as MMTU= 0.5 and PP=0.25).

Apart this relationship, the highest PP will show only the higher inclination to risk.

What do you think about?
/achab
  • #1,090
  • Mar 20, 2011 7:15am Mar 20, 2011 7:15am
Quoting achab
I also thought about this new feature to add, as well as one that gives you automatically the possibility to close the positive position if the cross is 60-70% near to the PP but with less than 5 (or 10) minutes to the 00 time (before to open a new position, I mean).

It's better to add the half of your PP in one hour and start again instead of to wait more hours to get the double, or less, imho.

/achab
Yes, I agree with that absolutely. What I did was trailing the positive trades at 25 for EURCHF, USDCHF and CHFJPY, 35 pips for GBPUSD, GBPCHF and EURJPY due to more volatility.
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #1,091
  • Mar 20, 2011 7:28am Mar 20, 2011 7:28am
Quoting achab
Hi guys,

I'd like to propose doing a poll about how our MMTU and PP sets.

My opinion starts from the relationship (ratio) between them, that should be >= 2.0 (such as MMTU= 0.5 and PP=0.25).

Apart this relationship, the highest PP will show only the higher inclination to risk.

What do you think about?
/achab
What would be a nice feature is to set PP according to the ammount of trades opened. Cause the higher lots you have active on one pair, the higher the "drive" or momentum to reach PP will be. And why not take more from the table if we can ? Means : for each trade open you calculate a PP of 0.025, so that 10 trades will mean PP 0.25, 20 would be PP 0.5 etc ...

For my micro / cent accounts at Alpari UK I have another little problem. I am only allowed to do 100 trades max, and use a total MMTU of 2.0 lots.
So 2.0 lots / 100 trades = 0.02 lots to run the account at its best. If I start using 0.03 lots, I only get 2.0 lots / 0.03 = 68 trades maximum, at 0.04 it is only 50 trades max ! Every one should check their broker limits in that cases, like minimum lots etc to avoid running out of options if the EA gets stretched to the limits / margin calls etc. Thus I will split up my margin on more accounts, means trading 2 times 2k @ 0.02 lots is the same as 4k @0.04 lots but having the advantage of 2x100 trades max instead of 50 on a single account

Demo at Alpari UK allows everything, so this is not comparable to the live account limits. For example the Alpari Pro accounts have much better spreads, but a minimum balance requirement of 20000 Euro and a minimum lot size of 3 lots (!). As the EA is scalable, that would be the same as trading 0.3 lots on a 2000 account ... and I trade my 2000 accounts with 0.02 at max !

Seeing a demo account rising from 100k to 160k in 14 days is very nice and greedy to watch, but would you do it with live money ? 100k @10 lots is same as 10k with 1 lots and 1k with 0.1 lots. I prefer to trade 0.01 with 1k money ... think about it.

If you take 100k and use 1 lot instead of 10, you'd have 6k earnings in 14 days and 1/10th of the risk going broke with a margin call ... I could happily live with 6-10 k a month and simply sit it out for some months/years ^^
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #1,092
  • Mar 20, 2011 9:08am Mar 20, 2011 9:08am
Quoting xxl205
Yes, I agree with that absolutely. What I did was trailing the positive trades at 25 for EURCHF, USDCHF and CHFJPY, 35 pips for GBPUSD, GBPCHF and EURJPY due to more volatility.
Hi guys,

I had tried something like this a while back, not sure if it was exactly what you guys are thinking.

What I did was modified one of the phantoms to close a trade if it reached a certain profit level. Kind of like the opposite of the cut loser option. If I remember I had it set to half the PP. So lets say for example if my PP was set for $30, any trade that reached $15 would close. I really didnt experiment too much with it because it didnt test too well, but I only tried one setting and for only a week or so maybe I should have given it more time.

Not sure about the idea. Is it good to close out the winners and keep the losers going? My trimming lately has been to take out an equal amount of winners and losers in the hope that one or two pairs with jump to get to the PP.

As for MMTU and PP. Still like to use MMTU 1.0 and a very low PP usually 0.1 or 0.2. That may limit the profit alot but I feel it also limits the drawdown. I usually need around 15-20 basket closes per week to make 1% and I usually do, sometimes alot more. Drawdown when on demo is not a big deal, but I always try to think how would I feel about 20% drawdown if it was real. So all my demos I try to trade as if it was real money at stake and use the account size I would try if I were to go real. Im sure xxl205 can relate to that as he is trading a live account.

One other thought, instead of trailing stop on each winning trade. How about a trailing PP. Start off with a higher PP. lets say 3%. As the profit goes up, set the PP to a level below so if there if a turnaround the basket will close with a small profit instead of going into drawdown. Maybe it could be set to 50% of current profit level. So for example if the basket reaches .5 set PP to .25, at 1.0 set PP to .5 and so on, until it reaches that 3% level. This would keep the potential of getting a real good basket profit and keep you from losing out on a decent profit and having to wait out longer for that higher level. Not sure Im a good enough coder to figure that one out but maybe Ill give it a try.

Good luck in your testing.
  • #1,093
  • Mar 20, 2011 9:14am Mar 20, 2011 9:14am
Quoting IamTHAT
Today I closed 2a when profit was around 7% . I had set 10% Profit Pct.
May be we should trail the basket to take advantage of the momentum.
IamThat,

I missed this post of yours. I just posted above about the trailing stop idea. Simple minds think alike LOL. Do you know how to code such a function into the EA?
  • #1,094
  • Mar 20, 2011 7:19pm Mar 20, 2011 7:19pm
Hi, all. First, I am newbie as a trader, but I like programming in C and C++, so I find easy to code on MQL, and I am learning by now. I am sorry if my english is not good enough, but I am spanish.

Well, I like the Phantom trading idea, as well as Basket trading. The problem to me is that we can't backtest it's results and tune up the system before forward testing it. And I am sure you see as a valuable tool if you can backtest it.

First, I did try to traslate the EAs coded for MT4, for running into MT5, but it was a hard an non profitable job, because some functions doesn't traslate directly and others simply are removed. As you know, MT5 allows backtest multicurrency EAs, but its Order management is not as I like (it allows only trade following FIFO management, and not in visual mode).

Then, I have a good idea: why not backtest on MT4, but reading the rates of the other currrencies from a file? I am coding an EA that reads the historical data from a pair and writes three data for every tick: DateTime, Bid and Ask. Unfortunately, this EA can only work on MT5, because it's the only one platform that downloads historical data on every tick and with different Bid and Ask data. The EA can be launched on backtest, so writing all the historical data from each pair doesn't spends more than a few minutes.

This is just the first stage. From here, I have two options:
a) Re-code the main EA (Phantom EA, by instance), in the way that every call to the function MarketInfo(...), changing as another function that reads the file wrote previously.
b) The more elegant. Create an indicator the reads the files and serves the data to the main EA. We can, too, show the data in a beautiful way, as bars or candles.

What do you think about it? Some better ideas? Thanks.
  • #1,095
  • Mar 20, 2011 11:01pm Mar 20, 2011 11:01pm
Quoting thorn
IamThat,

I missed this post of yours. I just posted above about the trailing stop idea. Simple minds think alike LOL. Do you know how to code such a function into the EA?
Simple mind? lol. It's a piece of shit most of the time.
What I have been thinking is about using mptm. presently I am testing another EA from T101 thread - AccountProfitProtect EA. If it works I'll let you know. There are many such EAs around here.
"Put all your efforts in becoming effortless"
  • #1,096
  • Mar 20, 2011 11:07pm Mar 20, 2011 11:07pm
Quoting thorn
Hi guys,

What I did was modified one of the phantoms to close a trade if it reached a certain profit level. Kind of like the opposite of the cut loser option. If I remember I had it set to half the PP. So lets say for example if my PP was set for $30, any trade that reached $15 would close. I really didnt experiment too much with it because it didnt test too well, but I only tried one setting and for only a week or so maybe I should have given it more time.

Not sure about the idea. I
I tried this too. Didn't work. Basket Profit trailing is what I am trying. Even single closure of 5% a week or 10 days should suffice. Such runs may not be that difficult to get. When trends are favorable we should exploit. Waiting for another hour may be too late.
"Put all your efforts in becoming effortless"
  • #1,097
  • Mar 21, 2011 1:49am Mar 21, 2011 1:49am
Quoting xere
Hi, all. First, I am newbie as a trader, but I like programming in C and C++, so I find easy to code on MQL, and I am learning by now. I am sorry if my english is not good enough, but I am spanish.

Well, I like the Phantom trading idea, as well as Basket trading. The problem to me is that we can't backtest it's results and tune up the system before forward testing it. And I am sure you see as a valuable tool if you can backtest it.

First, I did try to traslate the EAs coded for MT4, for running into MT5, but it was a hard an non profitable...
Hi xere and welcome.

I quote you but my answer is for anyone is following this thread.
Of course it would be nice to backtest but:

 

  1. The data will be different from one broker to another;
  2. The demo accounts are almost always different from live ones;
  3. Some broker has different type of live accounts;
  4. Spread is always different (even if the accounts have floating or fixed spreads);
  5. We have tested that even if you start the same EA, with identical setting, in the same moment, the results will be always different (Stochastic result).

So, we can say that, just in case you'll code the Phantom's backtests, you'll get only an idea about what the live trading will be.

If you are concious of it, it's ok, backtests will give us a better idea of how it works and why, but it will not be a panacea for sure.

In my personal opinion, it would be better to spend time and resources to translate it on MT5, because - before or after - we will have to go there. And this is sure.

Well, it's time to go for me.
Profitable week for everyone.
/achab

  • #1,098
  • Mar 21, 2011 4:16am Mar 21, 2011 4:16am
Well, thanks for your reply.

I should say that with the method I am coding, each one should build its own history, with its own broker data, and having account of the spread too (remember that I am building the history from MT5 due to its history contains different data for Ask and Bid rates). Then, this two question shouldn't be a problem.
  • #1,099
  • Mar 21, 2011 4:42am Mar 21, 2011 4:42am
Here you have my HistoryBuilder.ex5
Copy it to Metatrader 5MQL5ExpertsAdvisors and compile it.
Then open MT5, and in the Strategy Tester window, choose what pair and Period you want to build your ask-bid data, then click Start. When the backtest will be finish, you must have an EURUSD.csv (by instance, if this was your pair chosen) in the directory TesterAgent.....MQL5FilesHistory

Repeat the process for every pair you need. You can open the file in order to see what data is wrote on it: Datatime, Ask, Bid for every tick.

Next I must re-code the bot or made first an indicator that reads this data.
Attached File(s)
File Type: zip HistoryBuilder.zip   < 1 KB | 183 downloads
  • #1,100
  • Mar 21, 2011 5:00am Mar 21, 2011 5:00am
Quoting IamTHAT
Simple mind? lol. It's a piece of shit most of the time.
What I have been thinking is about using mptm. presently I am testing another EA from T101 thread - AccountProfitProtect EA. If it works I'll let you know. There are many such EAs around here.
OK, Ill check it out. I remember that thread, from years ago. Never could wrap my mind around how he actually traded that system.

You are right, we should take advantage of those high flying baskets when they happen. But still be able to take some profit if they turn south.
7 traders viewing now