Victor's Strategy | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Comprehensive Fibonacci Analysis by Victor Pershikov, MFTA 113 replies

Clear charts by Victor 15 replies

VJ <Victor's Journal> 0 replies

WRB/MC strategy:Coder needed to semi-automate my strategy 11 replies

Is opposite of losing strategy a winning strategy? 21 replies

Attachments: Victor's Strategy
Exit Attachments

Victor's Strategy

  • #1
  • First Post: Jun 24, 2007 2:51pm Jun 24, 2007 2:51pm
Hi

This is the first strategy i upload here. Please take a look over the EA and the graphic i posted, and tell me what you think about this kind of strategies.
I tend to write EA's that win a little but don't lose at all. I tested this on a 30 min chart from 2002-2007 it wins a little but never loses. Also i tested it month by month from 2005 and never losed. Is my approach ok?

I know that here are lots of guys that know a lot more things about this stuff, please take a look over this and let me know what you think. You can also come with ideas and make it more profitable. (I tested it on M-30, GBP_USD)


P.S This strategy motto: if i could be 100% sure that i can win 1pip from a trade than i'm sure that i've found a pattern in the graphic. If patterns exist, than i can predict in every single moment how will the graphic look after a while. (at least for the near future).

P.S How can i upload reports? I uploaded here only the picture from the report, i can't upload the html file.

Excuse my bad english .

Thanks Victor,
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester.jpg Size: 21 KB
Attached File(s)
File Type: mq4 dupe3vine4_pt100.mq4   3 KB | 1,041 downloads
  • #2
  • Jun 24, 2007 2:52pm Jun 24, 2007 2:52pm
I tested this starting from 400$, M-30, Gbp-Usd
  • #3
  • Jun 24, 2007 3:34pm Jun 24, 2007 3:34pm
That's probably the funniest equity curve that I've ever seen!

Do you have any trading results that you can post?
  • #4
  • Jun 24, 2007 3:44pm Jun 24, 2007 3:44pm
Hi, I tryed a lot of combinations. All look the same strait up . That's why i wanted you guys to take a look and tell me what's wrong.

I tested so: 2002-2007...strait up
2005 to 2007 month by month(ex 2005.01.01-2005.02.01
2005.02.01-2005.03.01
...


)
  • #5
  • Jun 24, 2007 3:46pm Jun 24, 2007 3:46pm
Quoting neamtu_victor
I tested this starting from 400$, M-30, Gbp-Usd

u have 2 warnings on that file...and the indicator dosnt show enythink
u have other info?
Rather miss a opportunity than to get involve in a bad trade!
  • #6
  • Jun 24, 2007 3:47pm Jun 24, 2007 3:47pm
Please try the EA i posted and tell me if you get the same results. If yes tell me what you think about this:

rare and small winnings but 100% sure are better than big and frequent winnings but not 100% sure ? )
  • #7
  • Jun 24, 2007 3:49pm Jun 24, 2007 3:49pm
Salcidon,

Yes i know about the warnings but it's not an indicator it's an expert. You drag it on gbp-usd 30 min chart, set the start-end date and run. Please let me know if it works.

Thanks
  • #8
  • Jun 24, 2007 4:08pm Jun 24, 2007 4:08pm
Looks good. What's the strategy behind it?
  • #9
  • Jun 24, 2007 4:19pm Jun 24, 2007 4:19pm
Hi,

The strategy is very simple as you can see in the EA:

-It looks for this pattern 3 consecutive bars up. If we found this pattern we attept to start a buy order with limit 12*Points.
-We do this if in the last 4 bars from the moment we attempt to start a new order we can find a high that is 20*Points up, so 8 over the limit we want to put.

If this conditions are accomplished we start a buy order, limit 12 points stop-200Points. That stop will never be hit. Was never hit from my backtests in 5 years. (2002-2007).

I tested this using 400$ so i calculated how manny lots i can trade and starting from 400 i use 0.2lots. If you want to trade starting from 10.000 you can set that to 4.8lots.

I posted this strategy as a sample of...never losing a trade. How can one use this to get more profit? Is there anywere a bug? I actuly won't trade this because it's not so profitable,...yet ).

Please let me know what you think, Thanks
  • #10
  • Jun 24, 2007 4:34pm Jun 24, 2007 4:34pm
Quoting nitman
Looks good. What's the strategy behind it?
Resume:

buys= 0.2 lot, (sl=200 pips, tp=12 pips) if:
highest(4days) is greater at least 20 pips from actual ask
and
only white candles on last 5 bars (M30 period)

The code is a mess and whoever did it is not professional (i´m not either)

alex.pt (aka Carllos)
"Hoe ur pips mate?" fxcarllos
  • #11
  • Jun 24, 2007 4:39pm Jun 24, 2007 4:39pm
Quoting neamtu_victor
Salcidon,

Yes i know about the warnings but it's not an indicator it's an expert. You drag it on gbp-usd 30 min chart, set the start-end date and run. Please let me know if it works.

Thanks
i downt get it ... i made it.. compile... drag it... but cant set the start-end date...
i just send u a private message
Rather miss a opportunity than to get involve in a bad trade!
  • #12
  • Jun 24, 2007 6:28pm Jun 24, 2007 6:28pm
Hi, well I'm no programmer, but it looks like you're only buying from the code.
My guess is that GPDUSD has been going up over the last few years (I backtested to '99) so if you're only buying of course you've got a good chance of doing ok.
Especially with a HUGE 300 stoploss

-Good idea though. Not enough people focus on EA's that target price action rather than indicators.


P.S. Looks even better if you change timeframe to H4, with tp=16.
I'm not really convinced as we're tweeking to make a pretty graph..

Regards
Paul
  • #13
  • Jun 24, 2007 7:12pm Jun 24, 2007 7:12pm
Quoting pjimmyinahouse
Hi, well I'm no programmer, but it looks like you're only buying from the code.
My guess is that GPDUSD has been going up over the last few years (I backtested to '99) so if you're only buying of course you've got a good chance of doing ok.
Especially with a HUGE 300 stoploss

-Good idea though. Not enough people focus on EA's that target price action rather than indicators.
Yeah, was going to say the same. GBPUSD has been in a bull run for the past five years, and a code that is only on the buy side is going to have a better chance of winning with that data sample. I'd like to see this tested on something that is a little more sideways (EURCHF, EURGBP) or it trade both buy and sell side.
  • #14
  • Jun 25, 2007 12:39am Jun 25, 2007 12:39am
Quote
Resume:

buys= 0.2 lot, (sl=200 pips, tp=12 pips) if:
highest(4days) is greater at least 20 pips from actual ask
and
only white candles on last 5 bars (M30 period)

The code is a mess and whoever did it is not professional (i´m not either)

alex.pt (aka Carllos)

Yeah, sorry i wrote 3 days, 5 is the corect value. The thing i was trying to demonstrate with this code is that there are "patterns" in a curency graph. IF this thing happens every time on every currency than i'm right. You don't need to know external events in order to predict where will the market be at a moment.

Quote
Yeah, was going to say the same. GBPUSD has been in a bull run for the past five years, and a code that is only on the buy side is going to have a better chance of winning with that data sample. I'd like to see this tested on something that is a little more sideways (EURCHF, EURGBP) or it trade both buy and sell side.

bluebuddha said that is weird that it only buys, no sells, that is true. This is what i found, the pattern happens only for buy, and in order to demonstrate my principle i downloaded the EURCHF and EURGBP history data and posted here the results pircutures, the pattern seems to be the same for those too. (strait up)

Thanks for all your answers, if you have ideas about how to make this more profitable please let us know.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: euro-chf-200-2007.jpg Size: 16 KB
Click to Enlarge Name: eur-gbp-2002-2007.jpg Size: 13 KB
  • #15
  • Jun 25, 2007 7:06am Jun 25, 2007 7:06am
Quoting neamtu_victor

bluebuddha said that is weird that it only buys, no sells, that is true. This is what i found, the pattern happens only for buy, and in order to demonstrate my principle i downloaded the EURCHF and EURGBP history data and posted here the results pircutures, the pattern seems to be the same for those too. (strait up)
.
Hi there

This is a very rare patern. I found it only 16 times in eurgbp 4.1.1999-9.4.2004 (DMY) (excell has cell limit so recent years were out).

Anyway, even so, it´s worth to work around the theme. I was working for past days on my new EA i called it "Realista" (portuguese name to Realistic). This EA works around same subject, thats why i came here to see your thread.
Basicly, we can set one of two things to work around this (percentage of profits we want to achieve at each trade, or pips profit and stoploss).

So, lets take your EA and its settings: TP: 12, SL:200. considering a margincall of 50%, the SL should reflect that margin call. If so, starting with a $1000 account we need 255 trades to reach 1 MILLION.

My guess is, that if we set your EA in severall pairs, and with some luck after 255 trades we would be millionaire.

I´ve built an excel file so you can play with this idea. I´ve been playing with it for some time. I was just missing some sort of a setup. Maybe your pattern is a close one.

Cheers

alex.pt
Attached File(s)
File Type: zip forum - how to make 1 million.zip   56 KB | 819 downloads
"Hoe ur pips mate?" fxcarllos
  • #16
  • Jun 25, 2007 8:29am Jun 25, 2007 8:29am
Hi Carllos,

Thanks for the interest in my post. I will test it some more on big periods,other currencies. After all to find 255 patterns are not so much if it's real and we are 100% sure that can not fail.
Anyway i want to make sure that after my tests i can guarantee that this can not fail. How much it makes is not so important if we are 100% sure that it can not fail .
How is the pattern you found?How it works? Mabe we can bind the 2 patterns and make stronger one
  • #17
  • Jun 25, 2007 9:19am Jun 25, 2007 9:19am
Quoting neamtu_victor
bluebuddha said that is weird that it only buys, no sells, that is true. This is what i found, the pattern happens only for buy, and in order to demonstrate my principle i downloaded the EURCHF and EURGBP history data and posted here the results pircutures, the pattern seems to be the same for those too. (strait up)
Just playing "Devil's Advocate". If it works with several sets of data, then we can do what Carllos suggested: use it whenever it appears on any pair your broker offers.
  • #18
  • Jun 25, 2007 9:57am Jun 25, 2007 9:57am
Quoting neamtu_victor
Hi Carllos,

Thanks for the interest in my post. I will test it some more on big periods,other currencies. After all to find 255 patterns are not so much if it's real and we are 100% sure that can not fail.
Anyway i want to make sure that after my tests i can guarantee that this can not fail. How much it makes is not so important if we are 100% sure that it can not fail .
How is the pattern you found?How it works? Mabe we can bind the 2 patterns and make stronger one
I don´t have a setup. That´s the missing point on my EA. All i have is the MM strategy. Basicly i built it to go from $1000 to $1 Million in about 200-250 trades (smal TP). Entry is now discrecionary, so i set EA to just go long or short as i want to. But, if there is a patern 99.9% correct, i´ll use it.
We shall not forget that we can randomly go long or short 5-10 pips TP, using a 200 SL and be >80% correct all time. If we use TP <= 5 pips we will be correct over 90% time. So, the problem here is.. the 10% where we will fail and the huge number of trades we have to be correct.
A perfect setup (about 99.9%) will take care of the 10%, the rest is MM and opening several accounts with profit. When we reach some profit level we open a new account that starts from zero, and we do it sistematicly. Even after some accounts blow, at certain point the $1 Million will be reached.

Is it clear how i´m trying to implement this?
"Hoe ur pips mate?" fxcarllos
  • #19
  • Jun 25, 2007 10:03am Jun 25, 2007 10:03am
Yes, I understand what you're saying carllos. The sistem has to be 99.9% correct. Lots of backtests have to be done .
How can i automatise the backtests to be done so:

one par set...one big period, month by month,from 10 to 10 days and so on...? Does metatrader have something like that? Or should i do them manualy?
  • #20
  • Jun 25, 2007 10:03am Jun 25, 2007 10:03am
Quoting neamtu_victor
... if it's real and we are 100% sure that can not fail.
Anyway i want to make sure that after my tests i can guarantee that this can not fail. How much it makes is not so important if we are 100% sure that it can not fail.
Neamtu_Victor -

Congrats on finding a rare pattern that, according to you, produces pips.

If it works, by all means, use it.

But you're complicating things - and maybe headed for disaster - by placing so much emphasis on being 100% sure that it cannot fail.

Of course it will fail from time to time. And that's the only thing that's 100% certain about the method.

Secondly, you're not leaving yourself much margin for error as you need about a 95% success rate to break even.

So maybe you can tweak the parameters a little bit and reduce those long odds.

I hope you can find a formula to make the numbers work. But banking on a 100% success rate is a complicating factor that's not going to work.
0 traders viewing now