Magix, Martingales, and Market Myths | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Neural Networks: Myths And Reality 19 replies

Martingales, Anti-Martingales, etc. 16 replies

Martingales: Forget Technical and Fundamental Analysis 46 replies

Demo Trading, Martingales, and Real Trading 4 replies

Boomerangs Martingales and Grids 1 reply

Attachments: Magix, Martingales, and Market Myths
Exit Attachments

Magix, Martingales, and Market Myths

  • #61
  • Aug 28, 2014 5:25am Aug 28, 2014 5:25am
Quoting realjumper
{quote} Ok thanks....that's pretty much what i thought,.....which goes to confirm what I was saying before, that those who use Martingale as a 'never lose' strategy, will do exactly that...lose....in the end. For those with experience, it can be a very useful tool

A lot like any other tool available in forex.

Some will try it and fail. Others will try it and succeed.

I don't base my opinion of others on the system they use...but I don't feel much pity for their losses if they walk into anything, blindly.
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #62
  • Aug 28, 2014 5:32am Aug 28, 2014 5:32am
Quoting Magix
{quote} A lot like any other tool available in forex. Some will try it and fail. Others will try it and succeed. I don't base my opinion of others on the system they use...but I don't feel much pity for their losses if they walk into anything, blindly.

Same......

Cheers.
Doing what you like is Freedom. Liking what you do is Happiness.
  • #63
  • Aug 28, 2014 12:57pm Aug 28, 2014 12:57pm
As per heated discussion in the other thread about Martingale, I will be following your TE closely with great interest! Best of luck
  • #64
  • Aug 28, 2014 1:41pm Aug 28, 2014 1:41pm
Quoting jmn5611
What is the definition of Martin for the sake of this discussion?
If I may, I would quote Ryan Jones (The Trading Game, chapter 3) who defines it simple as "The Martingale [category] simple states that as the value of an accoun is decreasing, the size of following trades increase".

But this definition does not state if we are talking about a tool, a strategy or a money management technique (his opposite, the "anti-martingale" is often used as that) ... what you do with it is what states what is in the end.

For the sake of discussion we have also accepted cost averaging as another technique/tool that can be defined as as subcategory of the martingale family.

So, if we accept it so, there other techniques to be considered too part of the "family": Pyramiding.

And if the lot size is mantained equal along entries (instead of increasing it) some people refer to it as "scaling into a trade" as far as they are adding to a winning position.

And if we also accept scaling into a position as a valid sub-species, another favored technique use to stage the entry int a position by means of waiting for a retracement is also para of the same thing ...

What I'm aiming at is that at the end many people are using derivatives of the same technique being the differences the relative increase/decrease of the size of the entries and/or the technique to select the levels where those entries are established ...

Magix has selected a derivative of SWB GRID as an example of the risk involved ... as the problem with these systems is to accept the risks involved.

He asked a question: "The first thing, I want you to state why this is or isn't a Martingale."

I know I'm going to be crucified but, no, this is not a Martingale as the number of levels has been limited.

It is only a multi-staged trade with target profit ($1) that will use up to 5,12 lots for each trade. The hard stop loss is the full account size independent of the accumulated profits. Risky as hell ... but with a risk limited to the size of the account.

Let me add some salt and pepper to the discussion:

  1. Which are the differences between the grid technique Magix is using in that EA and the rescue/money management techniques explained at FTI's Technical Analysis Fallacy (http://www.forexfactory.com/showthread.php?t=57639) ... I don't see anybody screaming "martingale!!!!!" there ... It is perhaps the Zen approach?
  2. Also ... the nice ALR technique being explored here (http://www.stevehopwoodforex.com/php...forum.php?f=81) is not shouted at as "martingale!!!!" in a forum notoriously known as anti-martigale ... perhaps the "hedge" flavour?

Cheers,

Sergien

Massive trading attack!
  • #65
  • Aug 28, 2014 1:52pm Aug 28, 2014 1:52pm
If we have an EA that uses martingale and the EA is set to close all of those trades at 3% DD, how can martingale be a problem.
  • #66
  • Aug 28, 2014 2:25pm Aug 28, 2014 2:25pm
Quoting sergien
{quote} If I may, I would quote Ryan Jones (The Trading Game, chapter 3) who defines it simple as "The Martingale [category] simple states that as the value of an accoun is decreasing, the size of following trades increase". But this definition does not state if we are talking about a tool, a strategy or a money management technique (his opposite, the "anti-martingale" is often used as that) ... what you do with it is what states what is in the end. For the sake of discussion we have also accepted cost averaging as another technique/tool that can...
Sergien, FTI uses Dollar Cost Averaging and no SL, but he uses low positions in regard to an account size....that is all, but I see it as a proven method by him

Zen approach he advocates is there because you will need it if you want to open highly leveraged position to average in and save if there is anything to be saved...
  • #67
  • Aug 28, 2014 2:38pm Aug 28, 2014 2:38pm
Quoting sergien
I know I'm going to be crucified but, no, this is not a Martingale as the number of levels has been limited.
In real life all Martingales are limited, so if you are going to use this definition then Martingale doesn't actually exist outside mathematical theory.
  • #68
  • Aug 28, 2014 3:07pm Aug 28, 2014 3:07pm
Quoting merquise
{quote} Sergien, FTI uses Dollar Cost Averaging and no SL, but he uses low positions in regard to an account size....that is all, but I see it as a proven method by him Zen approach he advocates is there because you will need it if you want to open highly leveraged position to average in and save if there is anything to be saved...
Yup. My point is he is using a Fibo-based levels grid with increased levels to "rescue" a position at break-even or little profit (i.e. $1). And in his own words due to his account size he is willing to iterate this up to multiple levels ... (that's me being ironic).

Quote
In real life all Martingales are limited, so if you are going to use this definition then Martingale doesn't actually exist outside mathematical theory.

... exactly my point! Martingale can only exits in the event of an unlimited account.

If your account is limited your need to compromise if you are willing to use a martingale-like approach.

If the maximum risk per trade (understanding trade as the sum of orders your averaging-down grid is going to open) is not stablished in advance then the risk is automatically defined as the total balance in the account.

Unexperienced traders don't want to define this risk level so it is defined for them even if they don't want to aknowledge it.

These techniques are tools ... but you have to adhere to basic safety principles before using a tool (such as limiting your risk).

Cheers,
Sergien
Massive trading attack!
  • #69
  • Aug 28, 2014 5:45pm Aug 28, 2014 5:45pm
Quoting jmn5611
{quote} I coded an EA like this once, it was an accident. It used a BB trigger in the direction of the monthly trend only. It had a brilliant 45 degree equity curve, with just one spike of serious drawdown. It traded 0.01 lots right across the board and waited 20 pips or two hours between signals for a TP of 20 pips. It was on the EURJPY.

Well...fuck this EA...

If it isn't going to blow up an account on 1 pair, maybe I should add more???

The pair selection??


If I went with a single currency, concurrency...EURUSD already in play...

Suggestions?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #70
  • Aug 28, 2014 5:46pm Aug 28, 2014 5:46pm
Quoting sergien
{quote} Yup. My point is he is using a Fibo-based levels grid with increased levels to "rescue" a position at break-even or little profit (i.e. $1). And in his own words due to his account size he is willing to iterate this up to multiple levels ... (that's me being ironic).
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #71
  • Aug 28, 2014 6:02pm Aug 28, 2014 6:02pm
Quoting sergien
Magix has selected a derivative of SWB GRID as an example of the risk involved ... as the problem with these systems is to accept the risks involved. He asked a question: "The first thing, I want you to state why this is or isn't a Martingale." I know I'm going to be crucified but, no, this is not a Martingale as the number of levels has been limited. It is only a multi-staged trade with target profit ($1) that will use up to 5,12 lots for each trade. The hard stop loss is the full account size independent of the accumulated profits. Risky as hell...

I still maintain that a "True Martingale", cannot be effectively run as applied to forex...

So...

Your Crucifixion...ain't gunna happen!

The idea that a single application has to be absolute without the use of applied knowledge, isn't a very effective use of previous data and statistics.

Seriously..if all martingales were a complete fail...why would anybody talk about them still?

Is it the lottery?

We randomly pick numbers and cross our fingers??

I simply cannot subscribe to this notion.

Are they a gamble?

Well...Yeah...they are.

When you take human interaction out of the equation, we are counting on 100% on automation. This, in itself is a gamble. We are under a specific notion that we are trading in a utopic scenario...

This will simply never happen.

So, these really boil down to, as best as will ever be, a Game of Chance.

But does this really mean that we can't load the dice or stack the cards in our favor?

With what we have been working on...when you see a trade based on an M5 logic, does it actually make sense on an hourly, H4 or daily chart...or is it completely random? What about any other time frame traded?

Is market so random that we can actually gear into the random..or is it so set in basic structure on a repetitious area that we can capitalize on retraces?



Maybe success on any grid or martingale strategy simply follows one of Newton's Laws of physics in that any action, requires an equal and opposite reaction?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #72
  • Aug 28, 2014 10:12pm Aug 28, 2014 10:12pm
Quoting Magix
{quote} Well...fuck this EA... If it isn't going to blow up an account on 1 pair, maybe I should add more??? The pair selection?? If I went with a single currency, concurrency...EURUSD already in play... Suggestions?
Over exposure is always a great idea when trying to blow an account. Load it up for more EUR and USD pairs, so when the spike comes justice will be sure and swift!
If you are good at something, never do it for free--Joker
  • #73
  • Aug 28, 2014 10:41pm Aug 28, 2014 10:41pm
Quoting sergien
{quote} If I may, I would quote Ryan Jones (The Trading Game, chapter 3) who defines it simple as "The Martingale [category] simple states that as the value of an accoun is decreasing, the size of following trades increase". But this definition does not state if we are talking about a tool, a strategy or a money management technique (his opposite, the "anti-martingale" is often used as that) ... what you do with it is what states what is in the end. For the sake of discussion we have also accepted cost averaging as another technique/tool that can...
I personally think a Martin can work. Here's my theory.

If you take an EA and trade it with the monthly trend( Gann and all the other legends went with monthly trend) then you will only blow your account if there is a major change in trend, or the EA trades too often. If you limit the EA to a very specific signal like the bounce of any momo indie in the direction of the main trend and limit the EA to certain amount of positions it can add per session or day, you can win.

Let's say the average monthly range of E/U is about 350 pips. You tell the EA to trade to not take the next trade unless there is an 18 pip gap between trades, and no more than two positions per day, on the trigger. So we need a trigger that is 18 pips apart, max of two trades per day. Take profit is when the total position is in profit by one dollar.

The EA is going to blow up sooner or later as sooner or later there will be trend reversal or a very deep pullback, but I think the odds are in your favor as you can withdraw profits at regular intervals.

Okay, you can now debunk my theory.
If you are good at something, never do it for free--Joker
  • #74
  • Edited 1:26am Aug 29, 2014 12:38am | Edited 1:26am
Quoting jmn5611
{quote} I personally think a Martin can work. Here's my theory. If you take an EA and trade it with the monthly trend( Gann and all the other legends went with monthly trend) then you will only blow your account if there is a major change in trend, or the EA trades too often. If you limit the EA to a very specific signal like the bounce of any momo indie in the direction of the main trend and limit the EA to certain amount of positions it can add per session or day, you can win. Let's say the average monthly range of E/U is about 350 pips. You tell...
Are you aware that what you have defined is equivalent to do swing trading with target profit of $1, with a initial stoploss defined by the size of the account and a trailing stop loss also defined by the size of the initial funds of the account?

Or let me put another way ... if I say that my trading capital is $100k and I fund an account with $1k5 to trade your idea (1.5% of my trading capital), knowing that statistically I have a 75% win ratio and a 1:5 risk/reward ratio if sistematically the account is reset after 4 weeks, what will you think?

I'm not debunking your theory ... actually what you have defined is one of the most basic escenarios normally used when using EA grids.

Yes, I think your idea can work ... if managed properly. And if you are fully aware that a percentage of the times you start the EA it is going to obliterate the account.

My personal opinion, and of course it is only an opinion, is in this case the approach to this tools has to be fractal. If you just see one account running a martingale-like EA as one of your trades, with proper money management, risk management and profit management (I know, too many management, but, hey! I didn't say this kind of EAs didn't need to be managed), and not as your only trading investment, it can work.

(Only one word about money management on marties ... normally it should be done at the account level, not at the order level, so it is the relative size of the accout to the overall trading capital, nor the operative used to fund the account, what defines the account funding a.k.a. the risk level, and then is the account funding what defines the intial lot size level ... not the other way around)

As is has been previously stated, not a tool for beginners.

Cheers,
Sergien
Massive trading attack!
  • #75
  • Edited 12:12pm Aug 30, 2014 11:38am | Edited 12:12pm
My martin theory: (AUD/USD)

The quick math is that you'd have $126 in margin cost against your balance to enter the trades ($1874)

At your highest entry of 9548 you have about $31 in drawdown ($1841)

adding the .32 trade ($1777)

Looks like about 7.1% margin at risk once you take the highest short trade drawdown included.

You'd still have 2820.6 pips to a margin call, or, roughly, 1.2340 AUD/USD level. (This assumes no stop loss...and really, if that 2800+ pips happens so fast you can't do anything about it, you've probably got more pressing issues than the status of your FX account anyway...)

Hmm...so....how unrealistic is a pure martingale really?

Currently the top of the short targets is a skoash under 9300 (9297 for those want real precision..9130 if you really want to be a hold out but we'll stick with conservative figures...you know...just for skeptics sake..)

So, this batch of trades can be closed there with a reasonable gain or can be held for a bit if you wish...All in all your worst entry standing on it's own will be about 150 pips to the good...but, averaging all of those to your true break even point...


https://www.evernote.com/shard/s22/s...y-TrackVia.png

Predicted result:

https://www.evernote.com/shard/s22/s...D_08252014.png

Target at .9250:

https://www.evernote.com/shard/s22/s...D_08252014.png
"Holy Grail" exists - accepting where is the first step.
  • #76
  • Aug 30, 2014 12:05pm Aug 30, 2014 12:05pm
Quoting sergien
{quote} Are you aware that what you have defined is equivalent to do swing trading with target profit of $1, with a initial stoploss defined by the size of the account and a trailing stop loss also defined by the size of the initial funds of the account? Or let me put another way ... if I say that my trading capital is $100k and I fund an account with $1k5 to trade your idea (1.5% of my trading capital), knowing that statistically I have a 75% win ratio and a 1:5 risk/reward ratio if sistematically the account is reset after 4 weeks, what will you...
I dunno, I think a newbie could learn a lot from a Martin. As long as the newbie understands that any monies used to trade should be considered hooker, booze or Vegas money, then s/he will be fine. Let them work on Martins till they get rich or they get tired, whichever comes first.
If you are good at something, never do it for free--Joker
  • #77
  • Aug 30, 2014 12:09pm Aug 30, 2014 12:09pm
Quoting VaBikePacker
My martin theory: (AUD/USD) The quick math is that you'd have $126 in margin cost against your balance to enter the trades ($1874) At your highest entry of 9548 you have about $31 in drawdown ($1841) adding the .32 trade ($1777) Looks like about 7.1% margin at risk once you take the highest short trade drawdown included. You'd still have 2820.6 pips to a margin call, or, roughly, 1.2340 AUD/USD level. (This assumes no stop loss...and really, if that 2800+ pips happens so fast you can't do anything about it, you've probably got more pressing issues...
I think it depends on the pair. I would use Martin on a sleepy pair like E/G.

If I were trading good money I would never use a Martin on a pair like G/N, as it can easily move 1000 pips in one direction.
If you are good at something, never do it for free--Joker
  • #78
  • Aug 30, 2014 12:14pm Aug 30, 2014 12:14pm
Quoting jmn5611
{quote} I think it depends on the pair. I would use Martin on a sleepy pair like E/G. If I were trading good money I would never use a Martin on a pair like G/N, as it can easily move 1000 pips in one direction.
Well, I think it can work on any pair - the key would be understanding extended range of the pair and adjusting the trade spacing accordingly.

Understanding extended range also contributes to determining the maximum volume of trades taken.

I was just looking at the batch of data I have on the GBP/AUD so far...haven't calculated like I've shown here on AUD/USD, but, it's doable...just a wider range...
"Holy Grail" exists - accepting where is the first step.
  • #79
  • Aug 31, 2014 5:02am Aug 31, 2014 5:02am
Guys do not forget about weekly and monthly pivots...they present excellent opportunities for breaking even or a small profit as the price more often than not retraces around them
  • #80
  • Sep 4, 2014 9:22am Sep 4, 2014 9:22am
Here is martingale at it's finest, high leverage, 1:500, no stop, and stepped within a few pips :
http://www.myfxbook.com/members/Magi...doo-all/917331
0 traders viewing now