Blackbeard's Grid

Blackbeard is a Martingale variation that I developed that keeps drawdowns extremely low. I do not trade it yet, but wanted to share it to the FF community.

Basic Martingale:

To refresh you on basic Martingale, suppose my TP is 100 pips and my SL is 100 pips with a 1 lot trade. I lose the trade, and now double up on the next trade with 2 lots. I continue doubling after every loss until I either TP or bankrupt my account. Let's look at the drawdown that would occur:

Trade 1: 1 lot -100 DD -100
Trade 2: 2 lots -100 DD -300
Trade 3: 4 lots -100 DD -700
Trade 4: 8 lots -100 DD -1500
etc....

Martingale Hedge:

See how lower DDs work with a Martingale Hedge, described here:

http://www.forexfactory.com/showpost...5&postcount=19

As the price moves back and forth across the channel, the drawdown is much lower because only half the floating losing positions can effect your equity at a time. This is because no positions are ever closed until TP.

Trade 1: Long 1 lots -100 DD -100
Trade 2: Short 2 lots -100 DD -200
Trade 3: Long 4 lots -100 DD -500 (note that T2 is not included because it is at 0 when T3 is at -100)
Trade 4: Short 8 lots -100 DD -1000 (note that T1 and T3 are not included because they are at 0 when T4 is at -100)
etc....

Blackbeard's Grid:

If you understand everything to this point, I will now show you the full advantage of the Martingale Hedge described above. The twist to Blackbeard is to have the TP at 200 instead of 100.... Now you can trade the hedge like this:

Trade 1: Long 1 lot -100 DD -100
Trade 2: Short 1.5 lots -100 DD -150
Trade 3: Long 1.25 lots -100 DD -225 (note that T2 is not included because it is at 0 when T3 is at -100)
Trade 4: Short 1.88 lots -100 DD -338 (note that T1 and T3 are not included because they are at 0 when T4 is at -100)
etc...

We are Martingaling with a DD of -338 vs. a DD of -1500 in traditional Martingale. Just set one Blackbeard pointing long and one pointing short with a TP of 100, and you will TP 100 after the first 100 pip move in either direction.

Remember that the TP on the losing Grid is 200, so now you can open an opposite position as the price move 100 pips toward the breakeven TP.

No matter how you dress it, all you're really saying is "hang on to losing positions forever", aka...how to lose. One day that range will end and your losing position will bankrupt your account. Also, you didn't even touch on how to define the range in the first place.

The range is 100. I am using Blackbeard as the backbone of a strategy I am working on. There are extreme cases where the price touches each side of the channel for extended periods. I am exploring how to mitigate this problem.

Here's a simple spreadsheet that demonstrates the improvement of Blackbeard (left) over standard Martingale:

I must be missing something here. When does the "enlightening" thingy kick in?

When you get to break-even?

How can you call 9 months ago 'recent'?

How do you know in advance the channel will be 100 pips?

Do I need a $300,000 account so trade No 1 Will be $1.00 (just in case it bounces 16 times)?

Roy

The spreadsheet is comparing my Blackbeard (left) with standard Martingale. It is meant to show Blackbeard's improvements over the standard... that after 15 losses you would have been down over 3 million pips with the standard, but with Blackbeard only 29,000.

The channel can be any size you feel comfortable with. It's the size of the stop loss you choose. All that matters is that the grid's take profit is twice as large as its stop loss.

Suppose you opened up two of these grids at once.... one long and one short. You take profit as one hits 100 (the size of the grid's stop loss,) but now the other has started it's drawdown cycle. Once it breaks even, you have banked your profit.

Why nitpick "9 months ago" isn't recent?

For a volatile pair like GBP/JPY, 1000 pip target (vs. 200) might be okay. Bank the 100 on the winning grid, and then once the price moves 1000 away from the losing grid, close for breakeven.

The benefit is the hedge changes from:
T1: 1 lots
T2: 1.5 lots
T3: 2.25 lots
T4: 3.38 lots
etc....

to:
T1: 1 lots
T2: 1.1 lots
T3: 1.21 lots
T4: 1.33 lots
etc...

So think about this..... you can place a bet with extremely high leverage that GBP/JPY won't touch for more than say, 15 times before it reaches 1000 pips away. If right, the account would nearly double.

What this means is that as the price is chopping back and forth through the grid, the draw down is kept extremely low (for example, -133 after 4 touches, but really -33 because you already bagged the 100 on the other grid) Assuming GBP/JPY keeps up its pace, the price will move 1000 away eventually. You could open up two grids at every 100 pip interval. I think a swap free account would be necessary to make this work.

Thank you for sharing these interesting, though perhaps not popular, ideas!

You have a done a good job thinking critically about ways to make the drawdown less and the system profitable.

Could you post some more diagrams or charts to help explain your ideas? Thanks tdion.

Glen

if we open two position buy and sell for start..and then if the price move down to the lower boundary..we close the sell position (if range 50,our margin add 50) the buy position that not close still -50,and we open again the sell position and if the price bounce again to the top we close the buy position (margin added 50 - total 100), the sell position that not close still -50....if the price bounce again.. we get the pips from the bounce...
maybe this trik can be join or colaboration with the earlier strategy...

If this strategy is good, then you have yourself a coder.

if it is started with winning position, will you made this beard grid?

since it is started with loosing position, the accumulated pip between the buy & sell may gives unnecessary burden to the added lot. i guess if it is started with winning position, the lot size can even be decreased accordingly based on how far it placed from the last position.... so it can press on beyond the Break even point.

edit: ooo i get it now, if the 1st position is win, case close.... this strategy is mainly for covering lossing position only..... i think

Gwan you start with two grids... one pointing long and the other short.

The market decides which one TPs....

So now you are at breakeven.... You will draw down if the price chops between the channel boundaries of the losing grid....

The beauty of it is that the market can go long or short on the losing grid.... Also, you can TP BEFORE the market hits the TP.... Your profit will be cut, but you will still make some $.


I'm not advocating this strategy, but isn't it worth exploring?

I think you may have something here. Instead of opening a sell, how about a buy?

If a chop occurs, you TP the buy.... now you can close the all positions with profit.

If no chop, it's just another hedge side-by-side with the original. No big deal. The logic here would be to make profits off the chops.

When there are no chops, your position eventually closes out at break even.

Anybody got an opinion on this?

TD,

Would you mind going into a bit better of an explanation. I'm following here with great interest in hopes of coming to a positive conclusion.

I just have a hard time understanding what you are saying as I think my terminology may be different than yours.

Can you please provide a swift example? Something that shows where your buys/sells would be placed and what your TP's would be?

Thanks.

.... somehow it is resemble with reversing trade dirrection, but in more agresive manner,

So this thread died? I thought it sounded like a good idea .

I am working on other stuff these days.

Anything good worth sharing?

Most people wouldn't care about what I'm working in. I wrote an AOL instant messenger client to send me IMs when my EA makes trades. I like AIM better than SMS because it hits my cel immediately. If the server goes down, I know fast.

The best "systems" I've researched lately can be found here:

http://www.forex-tsd.com/commercial-...hebest-ea.html

Also, I'm interested in the HAS method as WHTenn claims to have over 20 winners in a row this past week. WHTenn has been around awhile, so I think it's worth a look.

Good luck.

I was thinking over the same idea some months ago and I was just about to go live with it when I made a test using real numbers and I knew where the problem is:
It assumes 2 zero lines while in reality the pips difference between them is not zero but the width of the grid wich the losing position is passing and adding up to your negative balance till u hedge

Driving from Walmart home at 3 AM, I got the plan together for this system.

The first few posts are the rough draft.

The final idea is as follows:

Set up a grid of martingale hedges side by side. Semi-martingale, as TradeFXPro says. The theory is very good for this idea so read on and give your input. The only real problem I see is the negative swap on the shorts, but given GBPJPY's characteristic, it might not be a problem.

This is only GBPJPY.

The stop and reverse for each martigale channel is 50 pips, with 1000 pips take profit. The position size is 0.1, 0.1, 0.1, etc... until you have drawn down over 1000 pips in that particular channel, then it escalates to 0.2, 0.2, etc.... so you are looking at 30 bounces or so before you escalate to 0.3 (drawdown 2000 pips in the channel)

When you win a trade, you don't win much at all.

BUT

You have several channels open at one time, and assuming GBPJPY moves 1000 pips often, you will be closing channels out frequently.

Is this making sense?

The benefit is allowing the market to go where it wants, while you capture small gains in the process.

The graph below shows a series of channels (50 pips apart) and the price ranging around the grid. Green means close and buy within that channel, red means close and short within that channel. The chart doesn't show any take profits.... in fact, the price would need to move up 20 channels for the bottom grid to close for a take profit.

I really like the idea because it removes prediction from the trading equation, and is dependent only on volatility. You will need to take this idea further on your own, compensating for swaps and rollover.

The system would be a nightmare to trade and would need to be automated.