Dreamliner and Shads:

My understanding of the system is that each day it will hit either The Buy Take profit and consequently the Sell Stop loss or the opposite which make the demo days neutral +30 - 30 = 0 but you save the spread.
So I think adding the demo days or not do not make a difference unless it hits just both Stop loss or never reach either Take Profit on a single day.

Dreamliner I think in your calculation you added somehow pips from the demo days and that is why I have a different profit (either 210 or 270 or 330 pips but not 490 pips).

Now looking at daily charts we can clearly see the trend stays more than a couple of days at a time which make this anti-Martingale a winner on most weeks.
If some days it hits both SL but not TP I guess we could just start again by doubling the winner on those occasional days.
Also If it never reach any TP we could just adjust the lot size so the next day if it hits TP the serie will close with a pip step profit.

I will be out today as I am still on holiday but I am following this thread closely and back in 12 hours.

Sunwest

Thanks for your thoughts here, SunWest, you may be right in all. Does anyone know what pair Jimbo originally used in these results? I'm forward testing, starting tonight on Eur/Usd.

Glad to see this thread popping up and the original Jimbo's thread continuing.

Couple of inputs here. Couldn't make out what Jimbo was trading it originally on, but EUR/USD may not be the best candidate due to the lack of volatility in some days. With EUR/JPY or USD/JPY for example, the spread is only 1 pip higher but on most days you get much more volatility so more likely to hit your profits in either direction. So you can in fact increase the targets and stops by proportio to say 40 or even more.

A trend filter would help, but not really necessary in that you're relying on this lot increment to profit. Lastly, the initial demo of 1 buy 1 sell can just be removed all together. You make 0 pips and you pay the spread, why not just look at the previous days and what would of happened, to determine which side you double up on.

Thanks Sunwest for the files

Best,

doragio

I started forward testing on 14 pairs with .1 lot looking at the previous days winners using 30 T/P & S/L. 4 have already hit T/P in less than 20 minutes. I am also considering starting over if the trade max reaches .8 lot on a loss. Any thoughts?

Hi Doragio,

Yes, I am making the first trades demo trades, and every trade after the completion of a series.

I appreciate your comments on different currencies, I'm also testing with USD/JPY.

Had another thought while was taking a break and thought I'd just throw it out to this thread here. So here goes.

With this anti-martingale strategy, you effectively alternate the entries so as to avoid the harsh d/d's that a strong trend will bring you. What got me was why are we even doing an hedge type all together.

Here's what I mean.

When we start with the first doubling sequence, it goes something like this.

Buy 1 Sell 2
Buy 4 Sell 1
Buy 1 Sell 8
etc,.

This effectively translates into Sell 1, Buy 3, Sell 7, etc. So just getting the difference of the lots sizes of the 2 hedges and entering that as a straight buy/sell, you save yourself the spread on the other order and the potential problem of both stops getting hit while the one of the target does not. It may not look like an anti-martingale upon first glance, but since you are alternating the entries just like the hedge, it works out the same.

Are you doubling on winners, only until you are net positive? If so I would think 8 would be plenty.

So let me get this straight, just so I understand what you are saying:

Trade number 1, .01, win
Trade number 2, .02 win
Trade number 3, .01 lose
Trade number 4, .01 lose
Trade number 5, .01 win
Trade number 6, .02 win

Is that how you would do this? The only reason we used the hedge was to reduce the effect of the martingale strategy, but since this is anti-martingale possibly we don't need the hedge? I don't know.

I guess what I did was to look at all the winners this evening and enter .1 lot in their direction. The pairs that hit their stop loss will be doubled in that direction for tommorrow and the pairs that hit their take profit will not be traded tommorrow so we can determine the new trade direction. Is this correct?

Dreamliner,

Let's backtrack to the martingale progression, so we're on the same basis.

Originally the trades doubling go something like this, 1,2,4,8,16,32.
And then we added the anti version. So once we have a loss, we double up on the win side. So then let's examine a series of losing trades in net losses

You got

Buy 1 Sell 2 (net loss of 1 lot)
Buy 4 Sell 1 (net loss of 3 lot)
Buy 1 Sell 8 (net loss of 7 lot)
Buy 16 Sell 1 (net win of 15 lot)

And this clears out the sequence to let you start over. But look closely though, how is the other position hedging this martingale effect? To put this in perspective

Buy 1 Sell 2 = Sell 1 (loss)
Buy 4 Sell 1 = Buy 3 (loss)
Buy 1 Sell 8 = Sell 7 (loss)
Buy 16 Sell 1 = Buy 15 (win)

Compare these 2 sets of orders, your profits and losses are the same. And u minimize the chance of stops getting hit on both orders without a target getting reached; in those rare instances where the market gets to your spread threshold and turns.