How far can you go?

Assume you begin with $100 on a 1:100 leveraged account.
For whatever reason you think there's a trade that will gain you 500pips (if you're lucky), with a compounding approach and stacking orders on the way how would you place the orders? Would you pick every pullback or use a grid like approach? Further, what is the potential of earnings assuming that for .01 units around $16-$17 is required (assume you're compounding aggressively)?

Of course this will all depend on the drawdown foreseen/able to be endured/actually sustainable until margin call.

Essentially i'm trying to build a calculator or a bit of software to calculate the ideal stacking arrangement for a given trade.

As far as I can think right now, there are four factors: capacity for drawdown, starting balance, lot size and your target(pips distance from the first trade, followed by every subsequent trade).

This is a very interesting question. I thought about it a lot; especially in term of maximum risk and expectancy variation over the time (see http://www.forexfactory.com/showthread.php?t=411923). Unfortunately the optimal pyramiding depends on the future price action... More precisely on the depth of the pullbacks.

If the PA will be a straight move any grid makes a killing. The problem is that a deep pullback may push the trade severely underwater because of the high leverage. See the snow ball EA (http://www.forexfactory.com/showthread.php?t=239717).

If the price moves in large waves closing at what seems to be a top then re-entering later (with, for instance, a counter-trendline breakout) looks like the best option. The two problems here are of course that 1/ you risk closing the trade and see the price re-accelerating without you... far away from the target. 2/ re-entering too early. The point 2/ can be mitigated by fragmenting the entry. What shall be the size of this position? The same size? Bigger so that it is as if the previous position was still open (a virtual pyramid)? or smaller to make sure we keep some pips in case of reversal? It depends on what is/are the most important factor(s) for you among the profit, the DD, the risk of ruin, etc...

A hybrid approach is to grid more or less aggressively and close all but 1 or 2 positions (with SL @BE) as soon as the price enters a zone of potential reversal. Then wait for the pullback and re-enter with the same method until the target is reached. The problem here is to close when the price is strong that is just before the retracement starts (S/R hit, strong overbought condition, trendline break on the smaller TF, and thousands of hours of screen time for fine tuning the guts feeling). Otherwise the last entry will be in the loss and the second last entry will be almost BE, the third one will barely cover the loss. Well if there is a 3rd one!

What is now a certainty for me is that the management of the trade isn't predefined at the entry but it has to be adjusted while the PA unfold.

I think PipMeUp hit pretty much every relevant aspect in the question. Most importantly as he stated, the optimal compounding / reentry is dependent on the structure that future price action takes, and could only be determined in hindsight. You can constantly react to changes in the price structure (impulsive straight line moves vs. waves with pull backs), but those reactions are only good so long as price continues to advance in that way. Over the course of the 500 pips there's no reason you wouldn't see a random combination of straight thrusts (grid would perform best) and wave-like movements (pullbacks ideal) of different speeds, lengths, and proportions.

How about the concept The Semi-definitive guide to Pascal's Triangles in Forex for adding to positions? I'm looking for feedback on this concept. Two excerpts below:

Nice thread. I'm interested in what unfolds.

I've started looking into dynamic grids where their grid sizes adjust dynamically proportional to a factor like momentum. The results look good but grid results always do ;-) Doesn't stop bad entries of course.

Damn fellas, i wasn't expecting such an interesting read. Thanks to everyone who has posted so far.

I'm scouring the info you've all mentioned and/or supplied, so hang tight whilst I try wrap my head around it. I do indeed like the notion that the grid (for lack of a better term right now) is adjusted according by say - momentum, breakout zones...whatever, i reckon you could pick whatever mechanism you're comfortable with to make your judgement.

Let's suppose we use pullback entries, or perhaps 1,2,3's - or whatever. Now, the nature of the beast being fractal means that (margin allowing) you could take a pull back signal on the smallest, through to the largest timeframe, and assuming price DOES continue in your favour you could theoretically add positions every time frame. Even on the smallest timeframe for maximum accumulation – but lets not forget, exceptionally greater risk.

I'm incredibly interested in what's posted thus far, I admit I initially thought it was more of a "black&white" kind of question. From whence i spot an opportunity to visually demonstrate a proposed example, I will post here for all to see and discuss and if you're so inclined, follow.