Since martingale burns accounts, why not do the opposite?

Greetings to all fellow traders,

A while ago i started a thread " Is the martingale way the only way in forex?" link: https://www.forexfactory.com/showthread.php?t=683804

And as expected, I received massive bashing and warnings from FF members (I know they did it with pure good intentions) about how martingale sooner or later burns accounts mercilessly, and I totally understand and agree with their point of view, since martingale needs infinite or "enough" money to be working in theory, and that kind of money is hard or impossible to come up with at the first place.

Moreover, martingale risks the entire capital for small gains, and such gains and capital can go bust when things go wrong as they usually do. Some threads here in FF and other websites discussing martingale strategies and EA's suggest that a 'relativity safe' martingale with 5-10% monthly gains with passive settings can cut the deal, but I do not think that a 5% per month is worth risking the whole capital.

Since the vast majority of traders here agree that martingale strategies = margin, and martingale users are always stuck with huge DD and wished if things went differently, why not reverse the whole thing? Instead of gaining small tiny bits of profits per month and bust the account with profits made later, why not sustain small tiny bits of losses per month and double the account later, after recovering losses of course.

I know that the concept of anti-martingale or reverse martingale is nothing new, but I did not see any system which completely reverse actual traditional martingale EA procedure. For example, a conventional martingale EA starts with a buy or a sell based on a certain strategy or on random, then sets the TP at 10 pips, without a stop loss. If TP is hit the the EA enters a new trade on next candle. However if the market went against the initial position by more than a certain amount of pips, say 10 pips, a new position of an increased lot size by 1.2 multiplier is opened at next candle, and TP is adjusted for all positions so that the total profits = sum of lot sizes* 10 pips ( the intial TP amount) and so on till the common TP is hit, or a margin call happens. Being conservative with the settings by reducing initial lot size, TP and multiplier factor and increasing the pip step where new positions should be added, makes the system run a bit longer before the disaster eventually happens. While being aggressive in the settings accelerates bankruptcy. Bearing in mind that the more positions being added while increasing lot sizes at a certain factor, makes the TP level a bit easier to hit, but the TP becomes somehow far away and requires a big retracement to hit it.

So far so good, I believe most of traders agree with me till this point.

Now literally reverse the whole thing, open a position on random or as per a certain strategy, set SL at 10 pips without TP. If the SL is hit no problem, open another trade at next candle as per entry rules (not necessarily the same direction of last position). Now if the marker moves into our favour by a certain amount of pip step, add a new position at next candle with an increased lot size by multiplier factor and adjust the SL so that the total loss= sum of lot sizes* 10 pips ( initial SL amount). For the exit in profit, close all trades when a certain amount of $ profits is reached, you can set that into 50% of capital or 100% or what ever floats your boat. bear in mind the shared SL level now is also harder to hit when you go with many opened positions, similar to TP in original martingale. This should in theory give your trades more room to breath till the desired money target is acquired.

In conclusion, the tiny bits of monthly profits in original martingale are now the tiny monthly losses in reverse. And the margins in the original martingale are now doubling the account in reverse, or 50% DD in original martingale is now 50% profit in reverse.

In theory that sounds good, but in reality other factors may contribute to the desired outcome. For example what if the account is drained out before hitting the jackpot? What settings for lot size, multiplier factor, pip step, SL, money TP should be used? Which pairs are the best? Answers for these questions are decided by people who lost money using martingale without being able to double their initial investment, but nothing is guaranteed in forex as we all know.

This wild idea struck me while I was taking a ****, so I thought I would share it here since it makes some kind of sense, so please let me know what you guys think. And as usual, I welcome all kinds of constructive criticism and friendly bashing.

Peace.

You have the opposite problem. Martingale doesn't kill your account immediately, you can sometimes double or triple your account before you win. In the same way, you could burn through 2-3 accounts before getting anywhere. Without doing it on demo with 0.01 lots and seeing an average drawdown it would be impossible, but I think the whole Martingale, reverse or otherwise, is too luck based.

Very true, there is always a risk. The only way to understand the risks and rewards is to forward test it as you said to evaluate the outcomes on a demo account first. Maybe there are certain times or trading strategies where this approach works best.

I think I gonna open a new thread : ''Since 95% of traders lose money in Forex, why not do the opposite and just not trade?". Why do you think guys ?

I have tried it, it works sometimes. Last time I demoed it I was doing amazingly till NFP wiped me out (I was doing a crazy MM to amplify any risks), but apart from that most of my trades (50pip T/P and S/L) was doing OK.

You have to understand the open orders. The longer the trend has been going on, the more open orders are in the negative, because people don't close their trades when they should. There are also people who are trying to guess the top, and you have to keep track of the movement of retail traders from buy to sell. Oandas is probably the best, because those on Forexfactory are slightly better traders than most.

That's statement makes perfect sense, I've been introduced to Forex since 2008, till now my net loss is +30k $ (of course just extra money which did not effect my life style what so ever, but I would prefer if I did not lose it at the first place) . I don't claim that I have 9 years experience, because the only things i learnt so far are the mt4 standard indicators and how they function. Now for how to use them, just browse different threads and you will find infinite settings and combinations and theories in that regard. Other than that I feel like I am still at day one. So yeah, better not to trade based on statistics, especially with money you cannot afford to lose.

I tried a similar approach of the famous rule of 'Buy low, Sell high', for me I observe several pairs, then wait for a massive continuous trend of +1500 pips, then do my move there anticipating a retracement after such a huge trend. Till now it worked every time, but I am doing it with such MM that my DD was never more than 1% and my profits are 4% in the past 2 months. This method is long term and very boring, but only needs minimal observation and effort. I don't consider it a system yet, just trying it out.

Thanks, Rmk90, for bringing up this idea. It has interesting aspects, although not quite the way you intended in your first post.
My current thread (https://www.forexfactory.com/showthread.php?t=680556) deals with a counter trend strategy, which uses a stop-loss-secured grid, when the positions are going against the initial entry. So why not turn this idea and use a "profit grid" - at least as an add-on to the original strategy ?
I will give it a try in my next versions / backtests.

Regards, Oliver

No problem Oliver, this approach should work best while applied to suitable trading strategies, not as a stand alone trading style, hope it works out for you. I've had a quick look at your thread and it seems interesting, specially it's a counter trend strategy where you anticipate a decent reversal, so in my opinion this approach should magnify your profits if done right. Good luck and all the best.

Position sizing is about finding the right balance between preserving capital and maximizing gains.

Preserving capital is about keeping sizes small, making it easier to recover from drawdown. Larger sizes increase overall risk, and there needs to be the correct kind of justification in order to do so.

Maximizing gains is about having the largest pos sizes coinciding with the biggest pip winners (and/or smaller sizes coinciding with any losses). If there is no reason, in terms of market probabilities/behavior, to increase pos size -- for example, if you increase size to recover recent losses in your P/L -- then you are merely playing a game of hit or miss. The market takes no cognizance of the P/L of one insignificant retail trader.

Hence I don't see either martingale or reverse martingale, in itself, as providing any kind of edge. You'd need to find potential biases in market behavior first, and then ideally size your positions commensurate with the extent of the bias. For example, if there is independent (e.g. underlying fundamental) vindication backing the strength of a move, then you might be justified in pyramiding further into the move, ideally on suitable pullbacks.

Just my 2c FWIW.

Well said, and couldn't agree more.

RMK90, this shouldn't be in the rookie talk.
I use a very simple breakout system of each daily candle. SL in the middle and risk 0.5% - I usually run into a bad period of 20-30 losing trades in a row. ADjust you appetite for risk here :-)

Here is my strategy:
4R target strategy

Timeframe: Daily

Start with bracket trades on each candle (Pending order)
TP is 4:1R targets
SL is set by half of the daily range (candle)
Pending orders are valid for 22 days (one month on chart)
Pairs:
UC
EJ
GU
UJ
Gold (XAUUSD)
WTI

Expected win rate 26-27%
For BE 20% (doh)

Risk pr. trade 0,5 because of the low win rate. Bound to hit a 35 loss streak and large drawdowns if risk.
Not many can or will trade like this = my edge..

Before some of you might disagree with my strategy, it's how I trade and it works for me. I see no patterns in market, only random walk-levy flight. Markets usually go further than we expect and there is the "technical edge" (positive skewed distribution)

Go try and backtest this for 2017 - you be surprised how well this works.

Martingale works in specific times like for short terms so we can not always rely on it for it is very riskier and evry google on martingale will show us that many traders lost their money due to this reason so better be careful while applying the martingale.

The mods moved this thread twice from trading systems to trading journals then to rookie talk again.
Your strategy seams interesting, thanks for sharing and will look into it.

Using martingale makes you lose, then doing the opposite also makes you lose, maybe better not to use either of them. Or the best thing to say is: use any of them at certain situations or as per certain strategies which makes the best out of them, not just going in blindly. At least that's what I think. Higher returns against lower losses even at win ratio less than 30% may make it work, for the reverse martingale that is.

Rmk90: I do not know anything about Martingale really at least not compared to some of the members I have been reading but help me out here since price direction is so hard (for me at least) to get right I have been working on a martingale of sorts that makes a bet on movement. Here goes:

1. Eur/USD for example:

A. Buy .01 at market with Profit at 10 ticks and stop at 10 ticks;

B. If you win start a new series; but

C. If you lose sell .02 with a profit at 10 ticks and a stop at 10 ticks;

D. If you lose again Buy .04 with a profit at 10 ticks and a stop at 10 ticks

Obviously the number of ticks is arguable or modifiable. I used micros just for an example as well. I am trying to make a bet where we win if the market moves period end and it does not matter which way but a 20 pip move signals a win. Since your obviously thinking about this I mentioned your name above but anyone with constructive ideas is free to weigh in. Perhaps this has been tried before I dont realize that. Any thoughts?