Bob Llewellyn / The Forex-Assistant April 2014
Is a Martingale system bad?
When it comes to trading, there is nothing more reviled and berated than the martingale. However, a martingale recovery system is just a tool, good in some cases in the hands of someone that knows how to use that tool, and bad when misused.
First of all, a martingale is not a trading system. It doesn’t tell you whether to buy or sell, nor when to open or close a trade. If it isn’t a trading system, what is it? It is a recovery system. It’s not the best recovery system nor the worst. If you never lost a trade, you would never need a martingale or any recovery system because there would never be anything to recover.
First lets take a look at how the martingale system works. It was originally made for the ‘flip of a fair coin’. You pick heads or tails and the bet amount. Let’s say you pick heads to win for $1 and the fair coin comes up tails, your down $1. The next time you want to win and also get your lost $1 back so you bet $2, (double the bet), it loses again, your down $3. You can keep doing this doubling process until you run out of money.
The next bet would be for $4, and this time you get it right and make $4, one is your profit and 3 for the recovery of the lost amount. To get to the point where you won $1 you put $8 at risk. That risk increases as the iterations do. (iteration = one pass through a system or trade)
The illusion is that if you have a lot of chances, you are probably going to get it right at least once before you run out of money. The probability of getting a trade right is 50/50 (fair coin). Lets work with getting it wrong, so you will have a 50% chance of loss. If you double your bet and try again, you will have another 50% chance of loss. 50% converted to decimal is .5 and (.5 times .5) is .25 or 25%.
So you only have a 25% of loss on just 2 plays. A third flip of the coin or a trade would reduce the probability of loss by half or to 12.5%. The probability of loss keeps getting smaller but never hits zero. There is always a risk of loss. However, the size of the risk grows with every iteration. The probability of loss decreases by half while the trade size doubles. All to win one, dollar.
If you stop there as most traders do, you will determine that the martingale is a fools game. What is wrong with this assessment? All that we just determined was for a “fair coin”. The probability of loss diminishes by half because we were assuming the coin was fair. Here is the untold secret, ‘there is nothing fair about the forex market’. Here’s what fundamental traders know that technical traders gloss over, the distribution of outcomes may look random but they are not. There are often way to many factors to predict an outcome but there is always a cause.
There is also a resistance to change in the market so without some outside occurrence the prices would never change, and that would be the end of forex trading.
We don’t need to be able to predict a definitive out come however if we can increase the probability of a win, the probability of a loss is diminished. No more fair coin, the odds are now in your favor. This means that the probability of loss is diminishing at a faster rate than the amount of the risk. And that is a different story.
The next question is, if you can win more trades than lose, why bother with a recovery system as your winnings will over come your losses all by themselves. Guess what, that is a recovery system, probably the least effective but obviously the simplest.
Lets look at the benefits of both, to do that we will assume 100 trades with a 75% win ratio. For the straight trade $25 of the $75 won, will have to go to off set the $25 lost, making a profit of $50 out of 100 put at risk.
The Martingale would have $75 on the first trade of each set. The 25 losses would lessen the win rate by one only to increase it again with a win on the second go around. It actually does diminish a few dollars because about 6% of the time it will require a 3rd iteration. So the total efficiency is a profit of $73 out of the $134 put at risk. The amount at risk here is at a lessor risk than the straight trade because of the multiplying factor, but it is still at some risk.
There is a modified system that can help with the earnings to increase the efficiency a bit. The standard martingale only wins its base amount when it wins, no matter how large the trade is. The sequence of trades would look similar to 1, 2, 4, 8, 16, 32 … each doubling every time. To have one win for every iteration, we would use 1, 3, 7, 15, 31 … doubling the previous and adding one.
You will note that the sequence to the modified looks a lot like the standard but advanced one iteration and reduced by one trade size. However, since it pays $1 for every trade, you would have close to about a $98 profit for $192 put at risk. (again at a lower risk than the straight trade).
The conclusion then is that a martingale recovery system is useful and not so dangerous if your trading system is any good. If it is right most of the time, then a martingale recovery process may be of help but if your trading system is bad, no recovery system will make a bad trading system do good.