The Theory Behind “The Robust Pawn” :
The PAWN was coded strategically with an emphasise on preservation of capital rather than high returns. FX trading is a risky business unless an extensively thought out strategy including money management protocols are in place. We have given ourselves just a slight edge using a long/short strategy whereby the PAWN holds an open position either long or short in each of 28 major FX pairs at any given time. Let me explain some interesting and practical observations that have formed the basis of the PAWN.
While looking online at various Forex Signal Providers and algorithms I noticed one thing in common and that was the disclaimer stating that “past performance does not guarantee future results”. This got me thinking on how we could code a strategy that is so Robust that even if it were the first day of the Forex markets the strategy would perform profitably or at least break even, assuming that the movement in the Forex Market is completely random. Consider a scenario where at the end of the current bar on a chart (regardless of the timeframe chosen) the next bar is either going to close positive or negative from the previous bar. Assuming that this direction is completely random than it would be the same probability of flipping a coin. In other words if we were to flip a coin to make a decision at the end of the current bar, whether to buy or sell a given currency pair and hold it for 1 bar and close the position we would have a 50/50 chance or 1/2 chance of making a profit leaving aside spreads and commissions for a moment.
I saw a magician named Darren Brown who would ask one audience member to hold a coin in one hand behind his back and he would guess which hand the coin was in. The first time he did it was not that impressive as he could have gotten lucky. But he then repeated it 5 times and each time got the correct hand the coin was in. Based on the fact that each test was independent of each other, probability and statistics would conclude that if he were guessing he would have a 1 / 2^5 or 1 / 32 chance or just over 3% chance of getting all 5 correct. Now this is impressive but think about it in reverse and we would conclude that if Darren Brown were completely guessing at random then he would also have a 1 /32 or just over 3% chance of getting all 5 tests wrong. But using the edge he has in psychology and finding inefficiencies in the human mind he was able to defy all odds and get all 5 out of 5 tests correct.
Bringing this concept to currency trading and assuming that at this moment in time we are about to finish the first 240-min bar in the EUR/USD pair and at the end of the current bar we would have to make a decision to either Buy or Sell the currency pair, decided randomly like flipping a coin. Heads we go long and tails we go short. Also, assume that we put a profit target of 10 pips and a stop loss of 10 pips. When one of the target is hit than the position is closed. Based on this scenario and if the markets were completely random than we would have a 50% chance of making a profit of 10 pips and 50% chance of losing 10 pips. Now say we flip a coin 28 times for each of the major currency pairs and their crosses and make a decision to go long or short and place the same profit target and stop loss at 10 pips. What would be the probability of each of the 28 trades making 10 pips and resulting in a 280 pip profit?
1 / 2^28 = 1 / 268435456 = 0.00000037 % chance of getting all 28 trades correct and making a profit of 280 pips. This is an almost impossible task. Now lets reverse that theory and determine the probability of getting all 28 trades wrong and losing 280 pips. This probability would be the same as the probability of getting all 28 trades correct. Therefore, the probability of losing 280 pips is 0.00000037 %. What this tells us is that if we are to trade all 28 major currency pairs including their crosses and we were to flip a coin and make a decision to either buy or sell each currency pair at arbitrary intervals then for us to lose on all 28 currency pairs is extremely low as gains and loses would inherently balance out and therefore the probability of wiping out our trading account overnight is highly improbable. This theory also concludes that with this type of trading we are highly likely to breakeven at any point in time and therefore in all probability we expect to see a flat equity curve. This theory doesn’t account for spreads, commissions and correlation of course as it is theoretical so far but what this tells us is that if we adopt such a model than we have a high degree of capital preservation and safety even if we have never seen any historical FX data.
Now that we have theoretically mitigated the risk of losing our entire account balance overnight by trading 28 currency pairs simultaneously using a long/short strategy whereby we have 28 open positions, long or short at any given time we can now come to reality that we do in fact have historical data which allows us to exploit inefficiencies in the FX market as well as fundamental data such as interest rates which allows us to develop a strategy that provides us with just a slight edge from the random coin flip trading theory and that is exactly what THE ROBUST PAWN strategy is based on. Displayed in the back-tested results below it clear that each year we have on average anywhere from 55-60% profitable trades and therefore the PAWN is always just slightly correct similar to being the casino in a baccarat game or being the house when clients are betting on red or black in roulette. We have given ourselves just a slight edge and allowing for many losses and wins which inherently creates one of our most Robust trading strategies that focuses on capital preservation while still making small and steady profits rather than enormous but risky profits.