You have heard it before: Trade structure and money management are the most important components of a profitable trading strategy (see “Simple money management wins over time,” January 2012), and while you can’t dismiss the role played by entry and exit logic, order triggers are significantly overrated as the key contributor to profitable long-term results.
Surveys suggest that professional traders are successful about 60% of the time. Certainly that is not a bad success rate, but that is not why these traders make money.
Indeed, you can make a lot of money and be wrong 60% of the time. In the previous article, we presented the results of a test on the U.S. Dollar Index futures contract using a simple set of rules to ensure proper trade structure. There was no logic to entry/exit decisions at all. A coin flip initiated a sequence of trades. The results were a disappointing 38% success rate, losing on eight trades and making money on five.
Despite that dismal performance, the strategy made money. The reason was simple enough. Winning trades made more than losing trades lost. Winners averaged $1,100, and losers averaged $572. The net gain was $920 on a cash commitment of $3,500. The annualized rate of return was 70%.
Surely, all traders have heard the axiom, “You must make more on your winners than you lose on your losers to be successful.” It probably is the single most ignored element of the average trader’s approach and the main reason that 90% of traders, or more, fail over time.
Traders have losing trades and lots of them. It doesn’t matter what strategy you use. To assume otherwise is naïve. Still, most traders choose to operate in that zone, and shoot for that rate of success. It is liberating to embrace that no single trade has better than a 50/50 chance of producing profits, despite your best efforts. That is reality.
This brings us to something you can control: Money management.
One tenet of money management is to fund your account properly so you can sustain a series of losing trades. Because this discussion is not about trade selection, per se, we have to make some assumptions regarding the trading system we are using. Ours are simple. We’re going to have a goal of a two-to-one profit-to-loss ratio and expect a lot of losing trades.
One problem is knowing the duration of a series of losing trades so that we can know the maximum drawdown. (Drawdown is the dollar amount our account equity falls from its peak before it recovers to set a new high.) If we can’t sustain the losses from a string of losers, then we never will get to the point where we can capitalize on the string of winners when they occur.
A strategy that works uses a set framework for defining trades with realistic stop loss levels and profit objectives. That is one goal of a profitable trade strategy: Defining a realistic profit objective and a realistic stop loss for each trade. “Setting stops” (below) shows a pictorial representation of the strategy.
Shown is the daily closing value for crude oil. Plotted along with price is the 10-day moving average. The next lines above and below the moving average identify one standard deviation from the average. (Standard deviations are range measures defined by statistics that, assuming a normal distribution of price variations, contain a certain percentage of data points over time.) The next two lines out represent two standard deviations above and below the average.
The chart and the data that compose the chart give the trader a framework to seek out trade opportunities. Statistics tell us that 95% of the time the close will be contained within the outer boundaries of this chart, particularly over the long-term. However, we see that the market consistently moves up and down between these outer limits.
Be careful, however. Don’t get the idea that this chart is magic and provides a foolproof way to forecast price. It does not do that at all. What it does do is create a framework that makes sense for structuring profit objectives of at least two standard deviations and loss levels at no more than one standard deviation. For example, it would allow you to enter the market at any of the pivot points and instantly calculate a realistic profit target and stop loss level.
Consider the continuity in price action. For instance, on July 6, the market traded at minus-two standard deviations and continued to move higher through July 21 to plus-one standard deviation. It then moved down to the moving average on day 20, stopping out the trade with a two-standard-deviation profit. This is fairly typical. Once the market begins to move in one direction, it has a tendency to continue in that direction.
With this strategy and our assumptions in place, we can determine the impact of a string of losses properly.
To evaluate a series of losses, we need to construct a random series of trades. To do this, we’ll use Microsoft Excel’s random number function to assign a random number to a cell. Adjacent to this cell, we enter the following statement: “if (RC1<0.5,0,1).” If the number is 0, we assume the trade was a loser. If the number is 1, we assume the trade was a winner (see “Random results,” below).
We then copy these values down 100 rows. We repeat the sequence in the second and third columns. This provides three hypothetical trade results representing three hypothetical markets.
Each time we recalculate the spreadsheet, we get a new sequence of winners and losers. We recalculate the rows 10 times to generate 1,000 trades for each market.
The results are as expected. In the first column, the random number function generated 502 losers out of 1,000 trades. In column two, the random number function generated 484 losers. In column three, the random number function generated 499 losers.
The next step is to determine the longest string of losing trades from any of the three columns. The longest out of all this data was eight losing trades in a row. Operating on the assumption that this is a worst-case scenario, we can determine how much money we need to weather this potential storm.
The dollar value depends on the market we’re trading. If we are trading the U.S. Dollar Index, we need about $2,000 in margin to initiate a trade. If we assume we are going to make the margin amount on our winners and lose half that on our losers, we would need $8,000 plus the initial margin. This would require a total of $10,000 to trade just one contract.
Spread the risk
In the context of money management and risk control, we must consider diversification. Diversification spreads risk and reduces the capital needed on a portfolio basis.
We can demonstrate by returning to our spreadsheet and adding a fourth column. In this column, we add the values of the hypothetical trade result columns. The sum will range from zero to three. If it’s zero, we lost in all three markets. If it’s three, we won in all three markets. If it’s two, we lost one and won two. If it’s one, we lost two and won one. The results again are as expected. We have 126 zeros (out of 1,000 trades) in the fourth column. That is about 75% fewer zeros in any of the other three columns.
This is significant because if we lost on two trades ($1,000 x 2 = $2,000) and won on the other ($2,000 x 1 = $2,000), we broke even. If we lost on one trade and won on two, we made $3,000. If we won on all three, we made $6,000. The only losing situation is if we lose on all three trades on that particular row.
In the 1,000 trades, the longest string of losers on a portfolio basis is two. So, by trading three contracts with a 50% win and a two-to-one win-to-loss ratio, we reduce the longest string of losses by 75%.
With regard to final allocation estimates, if you intend to trade one contract, you should start with enough funds to sustain eight consecutive losses — plus the money necessary to margin the ninth trade. That is probably as much as five times what most traders allocate, assuming they only need to cover the initial margin.
If you incorporate diversification into your strategy, you can reduce the amount needed to cover the losing trades. If there were no diversification benefit, we would need to cover eight consecutive losses on each, plus the margin for the ninth trade, or $30,000. However, operating under the assumption that we can decrease the worst-case loss sequence by 75% through diversification, we can reduce the amount allocated to trade losses to just $6,000 (plus another $6,000 to cover the margin for all three markets on the next trade).
A word of caution: The real world will not necessarily match this idealized scenario. Although our tests have suggested a more conservative approach than most beginning traders adopt, going further would be prudent. Perhaps a starting contribution of twice the portfolio loss estimate ($12,000) plus the $6,000 margin requirement would be better.
One thing trading futures for 40 years teaches you: Traders have a hard time curbing their expectations. Leverage is the appeal that draws most of us in, and it is that attraction that causes many to flounder. With careful risk control and money management, you can make significant returns on your account. However, if you over-leverage and neglect trade structure for the latest “sure-thing” entry trigger, you undoubtedly will lose.
Joseph Stuber began his career in 1972 as a research analyst. He is an author and lifelong student of risk and risk management.