For short-term traders (intraday and swing, not position traders)
Those who claim that “it is not about percent of winning trades, it is about how much you make when you are right” conveniently omit any discussion of “profit factor” (gross profits divided by gross losses).
The purpose of this post is twofold, first to reveal the other side of the coin regarding money management and the quote above, with some stats and an equity curve to provide proof, and second, to show what form of money management I use in my position, swing, and day trading.
First off, there are some famed traders who believe percent of winning trades should take a backseat to the size of the average winner. Let’s compare two methods of money management using profit factor:
Win rate: 50%
Avg win to avg loss ratio (reward-to-risk): 2 to 1
Win rate: 67%
Avg win to avg loss ratio: 1 to 1
In both cases, profit factor = 2.0. Thus, there is no advantage in Trader A’s approach. Trader B may even have an advantage if he is able to turn over his inventory more times during a specific period, i.e. his trade frequency is higher.
Using 1R money management
The following is a forward test (i.e. in real-time, not a backtest) that I ran in ’12-’13 on a small universe of 30 issues, mostly commodities with some ETFs. The goal was to achieve a 67% win rate using r:r of 1:1. This meant that most trades would either hit +1R in profit, where R = initial risk, or be stopped out at minus 1R. If price reached +0.8R, s/l was raised to b/e. The majority of positions were exited in 1 to 5 days.
Only pullbacks were traded, no breakouts, both long and short. An edge had been discovered, it appeared consistent when a number of charts were eyeballed, and a test was in order. The same discretionary setup was used again and again. Over a reasonable sample size, its positive expectancy was borne out, as the below graph shows.
This is what short-term trading is all about: An edge and positive expectancy.
The objective, then, was not to hit doubles, triples, or home runs, but to hit singles consistently in order to generate an income of 2%/monthly. To do that, a minimum of 12 trades needed to be taken each month. This would produce eight wins and four losses, or four net wins each month. Using R of 0.5%, this equates to 2% profit per month (excluding commissions and slippage).
Looking at the results table above, the obvious question becomes: Why was 0.5% risk used when max drawdown was only 2%? The answer is that this was intended to be used for opm (other people’s money), where a more conservative approach to risk was sought. Also, over a larger sample, say 500 trades, perhaps the max drawdown would be greater than 2%. This is discussed in a moment.
With 1% risk, drawdown is about 4%, still quite manageable, and monthly return doubles to about 4%/month, or 60% per annum with monthly compounding. Due to the small universe (30 issues), I came close but did not quite average 12 trades/month. Adding another 10 issues so that the universe totals 40 issues should take care of that.
Before a test on a setup/strategy/system is complete, it is a good idea to determine the probability of having an x% drawdown over a large number of trades. Running a Monte Carlo simulation on 100,000 trades with 1% risk, a 67% win rate, and 1:1 indicated the probability of a 15% max drawdown to be < 1%. With 2% risk, that probability rises to 100%.
The why’s and why not’s of swing trading commodity futures
In the results table above, the reader might have noticed an R² (correlation) of 0.06. This means the results had virtually zero correlation with the S&P 500. This uncorrelated nature is the chief advantage — along with added leverage — of trafficking in commodities. U.S. stocks can be in a bear market, yet the swing trader in commodities would scarcely feel it in terms of trade frequency and expected results. This is why commodities make sense for a swing trader.
Why commodities do not make sense for swing trading is the risk of a weather forecast or other event putting a commodity into a lock-limit situation. This makes it impossible to exit a position if a stop is hit, and can sometimes last for multiple days. To be sure, these are uncommon events and do not affect all commodities. For this reason, following the ’12-’13 test period, a decision was made not to trade the above system in commodities.
However, a recent conversation with Linda Raschke, the Market Wizard, assuaged the lock-limit concerns. As a result, I will begin live trading the above strategy with the following changes:
a) Risk will be 1% instead of the 0.5% used in the test.
b) Instead of adding more issues to the universe of 30 tradables, a second strategy will be added in order to bring trade frequency to at least 12 per month.
c) Instead of the 1R protocol used during the test, the “trim and trail” method of money management will be used. Thus, a half position will be taken off at +1R, with the s/l moved to breakeven and then trailed.
The monthly objective will remain the same: +4R in profit, or 4% when using 1% risk.
One man’s approach to money management
Position trading. For growth stocks, I use a “scale-in, all-out” approach. This means I will begin with a 25%-50% starter position, then place one or two add-on trades. Exits are made with the whole position, hence “all-out.”
Swing trading. Because growth stocks trend so well, I don’t want to use a 1R approach. I tend to use an “all-in, scale-out” approach. This means taking half off at +1R, moving the s/l to b/e, and trailing it so that there is a chance of a large R-multiple move on the second half, i.e. the trim and trail method. When I swing trade S&P 500 and Nasdaq 100 issues which don’t trend as well as growth names, it is all-in, all-out, where I trade for 1R as was done in the above test.
Day trading. I never day trade stocks, preferring to use the futures market for a number of reasons. The vehicles are the Nasdaq, crude oil, and the euro. I will always take half off at +1R, move s/l to b/e, and then trail. Virtually every single futures day trader that I know takes half off at a reasonable profit. One of them calls it “adjusting my cost basis.”
In no way, shape, or form am I advocating a 1:1 money management method. This post was simply to illustrate how it’s not all about how much you make when you are right. The win rate is an important part of the profit factor equation.
Everyone is wired slightly differently. Some are more comfortable by taking 1R, booking a higher win rate, and moving on to the next trade. Others go for the bigger R-multiple wins, albeit with a lower win rate. For markets that don’t trend as well, I prefer the former, and for trending markets the latter.
There is no wrong way to trade, there are only results.