Personal Finance & Money Stack Exchange is a question and answer site for people who want to be financially literate. It only takes a minute to sign up.
Sign up to join this community
What should I aim for to achieve an effective and profitable trading system? Should I aim to get as high a percentage of winning trades as possible (ie. 100% winning trades)?
Also what percentage return should I be aiming for each year?
I plan to mainly trade individual stocks in the Australian market using both fundamental and technical analysis.
To achieve an effective and profitable trading system you should aim for a positive Expectancy Score. The higher the Expectancy Score the more profitable the system.
This value is an annualized Expectancy value which produces an objective number that can be used in comparing various trading systems. In essence the Expectancy Score is a combination of a trading system’s Expectancy - how much you expect to earn from each trade for every dollar you risk per trade, and Opportunity - how often your strategy trades.
Let’s imagine we have two trading systems that have two different Expectancy values:
On face value Trading System #2 seems more profitable, however if each system risked $500 on each trade and System #1 produced an average of 5 trades per week whilst System #2 only produced an average of one trade per week, then the weekly Expectancy Score or Profitability for each system would be:
Thus Trading System #1 would be more profitable.
So how do we work out a trading system's Expectancy?
We simply use the formula:
Expectancy = (Probability of Win * Average Win) – (Probability of Loss * Average Loss)
So if you have a Trading System which wins often, say 80% of the time, but because you are nervous of a winning trade turning into a losing trade you cut your profits short with an average profit of $200. But at the same time you let your losses run in order to hopefully let them turn into winners, and you finally bite the bullet and sell out once your loss hits an average of $1,000.
Your Expectancy for this system = (0.80 x $200) - (20% x $1,000) = $160 - $200 = - $40.
This means that it is expected that you would lose $40 every time you took a trade (on average).
On the other hand you start a new Trading System which wins less than half of the time, say 40% of the time, and you let your profits run with an average profit of $1,500. Even though your losing trades are more than your winning trades you keep your losses down to an average of $500.
Your Expectancy for System 2 = (0.40 x $1,500) - (0.60 x $500) = $600 - $300 = $300.
This means that it is expected that you would make $300 every time you took a trade (on average).
A good book to read more about Trading Systems, Position Sizing and Expectancy is Trade Your Way to Financial Freedom by Van Tharp. Here is also another link you can get more information on Expectancy.
Expected Annual Returns
Your Trading System Expectancy will also help you set your annual return goals.
I personally have started a new Trading Strategy at the start of September and am aiming for a 100% return on a geared account. Without the gearing I would have been aiming for about 25%. I am trading Australian shares as well but through a CFD account with margins ranging from 5% to 30%. I have taken a conservative approach in using an average margin of 25%, which is how I convert the 25% return (non-geared) to 100% return (geared).
I started with $10,000 and have so far taken 44 trades in 3.5 months (or 15 weeks) averaging 2.93 trades per week. (One trade being the opening and closing of a position).
These are my results so far:
Win rate =22/44 = 50%
Average Win = $419
Average Loss = $232
Expectancy = (0.50 x $419) - (0.50 x $232) = $209.50 - $116 = $93.50 per trade
Expectancy Score = $93.50 x 2.93/week = $274 per week
My return so far in 3.5 months is about 40%, and if I multiply $274 by 50 weeks (allowing 2 weeks over Christmas/New Years for no trading) my annual return at this stage is expected to be $13,700/$10,000 x 100% = 137%. So I am well on the way of meeting my target return of 100% per annum. (By the way I have also just gone through my maximum drawdown 3 weeks ago of 10.8%. My target is to try to keep drawdowns to a maximum of 15%).
There is a number of metrics you can look at but, in the end, the main things you need to understand about your system are:
Long term profitability
Note that the hit rate does not have to be high as you could only win 20% of the time but on average gain $100 when you win and only lose $10 when you lose.
Psychological difficulty
If your system has long periods of drawdown (a simple buy and hold strategy can be trading below its all time high for years) it can be difficult to stick to it in the long run so you need to make sure that this is in line with your ability to be patient.
Two metrics that can be useful in that regard are:
Risk / Return
You can't talk about returns without also mentioning risk. A strategy that can return 50% per annum but can also return -80% from time to time may or may not be appropriate to your risk appetite. In general the higher the return, the higher the risk. One measure of that is the Sharpe ratio, which is roughly calculated as the annualised return of the strategy divided by its volatility. 2 is a good Sharpe - some high frequency strategies can have Sharpe ratios as high as 5-10.
Beware the methodology
When testing your strategy, unless it is with real money, always make sure not to underestimate typical flaws. With backtests:
Understand statistics
Understand the statistics behind the performance and make sure you know how to separate luck from skill.
Let me give you an example: if I calculate the average performance per month of the S&P 500 over the past 50 years, there will be a month which has the highest average (say April) and a month with the worst average (say September). This is very likely just random and the probability that the pattern in the future is probably very weak. Deriving a trading system from that observation is unlikely to make you rich.
Statistical significance testing can be very important depending on the type of systems you plan to use. A typical measure is done using a t-stat or p-value.
One measure of that and doesn't say it is the only measure to be used. If you have a background of statistics, you should know how to use it or else learn it. Using statistics is a highly valued skill in financial markets, that is worth learning.
Dec 17, 2014 at 10:02