How to assess a trading strategy’s risk
-
- 0
- 1014
- 0
Evaluating a trading strategy is an essential element for long-term success. Once the trading strategy has been determined, it needs to be tested on a large sample of trades. There is no specific time frame to properly test a strategy, it will all depend on the trader’s trading style (if he is scalping or swing trading). The advantage of coding the strategy (see: expert advisor) is that the strategy can be tested over the long term with backtests available on trading platforms. If not, it has to be tested manually, by placing the orders on a demo account. Testing it oneself remains the best solution. Effectively, traders are in a real situation and can evolve their strategy according to their experience. We never repeat it enough on the forum, succeeding in trading takes a lot; a lot, a lot of work!
To evaluate a trading strategy, several criteria should be considered.
This calculation allows the average risk/return ratio to be calculated. In other words, it tells us whether or not the average gain expectation is higher than the risk (stop loss spread). A ratio below 1 is obviously to be avoided. A ratio of 1 means that this strategy is balanced, with 50% winning trades. The higher the ratio, the more losing trades the strategy supports.
With a ratio of 2 for example, the trader only has to earn 1/3 for his trading strategy to be nil since his expected gain is twice as high as his risk. In order to further analyse the strategy, this can be done for each asset. Thus, a trader can remove those assets that are less profitable or that do not match his average risk/return ratio from his strategy. Ideally, a ratio should be at least 1.5 and tend towards 2.
Drawdown represents the maximum historical drop recorded on a trading account over a given period (day, week, month, year). It is the difference between the highest and lowest balance of the account. Drawdown is indicated in value and/or percentage. It represents the account’s volatility. The higher the drawdown, the riskier the trading strategy. If it is too high, the trader should reduce the size of his positions that are not suited to the amount of his portfolio. In this case there is too much leverage. A high drawdown does not necessarily mean that the account has been in loss. For example, a trader trades with €1,000 and a few months later, his account is at €1,500. Following a bad series, his account falls back to €1,200. His drawdown is 20% ((1500-1200)/1500) which is worth €300.
This drawdown must be related to his portfolio’s performance. Here again, there is a notion of profitability in relation to the risk taken. There is no perfect drawdown, everything will depend on the investor's profile and risk aversion. However, drawdown should not exceed the portfolio’s performance. If it does, then the strategy should be changed immediately.
Let us take several examples. A trader’s trading strategy has enabled him to achieve a 50% return over one year. Over the same period, his drawdown is 25%. This means he risks 1 to win 2.
Using another trading strategy, he only achieves 15% performance but his drawdown is 5%. Although the performance of this strategy is lower than the first, from a mathematical point of view it is much more efficient. Effectively, he risks 1 to win 3.
The Sharpe ratio measures the return on the additional risk taken compared to a so-called "risk-free" asset. Without discussing risk-free assets (they do not exist, there is always a risk - see: government bonds), we are talking about benchmarks. We will therefore compare a strategy’s profitability to a benchmark index by integrating the notion of additional risk. For example, you could compare your strategy to the CAC40 index. Performance above that of the CAC40 does not mean that your strategy is good. Effectively, the marginal risk of your trading strategy must be taken into account.
The Sharpe ratio formula is S = (R - r) / E: where R is the rate of return of the portfolio under consideration, r being the chosen benchmark for comparison (generally the risk-free investment rate), and E the standard deviation of the rate of return of the portfolio under consideration.
A ratio greater than 1 means that your portfolio's out performance over the risk-free investment is not at the cost of too high a risk. The higher the ratio, the better the portfolio.
A ratio below 1 means that the portfolio under performs the risk free investment. You shouldn’t invest.
A ratio between 0 and 1 means that the excess return over the risk-free rate is lower than the risk taken.
To calculate the Sharpe ratio, calculate your portfolio’s standard deviation. Standard deviation is a measure of volatility and is the square root of the square average of the deviations from the average over a given period (or the square root of the variance). Let’s consider an example: Over the first three months of the year, the CAC40 posted 10% performance. Let’s suppose that portfolio 1 shows the following monthly performances with performance over the period of 12%:
- January: 10%
- February: 5%
- March: - 3 %
The series average is (10+5-3)/3 = 4%
The variance of the portfolio is: ((10-4)² + (5-4)² + (-3-4)²) / 3 = 28.66%
The standard deviation is therefore equal to: V(28.66%)= 0.5353
The Sharpe ratio of the strategy is therefore: (0.12/0.10) / 0.5353 = 2.24
The strategy’s profitability is therefore good compared to the additional risk taken.
Let's take another portfolio example (2), this time with the following monthly performances and performance over the period of 30%:
- January: 20%
- February: -15%
- March: 25 %
The series average is (20-15+25)/3 = 10%
The variance of the portfolio is: ((20-10)² + (-15-10)² + (25-10)²) / 3 = 316.66%
The standard deviation is therefore equal to: V(316.66%)= 1.7794
The Sharpe ratio of the strategy is therefore: (0.30/0.10) / 1.7794 = 1.68
The strategy’s profitability is therefore good compared to the additional risk taken.
Let's take another portfolio example (3), this time with the following monthly performances and performance over the period of 12%:
- January: 20%
- February: -7%
- March: -1 %
The series average is (20-7-1)/3 = 4%
The variance of the portfolio is: ((20-10)² + (-7-10)² + (-1-10)²) / 3 = 170%
The standard deviation is therefore equal to: V(170%)= 1.3038
The Sharpe ratio of the strategy is therefore: (0.12/0.10) / 1.3038 = 0.92
The strategy’s profitability is therefore not good compared to the additional risk taken.
Portfolios 1 and 3 have identical performances. Both outperform the CAC40 but portfolio 3 does so at too high a risk. Portfolio 1 is therefore better than portfolio 2.
Portfolio 2 has much better performance than portfolio 1. Yet portfolio 1 is better than portfolio 2. Indeed, the out performance of portfolio 2 is made in return for a much greater risk.
To evaluate a trading strategy, several criteria should be considered.
Average Gain / Loss Ratio
This calculation allows the average risk/return ratio to be calculated. In other words, it tells us whether or not the average gain expectation is higher than the risk (stop loss spread). A ratio below 1 is obviously to be avoided. A ratio of 1 means that this strategy is balanced, with 50% winning trades. The higher the ratio, the more losing trades the strategy supports.
With a ratio of 2 for example, the trader only has to earn 1/3 for his trading strategy to be nil since his expected gain is twice as high as his risk. In order to further analyse the strategy, this can be done for each asset. Thus, a trader can remove those assets that are less profitable or that do not match his average risk/return ratio from his strategy. Ideally, a ratio should be at least 1.5 and tend towards 2.
Drawdown
Drawdown represents the maximum historical drop recorded on a trading account over a given period (day, week, month, year). It is the difference between the highest and lowest balance of the account. Drawdown is indicated in value and/or percentage. It represents the account’s volatility. The higher the drawdown, the riskier the trading strategy. If it is too high, the trader should reduce the size of his positions that are not suited to the amount of his portfolio. In this case there is too much leverage. A high drawdown does not necessarily mean that the account has been in loss. For example, a trader trades with €1,000 and a few months later, his account is at €1,500. Following a bad series, his account falls back to €1,200. His drawdown is 20% ((1500-1200)/1500) which is worth €300.
This drawdown must be related to his portfolio’s performance. Here again, there is a notion of profitability in relation to the risk taken. There is no perfect drawdown, everything will depend on the investor's profile and risk aversion. However, drawdown should not exceed the portfolio’s performance. If it does, then the strategy should be changed immediately.
Let us take several examples. A trader’s trading strategy has enabled him to achieve a 50% return over one year. Over the same period, his drawdown is 25%. This means he risks 1 to win 2.
Using another trading strategy, he only achieves 15% performance but his drawdown is 5%. Although the performance of this strategy is lower than the first, from a mathematical point of view it is much more efficient. Effectively, he risks 1 to win 3.
Sharpe ratio
The Sharpe ratio measures the return on the additional risk taken compared to a so-called "risk-free" asset. Without discussing risk-free assets (they do not exist, there is always a risk - see: government bonds), we are talking about benchmarks. We will therefore compare a strategy’s profitability to a benchmark index by integrating the notion of additional risk. For example, you could compare your strategy to the CAC40 index. Performance above that of the CAC40 does not mean that your strategy is good. Effectively, the marginal risk of your trading strategy must be taken into account.
The Sharpe ratio formula is S = (R - r) / E: where R is the rate of return of the portfolio under consideration, r being the chosen benchmark for comparison (generally the risk-free investment rate), and E the standard deviation of the rate of return of the portfolio under consideration.
A ratio greater than 1 means that your portfolio's out performance over the risk-free investment is not at the cost of too high a risk. The higher the ratio, the better the portfolio.
A ratio below 1 means that the portfolio under performs the risk free investment. You shouldn’t invest.
A ratio between 0 and 1 means that the excess return over the risk-free rate is lower than the risk taken.
To calculate the Sharpe ratio, calculate your portfolio’s standard deviation. Standard deviation is a measure of volatility and is the square root of the square average of the deviations from the average over a given period (or the square root of the variance). Let’s consider an example: Over the first three months of the year, the CAC40 posted 10% performance. Let’s suppose that portfolio 1 shows the following monthly performances with performance over the period of 12%:
- January: 10%
- February: 5%
- March: - 3 %
The series average is (10+5-3)/3 = 4%
The variance of the portfolio is: ((10-4)² + (5-4)² + (-3-4)²) / 3 = 28.66%
The standard deviation is therefore equal to: V(28.66%)= 0.5353
The Sharpe ratio of the strategy is therefore: (0.12/0.10) / 0.5353 = 2.24
The strategy’s profitability is therefore good compared to the additional risk taken.
Let's take another portfolio example (2), this time with the following monthly performances and performance over the period of 30%:
- January: 20%
- February: -15%
- March: 25 %
The series average is (20-15+25)/3 = 10%
The variance of the portfolio is: ((20-10)² + (-15-10)² + (25-10)²) / 3 = 316.66%
The standard deviation is therefore equal to: V(316.66%)= 1.7794
The Sharpe ratio of the strategy is therefore: (0.30/0.10) / 1.7794 = 1.68
The strategy’s profitability is therefore good compared to the additional risk taken.
Let's take another portfolio example (3), this time with the following monthly performances and performance over the period of 12%:
- January: 20%
- February: -7%
- March: -1 %
The series average is (20-7-1)/3 = 4%
The variance of the portfolio is: ((20-10)² + (-7-10)² + (-1-10)²) / 3 = 170%
The standard deviation is therefore equal to: V(170%)= 1.3038
The Sharpe ratio of the strategy is therefore: (0.12/0.10) / 1.3038 = 0.92
The strategy’s profitability is therefore not good compared to the additional risk taken.
Portfolios 1 and 3 have identical performances. Both outperform the CAC40 but portfolio 3 does so at too high a risk. Portfolio 1 is therefore better than portfolio 2.
Portfolio 2 has much better performance than portfolio 1. Yet portfolio 1 is better than portfolio 2. Indeed, the out performance of portfolio 2 is made in return for a much greater risk.
