CHAPTER 6: ANALYZING YOUR RESULTS

After backtesting your strategies, you need to analyze the results to see if they are profitable. In this chapter, we will cover some basic metrics that you can use to evaluate your strategy's performance.

Analyzing the results of your backtesting is a critical step in evaluating the effectiveness of your trading strategies. It helps you to identify strengths and weaknesses in your approach and to make any necessary adjustments to improve your strategy's performance.

There are several metrics that you can use to evaluate your strategy's performance, including the following:

1.       Profit and Loss (P&L): This is the most basic metric for evaluating a trading strategy. It measures the total profit or loss generated by the strategy over a specific period.

2.       Win Rate: Win rate is the percentage of trades that are profitable. A high win rate does not necessarily mean a profitable strategy.

3.       Turnover Period: It is typically calculated as the average number of days between two actions. A shorter turnover period is generally seen as more efficient, as it indicates that the trading strategy is more robust.

4.       Maximum Drawdown: Maximum drawdown is the largest decline in value experienced by a portfolio or investment strategy from its peak value to its lowest value.

5.       Sharpe Ratio: The Sharpe ratio is a measure of risk-adjusted return. It takes into account the return of the investment relative to the risk involved.

6.       Average Profit and Loss per Trade: This metric helps you to determine whether your strategy is making enough money per trade to justify the risk.

7.       Average Holding Time: Average holding time is the average length of time a trade is held. This metric can help you to optimize your strategy's holding period.

By analyzing these metrics, you can gain insights into the performance of your strategy and make adjustments as needed to improve its profitability. It is important to remember that no strategy is perfect, and there is always a risk of loss in trading. However, by backtesting and analyzing your results, you can reduce the risk and increase the chances of success.

In this demo version, we provide several simple metrics to help users understand some basic ideas about how to evaluate a trading strategy. For more complicated metrics, we will add them in future releases. In this chapter, we will use another trading strategy – “Golden Cross and Death Cross” – as an example to explain how to analyze backtesting results.

Golden Cross is a bullish technical analysis pattern that occurs when a short-term moving average crosses above a long-term moving average. For example, a stock's 50-day moving average crosses above its 200-day moving average. This pattern is widely used by traders and investors to identify potential buying opportunities.

The Golden Cross suggests that the momentum of the stock is turning bullish, as the short-term moving average is starting to rise above the long-term moving average. This can be seen as a bullish signal, as it indicates that the stock's recent price trends are starting to shift to the upside.

The "death cross" is a technical chart pattern that occurs when a stock's short-term moving average crosses below its long-term moving average, indicating a bearish trend. For example, the stock's 50-day moving average crosses below its 200-day moving average. This can be seen as a signal of increased selling pressure in the market and a potential downturn in the stock's price. Some traders use the death cross as a signal to sell their positions or to take a short position in the stock.

Let’s set parameter scanning range in “Scan” worksheet like this:

After running the VBA tool, we will get a series of parameter combinations in “Pick” worksheet like this:

The VBA Macro will pick the top one parameter combination (stored in cells AO2 to AS2) to print out a chart:

The Maximum Drawdown looks a little bit ugly. And we can review this parameter combination‘s performance metrics in worksheet “Pick”:

After a long run, since 2011-01-03 to 2023-02-23, which is a total of 3056 trading days, and with the following parameters set: Momentum at 0, Take Profit at 25%, Stop Loss at 6%, Short Term SMA at 20 days, and Long Term SMA at 50 days, we will get an Annual ROI at 18.56%, not bad, eh? During this period, there were 31 trade deals completed, with 14 of them being long positions and 17 of them being short positions. Stop loss was triggered 17 times, while take profit was achieved 14 times. The ratio of stop loss vs. take profit was 54.84% vs. 45.16%. On average, it took 98.6 days to complete one full round of buying and selling.

Now the ugly Drawdown and the Annual ROI at 18.56% (on par with Warren Buffett's performance) come together, can you accept that? Some people may enjoy high returns and can bear temporary big losses, while others may prefer stability and may not care if the profits are low. It depends on individual's investment philosophy; we cannot say whether it is right or wrong.

Profit and risk are inseparable in the market, and high profit comes with high risk, which is a normal situation in the real trading world. For example, investing in a startup company or a volatile cryptocurrency may offer the potential for high returns, but also comes with a high level of risk. Conversely, investing in a stable, blue-chip stock or a government bond may offer more moderate returns, but also carries a lower level of risk.

While this VBA backtesting tool can save time on calculations and charting, it cannot replace your own decision-making process and critical thinking.

Still remember the TSLA’s curve in chapter 4? Let’s review it here:

Since 2013-08-08 to 2023-01-24, the backtesting Annual ROI of TSLA using the “WhiteSoldiers_BlackCrows” VBA Macro has been as high as 39.97%, and it has gained a total of 2301.07%. This performance is not only better than Warren Buffett’s average of 17%, but also higher than the original Tesla stock’s ROI of 32.27% during the same period.

This return is based on the parameter combination stored in cells AO2 to AQ2 of the "Pick" worksheet: Momentum at 0.1, Take Profit at 20%, and Stop Loss at 12%, which is indicated by a red frame as shown below:

If you want to avoid a large drawdown, try using a different set of parameters indicated by the green frame in cells AO11 to AQ11. Write down these values, go to the 'Scan' worksheet, and enter them into the Take Profit, Stop Loss, and Momentum cells. In this case, set both the 'Take Profit % Minimum' and 'Take Profit % Maximum' to 9, and ignore the 'Take Profit % Step'. This setting tells the VBA Macro to run this specific parameter combination only once, without running all other loops. Similarly, set both the 'Stop Loss % Minimum' and 'Stop Loss % Maximum' to 9 (occasionally, here it is the same as the TP group, but in other cases, it may be other values), and ignore the 'Stop Loss % Step'. Then set both the 'Momentum % Minimum' and 'Momentum % Maximum' to 0.1, and ignore the 'Momentum % Step'. Now run the VBA Macro, the program will scan these settings particularly, and ignore other groups.

After pressing Control + Y, you will quickly receive a new result:

the Annual ROI drops to 27.44%, with 135 trade deals (almost double the number of trades than the top-gun setting). The Take Profit to Stop Loss ratio is now 60% to 40%. The drawdown…

also looks more comfortable and less severe, eh?

In conclusion, backtesting cannot accurately predict how much profit we will make in the future, but it can certainly calculate how much we have lost in the past if we have used a wrong trading strategy, or how much we have missed out on if we own a strong trading strategy but picked some weak parameters, heartbroken, eh?

As an example, let's consider TSLA's historical data, which we split into a Training set (from 2013-08-08 to 2020-03-23) and a Testing set. If we pick the parameter combination with serial number 229 and strictly follow the rules of the "Three White Soldiers and Three Black Crows" strategy, obeying every signal and plan to act from the Testing set's day one (2020-03-24) until 2023-01-24, which amounts to a total of 715 trading days…

we could have gained…

an explosive annual ROI of 92.56%！

Backtesting is a cold-blooded truth teller, use it wisely. The trickiest challenge is how to pick the right parameter combination, such as whether to pick serial number 229's parameters instead of the top one in Training set, which is serial number 231. We never know what will happen tomorrow, even if we have a powerful tool like backtesting. Ultimately, success depends on our own experience, mindset and perhaps even… fate, eh?

As you analyze the results of your backtesting, you might start to feel like a stock market Sherlock Holmes, but don't forget to keep it fun! Just like in life, sometimes you win, sometimes you lose, and sometimes you're just stuck in a holding pattern. But with enough persistence and analysis, you'll eventually uncover the clues to a profitable trading strategy. So grab your magnifying glass, put on your deerstalker hat, and let's solve the mystery of the stock market! (Disclaimer: No actual detective skills required.)