Trading Systems - What are they & how to evaluate?

A lot of professional traders keep asking, what is my trading system? I actually have no clue what they are asking for. Can explain me what it is, and if if there is a system what is considered a good system? How do you evaluate it?

3 Likes

Tarzen, 

A Trading System can be anything that sets a predefined rule for all your trading actions. For example if you go by a simple 9 day and 21 day moving average cross over system ..

1) You Buy - When the 9 day MA Crosses over 21 day MA with 9DMA > 21 DMA

2) You Sell - When the 9DMA crosses over the 21 DMA with 9DMA < 21 DMA.

So if you follows these conditions to trade, then you pretty much have a trading system for yourself. 

Trading systems come in various shapes and sizes :-) The one above is a sitter..quite simple to follow. At the other end of the spectrum you find highly complicated quant based systems. 

Now the point is, irrespective which trading system you follow you need to have few stats to go by it, which can be generated by back testing..some of them on top of mind are...

1) How many trading signals did the system generated over the last 2 or 3 years period. Some people go back upto 10 years. You may want to check if the system is generating far too many or very few signals per year

2) How many were profitable. What has been the biggest profitable trade? 

3) How many resulted in a loss. What has been the largest loss making trade?

4) What is the maximum drawdown of the system ?

5) What is the risk of the system in terms of volatility?

6) How susceptible is the system to systematic shocks - stress testing for a black swan event.

7) How does the equity curve of the system compare with the benchmark such as Nifty or Sensex. Are you better of holding a NIFTY ETF instead?

8) If you have invested 5lac in your system what has been the absolute and CAGR return?

9) Does it require heavy duty infrastructure to deploy? If yes, are you profitable post expense, post tax?

If you are trader, you most certainly need a trading system to trade. Without which you will have a hard time spotting opportunities in markets!

Good luck. 

4 Likes

A trading system is a precise set of rules that automatically defines, without any human intervention, the entry and the exit on the markets. These rules can be on paper, in one’s mind, or can be coded as an algo.

If you want to code a trading system you need to have a software that easily performs all the programming and testing. Some of the popular coding scripts available for Indian traders are Amibroker’s AFL, Easy language and Tradescript. Coding a system let’s you run it over data and see when and where to apply them, that is, statistically test it. Examples of some trading systems are moving average crossover , ATR based etc.

Some of the basic parameters to evaluate a trading system are:

1) Average Profit- An important indicator of the goodness of a trading system is the average trade that is the net profit divided by the number of trades. Average trade tells us how much money we make or lose per trade. In absolute terms the average trade should be capable of covering slippage and commissions and then still leave some profit for the trader.

2) Percent of profitable trades- The percent profitable trades number expresses the number of winning trades out of the number of the total trades. It is important not by itself (a trend following system can have a low percent profitable trades number such as 35% and still be a viable system) but because it can be used to gauge how the system is balanced in relation to the average winning trade/average losing trade ratio. Trader needs to be comfortable with the percent of profitable trades to actually follow his system.

3) Profit Factor- Profit factor is a perfect indicator for comparing different systems or the same system plotted over different markets. Profit factor is gross profit divided by gross loss and basically reveals the size of gross profit in relation to gross loss, higher the better. Usually a healthy trading system has a profit factor of 2, an average winning trade/average losing trade ratio of 2 and a percentage of profitable trade’s number equal to 50%. But there are also good systems with a profit factor of between 1.5 and 2.5.

4) Drawdown -A broad definition of drawdown could be the largest loss or the largest losing streak of a trading system, whichever is the biggest. In a more graphical way we can depict drawdown as the dip in the equity line between a highest high point and the successive lower point before a new high is made.

Following is an example of trading system report with above parameters:

                                                                  Total Trades           Long Trades              Short Trades  

    

In the above Systems Report we start with initial capital of 100000.We have a Net profit of 1305725 over 3142 trades.

1) Avg. Profit/Loss- (Net Profit/Total No Trades)=1305725/3142 -- 415.57

2) Percent Profitable-(42.23%) -Total no of Winners/All trades-1327/3142----42.23%

3) Total Profit- Its the sum of all the profitable trades--3272390.00

4) Profit Factor-Total Profit/Total Loss--3272390/1966665=1.66

5) Max Trade Drawdown-Maximum Drawdown of a single trade.(17050)

6) Max System Drawdown-Total drawdown of the entire system.(48470)

7 Likes

Most common way to evaluate a system would be to backtest it, a good tip here would be to ensure you dont retrospectively optimize your systems. This could make all systems look profitable.

3 Likes

In my opinion, the best way eo evaluate a trading system is to look for the consistency of returns.

2 Likes

A successful trader is observant. You need to have a system. You need to have rules. You need to build a coherent trading system which is based on rules devised by your discretionary judgments by every day observations. For example, if you cannot deal with high risks you must make it a rule to deal solely with medium level of risk. If you are comfortable with commodity portfolio then mark it as another rule to adhere. In simple terms it means that you need to systematically plan before you invest.

The idea is to plan way in advance. The successful traders have always planned for the future. The best players do not blindly invest. Hence rules are must for a successful future in the trading business.

Your rules should help you gauge the right options the day there is a sudden 5% rise in the market. In case the market goes down by 10% you should be prepared to make the right moves. Once you get your rules and upload it to the trading software you are assured of well-calculated moves that never fail.

Your rules for entry and exit are crucial. They play an important while you face down grades. Even your rules to buy or sell should be accurate while losing trades across your portfolio. Your plan should be well-chartered, organized and precise before you invest. Success lies in planning and strategizing in advance. Trading systems have proven to be useful in planning and strategizing. They act as the perfect defense against market loss.

Everyone wants to know how to make money in share market? But no one looks into the future. Very few plan before entering the trading business. The tact lies in having a vision and planning for the future. You need apt trading systems that give you the edge over others. Trading systems have changed many lives. The right rules, trading system and vision is the perfect mix to trading success.

So the next time you decide to invest I hope you have a good trading system in mind.

Happy Trading with Trading System.

1 Like

In my opinion, the best ways to evaluate a Trading system is Monte Carlo Simulation and Walk Forward Optimization. Obviously, this is apart from the standard backtest report parameters like CAGR, Drawdowns etc.

Monte Carlo simulation is a process which performs repeated execution of pre-defined set of steps by adding randomness to the input parameters at each iteration. The results are noted down at the end of each iteration which forms the basis of probabilistic analysis of the desired result. In Trading terms, Monte Carlo simulation is performed to forecast the success of a backtested trading system. In order to ensure that your trading system is robust,  backtesting should be performed multiple times by adding variations to your trading rules or data. If it gives consistent result each time, then it has a higher probability of generating profit.

Read the below articles to understand step by step process to perform Monte Carlo simulation:

http://tradingtuitions.com/monte-carlo-simulation-in-trading-step-by-step-tutorial/

http://tradingtuitions.com/monte-carlo-analysis-in-amibroker/

Walk Forward Optimization is an important method to determine the robustness or credibility of your trading system. Like simple Optimization, it is also a process to determine the best parameters for your trading system. However, in Walk Forward Optimization the system is optimized for a particular period of data to find the best parameters, and the obtained parameters are tested on the data following that period (known as forward data). If the parameters look profitable in both the data sets, then the system is considered to be trustworthy. This process is iterated over multiple chunks of data to arrive at the best parameters. The basic purpose of Walk Forward Optimization is to avoid curve fitting in simple optimization.

http://tradingtuitions.com/walk-forward-optimization-and-testing-amibroker-tutorial/

You dont need trading system to trade, if you are a small trader who wants to earn not more than 1-5 lakhs month. Trading system must be required when trade million/billion of dollars to place to your 1000s of trades effectively and efficiently.

Manish, I strongly disagree with you. You are talking about earning about 1 to 5 lakhs per month through trading. Honestly out of my 10 years of experience I dont know anyone who earns that kind of money through trading consistently !
So much so that I believe its impossible. Besides, assuming you are the smartest kid on block the best you can generate is about 3% a month, which in itself is a dream…as it translates to 36% a year. So going by what you are saying…to earn 1-5 lakhs (lets keep it at 2.5 lacs) and assuming 3% return…you need to deploy a capital of 85laks! Forget making this amount if you don’t have a decent trading system! In fact if you dont have a trading system…banks provides one for free of cost which earns you 8-9% per annum…its called the Fixed Deposit my friend.

I think its a long way for me to go!

The only thing that I can understand on this table is: Initial Capital, Ending Capital, Net profit and Net profit%. Can you take one of the cases and explain me how the rest of the values are calculated?

Thanks for the explanation :slight_smile:

A very comprehensive answer, Would require some concentration to understand, Thanks

You can understand these parameters here http://tradingtuitions.com/understanding-amibroker-backtest-report/

Agree and don’t agree at the same time Nachiket :-)…what if you have a trading system that consistently gives you 5% return on a trade but identifies 1 opportunity per year ?

Indeed, you’re right, check for consistency should be on statistically significant number of trades.

Rather, better yet is walk forward with optimization for consistency. ha?

isn’t Walk forward more closer to speculation, since it cant take into effect market events.

By walk forward, I mean optimization not paper/live trading.
http://en.wikipedia.org/wiki/Walk_forward_optimization
https://www.amibroker.com/guide/gifs/walkfwd2.gif
It’s sort of multiple backtests of your system. It gives you a good idea if your system is really robust.