What is a good trading strategy?
Thus, a good trading strategy is not necessarily a strategy that gives you a high CAGR, returns, or high gains per trade. Even a “mediocre” trading strategy might serve a useful purpose in a portfolio of trading strategies.
By diversifying into different time frames, assets, and market directions you get closer to the holy grain in trading. A good trading strategy might not be particularly good on its own, but fill a purpose in a portfolio of different trading strategies because of its correlation attributes.
First, there is no perfect trading strategy:
The first thing you need to understand is that the perfect strategy doesn’t exist. Many keep looking for the perfect indicator or the perfect setup, or they keep fiddling with one or just a few strategies to improve them.
They will ultimately be disappointed, or they end up with a curve-fitted strategy that doesn’t live up to live trading expectations.
The perfect strategy doesn’t exist and a good strategy is neither perfect nor profitable at all times. A good trading strategy complements your other strategies:
- Why build a portfolio of quantified trading strategies
- Is this the Holy Grail of trading?
- What does correlation mean in trading?
This article briefly discusses what we consider the main elements of a good strategy and how you should go about with your trading.
What are trading strategy parameters?
Trading strategy parameters are variables or inputs that traders use to define and customize their trading strategies. These parameters can include factors such as entry and exit points, stop-loss and take-profit levels, position sizing, risk tolerance, and indicators used for analysis. By adjusting these parameters, traders can adapt their strategies to different market conditions and optimize their trading approach for better performance.
Why do strategies stop working?
First, a party spoiler:
The reality is that strategies stop working, and you constantly need to fill in with new strategies.
Why do strategies stop working? They stop working for a variety of reasons, for example:
- Curve fitting. Your strategies might be too complex and have too many variables. This often results in curve-fitted strategies that are unlikely to perform well on future data.
- Markets inevitably change and evolve. An example could be the immense growth in data power that makes the market more efficient and probably also more short-term.
- Markets go in cycles.
- The market is immensely competitive. In 2020 hedge finds spent almost two billion on alternative data to gain an information edge over the market. Even cheap computers can run a lot of data and optimizations. Just a decade ago, this was not possible.
A general piece of advice:
Many opt to either focus on one strategy or one market. We believe this is the wrong approach.
First, one strategy is not enough, and you may risk curve fitting the strategy. How does this happen? It happens because you constantly test and tweak the strategy.
Second, it makes you vulnerable to change in market dynamics. If you trade one market, you are at the mercy of one or just a few strategies. Being a specialist in one market, for example the S&P 500, you are left with nothing if your strategy (ies) stops working.
Specialist knowledge is good, but it’s far better to spread your knowledge into other markets. Knowledge is something that comes with time and experience.
Remember this: a one dollar profit in natural gas is the same as one dollar profit in the S&P 500. This is pretty obvious, but yet many ignore this simple truth.
Instead of choosing a market, we suggest looking at as many markets as you possibly can. Start with looking at each market’s characteristics, all markets behave in their own manner, and any market might contain trading edges you were not aware of. Usually, it pays off to turn every stone in the search for edges. Take your time and be patient.
By trading many markets and strategies, you are most likely less exposed to strategy deterioration. Moreover, by trading many strategies, you are indirectly forcing yourself to make simple strategies:
Simplicity trumps all in trading:
My advice is that there are better ways to spend your time than searching for the perfect newfangled nuclear-powered indicator that works perfectly in past years’ markets….Keep it simple. Simple time-tested methods that are well executed, will beat fancy complicated methods every time.
During our two decades of trading and investing, we have worked or been in contact with a wide variety of traders and investors. From what we can see, the best traders are those who manage to keep trading simple.
What do we mean by keeping trading simple? Simplicity means keeping your strategies at a minimum number of variables and criteria. Trading edges based on 1-3 variables/conditions are more likely to stand the test of time than those based on a myriad of conditions: the more variables, the more likely you curve-fit.
You don’t need any fancy software either. As a small private trader, you’re not able to compete on speed and certainly not on sophistication. A trading platform like Amibroker or Tradestation is all you need. You don’t need to develop your own trading software, like many software developers, unfortunately, tend to do. As a rule of thumb, programmers and coders are bad traders.
It’s not the software that gives you an edge, but your thinking, mindset, and knowledge. Make sure you spend time on developing strategies, not on software.
A good trading strategy is not curve fitted
As mentioned above, the fewer variables in a strategy, the less likely you curve-fit the strategy. Many variables make a strategy inflexible and not prone to deterioration when markets change just a little bit.
Computer systems are ordered and only do what you tell them to do, while people behave in less predictable terms. And markets are driven by people and psychology in the short term, while fundamentals are the main drivers in the long term.
However, optimization is often smart when done correctly, something we will get back to in a later article.
A good trading strategy doesn’t necessarily work in other markets
If you are looking at a “one strategy fits all”, you are searching in vain. Something that works in the S&P 500, might perhaps be worthless in lean hogs, for example. A strategy on consumer stocks might be useless on oil stocks.
Don’t expect many strategies to work on other markets. Don’t ignore a strategy just because it doesn’t work on all or in any other market.
Quite the opposite, we consider this good. Why is this good? Because inefficiencies are normally unique for each market.
A good strategy has neither a perfect nor random equity curve
What is the equity curve? The equity curve is a graphical curve of the accumulated profits of your strategy.
The curve is one of the most important factors in a trading strategy as it gives you a visual impression of how the strategy(ies) performs.
You want a rising equity curve from the lower-left corner to the upper-right corner. But you don’t want a very straight one, and you certainly don’t want an erratic one.
Why wouldn’t you want a straight one? A very straight equity curve might indicate you have curve fitted, or the strategy has performed well in the past due to chance and randomness. The curve below is a perfect example of that:
Even worse, such an equity curve might be the result of coding errors in your strategy.
One other aspect of the equity curve is the difference between “closed trade equity” or “mark to market”. The first one only shows the results on closed trades, while the latter shows the unrealized losses and gains. The difference can be substantial. Below are two equity curves based on the same strategy:
The above is the closed trades equity, and the one below is the mark to market:
Quite a difference!
Opposite, you don’t want an equity curve that looks completely random either.
For example, a strategy might produce an average gain of 0.66%, which you find very good, but the equity curve might look like this:
Although the strategy is profitable, you don’t want to trade something like that.
To read more about equity curves, we recommend reading this article about the equity curve.
A good trading strategy has a profit factor above 2
We like to use the profit factor when we construct trading strategies, although the profit factor is partially a result of the equity curve. If the equity curve is smooth, then the profit factor most likely is high.
What is the profit factor, and how is it calculated?
The profit factor is the ratio between gross profits and gross losses: the cumulative profits of winners divided by the losers’ cumulative losses. We like to see strategies that produce a factor of at least 2. This gives you a margin of safety in case the strategy deteriorates.
Nevertheless, if you have an arsenal of many trading strategies running on automated software, don’t be afraid to lower the profit factor threshold to, for example, 1.8 if the strategy adds diversification to your portfolio of strategies.
Is a good strategy based on logic?
However, markets are complex and not easy to understand. The financial media writes every day about why the markets went either up or down. Most of this is utter bullshit. Hardly anyone knows why a stock or index goes up or down.
When a journalist calls some pundit about why the market was strong, the pundit has to answer something. He can’t say “I have no clue”. Most market action is just noise and randomness.
While we prefer to use strategies that we believe are logical, we sometimes add strategies that we don’t understand why work. Markets are sometimes not logical, and we accept that we only understand a small fraction of the action taking place every day.
Instead, we prefer to focus on the robustness of the strategies by using some simple tests, such as a visual look at the equity curve and the mathematical profit factor. However, they are somewhat correlated. The best test of robustness is the out of sample test:
A good trading strategy passes the out of sample tests:
A strategy should always be run out of sample. What is an out-of-sample backtest? Out of sample is when you test your strategy on data not used in the backtest.
This can be done in two ways:
- Split your dataset into two periods. If your dataset is from the year 2000 until December 2020, backtest and develop the strategy based on the data from the year 2000 until, for example, 2017. When you believe you have a tradeable strategy, test the data using the data fra 2018 until December 2020. This way, you are testing the strategy on unknown data, i.e. “out of sample” data.
- We believe the best out-of-sample approach is to test the strategy live using a demo account for several months. Paper trading is never a perfect substitute for live trading, but you get to see every trade and later assess the drawdown. This has the advantage of seeing with your own eyes how it performs over a period of several months, instead of a one-minute test on the software platform.
Which type of trading strategy is the best?
No type of trading is better than the other. A dollar made is a dollar made whatever strategy you are trading. Ultimately, what matters is that you are confident in the strategy and can execute it according to the rules.
The most important matter is that you understand why you trade the strategy, back-tested it, and tested it out of sample, and you are confident you can stick to the strategy in both good and bad times. This also involves having a sense of when the strategy needs to be stopped trading. 50% of marriages seemingly end in divorce, and likewise, most strategies come to an end.
What type of trading is most profitable?
One market is usually different from another one as we mentioned above. You can’t expect to use strategies from the stock market successfully in natural gas, for example.
Thus, no type of trading fits all markets, and don’t reject ideas just because it doesn’t fit all markets.
Day trading, swing trading, momentum, mean-reversion, or trend-following. It doesn’t matter. Be open, flexible, and agnostic. There is no right or wrong as long as you can make money.
A good trading strategy adds diversification to your portfolio
Correlation in trading is the cornerstone of the portfolio theory, or perhaps we should say uncorrelation. The main idea of any diversification is to spread your eggs in different baskets, and you don’t want to drop all your baskets at the same time.
Likewise, any trading strategy must add diversification and preferably be uncorrelated to the other strategies. If the strategy correlates, then you might actually increase the risk of your portfolio of strategies.
Does the strategy increase the opportunity cost?
And in the real world, you have to find something that you can understand that’s the best you have available. And once you’ve found the best thing, then you measure everything against that because it’s your opportunity cost.
– Charlie Munger
The quote above from the well-known investor charlie Munger sums up well the definition of opportunity cost. Why would you allocate resources to ideas and investments that are not your best?
Thus, it would be best if you always aimed to increase the opportunity cost. Always ask yourself: Does this strategy contribute both to diversification and profitability?
Why do most traders fail? (How to fail as a trader)
The market is not meant to produce many successful traders. The number of unsuccessful traders is always vastly bigger than successful ones.
In the short term, the market is a zero-sum game. No one tells you this before you start, but this is the simple truth of the markets. Yes, the stock market has an upward drift with a slightly positive return every day that long-term investors can gain from, but as a trader, you are more interested in exploring short-term patterns. This requires work and testing.
If you buy and hold, you can use this tailwind by compounding. But it requires patience. Unless you already have a big amount of capital today, you will need to invest for decades to make it compound.
Warren Buffett presumably has made 99% of his wealth after turning 50, 97% after he turned 65, to put this into perspective. The compounding effect grows bigger the longer you wait. However, by then, you might already have dropped dead.
Trading offers the opportunity to scale and build capital fast. Of course, only if you have profitable strategies. This is the lure of trading. Be careful.
Trading strategy examples (a list of strategies)
We have a pretty big list of trading strategies on this page:
All of them are written between 2012 and 2020, and many of them are still profitable today, while others have failed.
We have chosen to charge a small fee for our best trading ideas and edges:
Conclusion: What is a good trading strategy?
A good trading strategy is part of many different trading strategies. Even an average trading strategy can be a perfect fit together with other strategies!