השכלה

INDEX:BTCUSD   ביטקוין / דולר אמריקאי
What makes a profitable automated strategy?

Probably the biggest misconception for trading perpetuated in the mainstream is that you need to have a greater than 50% win rate to be profitable.

This is followed by a close second, of constantly assuming you need to have a risk-reward ratio of greater than 1:1 (e.g. 1:2, 1:3 etc). This one is perpetuated mostly by forex and stock market gurus.

By the end of this article, I hope to dispel these myths and aim to shed some truths on how to assess a profitable strategy.

Why your win rate doesn't matter:

Let's simplify this down using an example. Consider the following two strategies. Which one would you rather trade?

• Strategy A: 50% win rate - When you win you make 2 dollars, but when you lose, you lose 1 dollar
• Strategy B: 50% win rate - When you win you make 5 dollars, but when you lose, you lose 1 dollar

This one was a very obvious case of choosing Strategy B. In this case, both strategies have the same win rate, but Strategy B nets you 5 dollars per win, whereas Strategy A only makes you 2.

Let's take another example. A little less obvious this time. Which one would you rather trade here?

• Strategy A: 90% win rate - When you win you make 1 dollar, but when you lose you lose 50 dollars
• Strategy B: 10% win rate - When you win you make 200 dollars, but when you lose, you lose 1 dollar

Now the 90% win rate strategy may look attractive on the surface, but when you dig into it, you realise that you could get a massive 50 dollar loss in the 10% of times you do lose! For those of you who chose strategy B, this is the correct answer.

One way we can assess the above strategies is using Expectancy. The formula for Expectancy is as follows:

(Win % x Average Win Size) – (Loss % x Average Loss Size)

We can calculate the expectancies of the strategy below:

Strategy A:

(0.9 * 1) - (0.1 * 50) = -4.1

Meaning you are expected to lose an average of \$4.10 per trade using strategy A. Not a good sign.

Strategy B:

(0.1 * 200) - (0.9*1) = 19.1

Meaning you are expected to win an average of \$19.10 per trade using strategy B. This is a major winner here!

As you've probably realised. It is possible to have a profitable strategy using a low win rate. Many trend trading/breakout strategies tend to have lower win rates, but with larger rewards to risk, whilst mean-reversion strategies tend to have higher win rates with less frequent but larger drawdowns.

The backtest shown in this post shows an example of a low win rate, and high win amount strategy using the Smoothed Heikin Ashi Trend on Chart indicator which I have developed, with an overall positive expectancy, backtest (note, no strategy is perfect, should not just blindly trust backtest data).

Why you may still choose to define a risk/reward

• Better consistency of your strategy
• Psychological factor of knowing that you can be expected to lose only x amount (assuming no slippage etc)

As an aside, note that defining a fixed risk-reward may hurt your win rate (which could impact your expectancy) so it's important to backtest to see if you get better results with defined risk-reward parameters. This is beyond the scope of the current article, but an important consideration.

Why do traders gravitate toward a higher win rate?

The simple answer here is that everyone wants to be a winner! It's human nature to want to be right, whether this be about a market direction or when to open or close a trade. It's often easier to brag about how much you win whether that be on social media or just feeling good about yourself.

For algorithmic traders, having a higher win rate may also provide psychological benefits, as losing 20 times in a row can sometimes be very daunting for traders and can throw doubt into the efficacy of your system.

I hope that through this article, I have managed to convey that it may be prudent consider strategies with low win rates also, as these can be very profitable in their own right.

Digging further:

This article is only just scratching the surface of how to create and validate if a strategy is something that you should consider trading. There are many aspects of backtesting including Monte Carlo simulation, understanding standard deviation of returns and risk, Sharpe ratio, Sortino ratio, walk-forward analysis, and out-of-sample analysis to name a few that you should conduct before you evaluate a strategy as suitable for live trading.

If you've made it this far, thanks for reading. If you like the content, feel free to like and share, as well as check out some of the free scripts, strategies and indicators that I have published under the scripts tab.

Thank you!

Disclaimer: Not to be taken as financial advice, anything published by me is purely for education and entertainment purposes
כתב ויתור

המידע והפרסומים אינם אמורים להיות, ואינם מהווים, עצות פיננסיות, השקעות, מסחר או סוגים אחרים של עצות או המלצות שסופקו או מאושרים על ידי TradingView. קרא עוד בתנאים וההגבלות.