Last Updated on September 15, 2023 by Mark Ursell
Every trader should understand their edge in their trading. If we have an edge, we can make a profit trading the financial markets. In this article, I look at how to quantify this edge.
I analyse four different technical indicators and see how they have performed as a short-term entry signal on the S&P 500.
Video
If you would like more information about this test, feel free to look at the YouTube video I recorded.
A Good Trading Strategy
A good trading strategy must generate a profit. At a minimum, this profit must allow us to:
- Pay for all the losing trades
- Pay for the cost of trading
- Generate a return that is proportionate to the time and money invested in the strategy
- Generate a return that is proportionate to the amount of risk
When we are developing a trading strategy we obviously want to know that it has an edge. However, it can be difficult to be sure whether this edge is genuine or not.
What is a Trading Edge?
A trading edge is an advantage that we can use to make our strategy profitable. This advantage might be to identify high probability turning points or continuation areas. An edge might be a single factor or a combination of factors.
A trading edge will vary by market, by timeframe and by trading strategy. A trading edge might work on one market but not in another.
In this article, I am going to focus on getting an edge with a trade entry.
Looking at the Charts
Humans are great at spotting patterns. This ability helps traders to identify all sorts of shapes and patterns in their charts. This might be a technical indicator crossover, double-top or Japanese candlesticks.

However, our eyes are naturally drawn to the points where the price changes direction. We automatically filter out the times when the trading signal did not work.
Looking at charts is therefore great for identifying a potential edge. But not as good at telling us whether this edge has been able to beat the market over the long-term.
How to Calculate Your Edge
To calculate the edge we need to have some way of studying what has happened in the past.
One way to do this is to scroll through a chart and keep a list of what happened whenever our edge occurs. This is absolutely fine but it needs to be done very carefully and it takes a lot of patience.
I prefer to use some form of analysis software. I carried out my analysis using a Tradinformed Backtest Model. These are built using MS Excel and are a great way to test all sorts of different trading strategies. They can be used with technical indicators, Japanese candlesticks and chart patterns. They are quick to set up and allow you to test lots of different variables.
Technical Indicator Edge Analysis
I did a simple test of four popular technical indicators: Stochastic Oscillator, RSI, MACD and Bollinger Bands. If you want to know how to calculate these, I have an eBook available: 21 Technical Indicators.
The test was carried out on the S&P 500 on the daily timeframe between 1996-2015. I looked at long-only trades. To compare each indicator I set up a time close. In the first scenario I closed the trade after one day. In the second scenario I closed the trade after two days. In the third scenario I closed the trade after three days. The entry signals were:
- Stochastic Oscillator – Indicator crossed above the 20 line
- RSI – Indicator crossed above the 35 line
- MACD – Indicator crossed above the 0 line
- Bollinger Bands – Price crossed above the lower Bollinger Bands
Results
Metric | Stochastic | RSI | MACD | Bollinger Band |
---|---|---|---|---|
Net Profit | $-2,655 | $-3,279 | $-5,857 | $16,037 |
Profit Factor | 0.94 | 0.91 | 0.69 | 1.3 |
Winning Trade | 79 | 56 | 40 | 84 |
Losing Trades | 83 | 65 | 41 | 69 |
Win Percentage | 48.80% | 46.30% | 49.40% | 54.90% |
Metric | Stochastic | RSI | MACD | Bollinger Band |
---|---|---|---|---|
Net Profit | $-4,320 | $-14,038 | $-6,722 | $22,814 |
Profit Factor | 0.93 | 0.71 | 0.75 | 1.36 |
Winning Trades | 83 | 52 | 43 | 74 |
Losing Trades | 74 | 61 | 39 | 65 |
Win Percentage | 52.90% | 46.00% | 52.40% | 53.20% |
Metric | Stochastic | RSI | MACD | Bollinger Band |
---|---|---|---|---|
Net Profit | $4,503 | $-7,435 | $-3,598 | $28,951 |
Profit Factor | 1.05 | 0.85 | 0.9 | 1.47 |
Winning Trades | 92 | 50 | 45 | 71 |
Losing Trades | 62 | 56 | 37 | 59 |
Win Percentage | 59.70% | 47.20% | 54.90% | 54.60% |
Conclusions
The results showed that out of the four technical indicators, only Bollinger Bands showed a positive edge in the scenarios tested. To me this makes Bollinger Bands worthy of further investigation.
Regarding the other three technical indicators, I was surprised that the RSI was the worst performer. I expected it to do well in this analysis.
Overall, this test is a good starting point but much more work needs to be done to see whether these results are repeatable over different markets and different timeframes.
Purchase the Spreadsheet
If you are intrigued by this test you might want to do your own trading analysis. You could test your ideas about the 5-minute bars on the E-mini or a multi-year investment strategy.
If you would like to do this you can see the latest models in the Tradinformed Shop.
I just finished a Dip course in trading,I asked one of the lectures what he though of back testing,by his response I don’t think he has ever back tested,I find it surprising.I have started using your excel back testing,it’s not surprising the results as I have done back testing before .I use bollinger bands and psar,I don’t depend to much on RSI,I used to trade without any indicators ,thanks for the useful insight.
Hi Kieran, Thanks for commenting. This is only a limited test so I am not saying that the RSI is not a useful indicator. More analysis could show different results. I agree about backtesting, it is not perfect but it is better than the alternative of not testing. I am personally a fan of Bollinger Bands and PSAR as well. Regards, Mark.
Hi Mark, Just purchased your Spreadsheet. I have put my data on it but i am struggling to understand how to figure out how to backtest Long positions and short positions with my own rules….can you please give me some guidance please?
Hi Marco. Changing the entry rules is probably the hardest thing to get used to in backtesting. I have just published a new guide How to Use a Tradinformed Backtest Model. You might find this useful to help you understand the model.