We all know there are no magic indicators but there is one that certainly acted like magic over the past 10 years or so. What indicator is it? Our reliable RSI. In this article we are going to look at two trading models that were first talked about in the book, “Short Term Trading Strategies That Work” by Larry Connors and Cesar Alvarez. It has been well established in various articles that a 2-period RSI on the daily chart of the stock index markets has been a fantastic tool for finding entry points. Sharp price drops in the S&P E-Mini futures during bullish markets have historically (since the year 2000) been followed by reversals. These reversals can often be detected by using the standard RSI indicator with a period value of two. Place this indicator on a daily chart and look for points when the indicator falls below five, for example. These extreme low points are buying opportunities.
RSI(2) System
We can turn this into a simple trading model to test the effectiveness of the RSI(2) indicator on the E-mini S&P. In short, we wish to go long on the S&P when it experiences a pullback in a bull market. We can use a 200-day simple moving average to determine when we are in a bull trend and using a 2-period RSI to locate high probability entry points. We can then exit when price closes above a 5-day simple moving average. The rules are clear and simple:
- Price must be above its 200-day moving average.
- Buy on close when cumulative RSI(2) is below 5.
- Exit when price closes above the 5-day moving average.
- Use a $1000 catastrophic stop loss.
The system backtest was performed from September 1997 through March 2012. A total of $50 for commissions and slippage was deducted per round trip. Below is a chart of what this system would look like along with the system results.
RSI(2) System Results
Net Profit: $17,163
Percent Winners: 67%
No. Trades: 64
Ave Trade: $268.16
Max Drawdown: -$5,075
Profit Factor: 1.90
These results are great considering we have such a simple system. This demonstrates the power the RSI(2) indicator has had now for well over a decade. Just with this concept alone you can develop several trading systems. For now, let’s see if we can we improve upon these results.
Accumulated RSI(2) Strategy
Larry Conners adds a slight twist to the RSI(2) trading model by creating an accumulated RSI value. Instead of a single calculation we will be computing a running daily total of the 2-period RSI. In this case, we are going to use the total of the 2-period RSI for the past three days. When you keep an accumulated value of the RSI(2) you smooth out the values. Below is a chart comparing the standard 2-period RSI indicator with an accumulated 2-period RSI indicator. You can see how much smoother our new indicator is. This is done to reduce the number of trades in hopes of capturing the quality trades. In short, it’s an attempt to improve the efficiency of our original trading model.
The rules are:
- Price must be above its 200-day moving average.
- Buy on close when cumulative RSI(2) of the past three days is below 45.
- Exit when RSI(2) of the close of current day is above 65.
- Use a $1000 catastrophic stop loss.
Accumulated RSI(2) System Results
Net Profit: $17,412
Percent Winners: 67%
No. Trades: 52
Ave Trade: $334.86
Max Drawdown: -$4,850
Profit Factor: 2.02
S&P Cash Market
What would the 2-period RSI system look like trading 100 shares of the S&P cash market going back to 1993? It does rather well.
Conclusions
So which one is better? The accumulated strategy worked as intended. It increased the efficiency of the standard RSI(2) trading model by reducing the number of trades, yet produced about the same amount of net profit. As a bonus, the drawdown was slightly smaller. While both systems do a fantastic job, the accumulation strategy may do a slightly better job. The Accumulated RSI(2) strategy will work well on the mini Dow as well as the two ETFs, DIA and SPY.
The EasyLanguage code is available below as a free download. There is also a TradeStation workspace. Please note, the trading concept and the code as provided is not a complete trading system. It is simply a demonstration of a robust entry method that can be used as a core of a trading system. So, for those of you who are interested in building your own trading systems this concept may be a great starting point.
Both strategies look the same to me and there is no cimmulative RSI there, just a regular RSI. Furthermore, 52 trades in 15 years is far from a significant sample.
Rick, I just updated the the text file. It was missing the cumulative RSI strategy. Thanks.
Thanks for the simple system. I do use RSI and love them. Rick why don’t you backtest the last 100 years and let us know. 15 years is pretty darn good! Thanks again for sharing!
[…] by Larry Connors’ Cumulative RSI(2) (found in this post), I decided to test out how a Cumulative DV2 indicator would work. This is a frictionless test […]
[…] by Larry Connors’ Cumulative RSI(2) (found in this post), and the results of my Cumulative DV2 (found in this post) I decided to test out how a […]
After reading your article it left me wondering how $1000 stop-loss ensured when buying at closing price?
No stop is ever ensured. This is part of the risk of holding trades when the market is closed.
That means that your simulated returns are incorrect and the system is not profitable at all.
Can you explain why the simulated returns are not correct and what the true simulated results are based on your runs?
Because one can’t control the gap between closing and opening price the simulation you offered might not be profitable at all. Please run the simulation at opening price instead of closing price.
If the price gaps below our stop we will be taken out at a loss larger than our stop. Removing the stop produces “better” results. I’ve also been trading a similar system which drills down to the 5-minute chart. I assure you, it’s profitable. I encourage you to test the concept on your own platform. Thanks!
Or $1000 stop-loss condition needs to be removed when running at closing price.
For those using TC2000 or Telechart, this is from Bruce at worden.com regarding accum RSI(2):
This is very simple if you want the running total of a non-Wilder’s Smoothed RSI:
3 * RSI2.3
Have you testing for 15 minutes and why you go out at 65
Thanks
I’ve not testing a 15-minute time frame. The exit at 65 is not an optimized number – I just picked what I thought was reasonable. Of course you are free to test and modify the parameter to your liking.
Hi Jeff, I’m a novice trader and i’m wondering how can we actually “Buy on close when cumulative RSI(2) is below 5.” as stated for the rules in RSI(2)??
Since we cannot predict today’s closing price, does it mean buying tomorrow’s stock at today’s closing price?
Thank you 🙂
Good question. If you trade futures, things are a bit easier. I mainly trade futures and placing orders at the end of the day is simple. The market closes, your computer program calculates the results and places the order. Because the futures markets are open after the daily session ends, your order is filled. For stocks within TradeStation, you could simply have your system calculate the results of the RSI(2) 5-minutes before the close of the market. This can be done via programming or adjusting your session times on your trading platform. Your order is place before the close and your order is filled near the end-of-day price. Hope that helps.
Thank you so much for the reply. This definitely helps a lot!
Jeff, thanks so much for posting this. Love your blog.
Re this system, I see that the eld is Long only. Have you tested this going short as well over 90?
Hello JB. Sorry for the long delay in getting back with you. I overlooked your comment. I have done some testing with going short and it was not very productive. However it might be worth looking into again, as it’s been many years. Thanks for the idea for a future article!