This week’s idea comes from Sintia, who has been experimenting with SuperTrend and Parabolic SAR technical indicators over the past year. Parabolic SAR is a very cool indicator that was invented by the great J. Welles Wilder, Jr.. Parabolic SAR (SAR = stop and reverse), or PSAR as I like to abbreviate, is used to find market reversals. It is early to identify a trend change, but it generates plenty of false signals.
SuperTrend utilizes ATR (Average True Range) calculations to identify trends. I first learned about SuperTrend on Futures.io, but I do not know the origins of this indicator and my research turned up nothing. If you know, please leave a comment below, as I love learning the history of technical analysis. SuperTrend has a lot of lag and is slower to market changes, but the signals are usually good. The idea is that SuperTrend and PSAR balance each other to (hopefully) give better trading signals.
Thanks to Sintia for this trading idea! Let’s see how it goes…
Phase 1: Plan & Design
1. Trading Idea
The big idea is to use the SuperTrend and Parabolic SAR indicators to provide trading signals. This system goes long or short, but I am going to include options for long only, short only, and both, so that I can test on equities in the future. SuperTrend uses an Average True Range (ATR) to determine the overall trend and the PSAR is used to confirm entries. I for this system will use an ATR-based stop loss.
2. System Definition
- Futures: 1 contract
- Forex: 10,000 lot size
- If SuperTrend is ‘up’ and Parabolic SAR is transitioning to ‘up’, then go long
- If SuperTrend is ‘up’ and Parabolic SAR is ‘up’, then go long
- If SuperTrend is ‘down’ and Parabolic SAR is transitioning to ‘down’, then go short
- If SuperTrend is ‘down’ and Parabolic SAR is ‘down’, then go short
- Stop loss amount = ATR * ATR Multiplier * Full Point Value * Number of Shares/Contracts
- Long exit: exit when Parabolic SAR indicates a shift to a short trend
- Short exit: exit when Parabolic SAR indicates a shift to a long trend
Profit Target: None
- Trade_Direction: long, short, or both
- AFStep: Acceleration factor for Parabolic SAR
- AFLimit: Acceleration factor limit for Parabolic SAR
- ATR_Length: number of bars for SuperTrend’s ATR calculation
- ATR_Multiplier: number multiplied by the ATR as part of the SuperTrend calculation
- Median_Periods: the number of bars for SuperTrend’s average median calculation
- ATR_Stop_Length: number of bars for ATR calculation used in stop loss
- ATR_Stop_Multiplier: multiplier used to calculate the stoploss amount
3. Performance Objectives
The system will meet the following objectives:
|Strategy Type (trend, mean-reversion, |
day, swing, etc.)
|Risk of ruin||0%|
|Profit Factor||>= 1.5|
|Win Percent||>= 35%|
|Max Drawdown %||< 35%|
|Profit/Drawdown Ratio||> 2.0|
This idea is S.M.A.R.T.: Specific, Measurable, Achievable, Realistic, Time-bound
4. Market Selection
|Energies||Crude Light, Natural Gas, Gasoline, Heating Oil||CL, NG, RB, HO|
|Currencies||Futures: Euro FX, Swiss Franc, Japanese Yen|
|EC, SF, JY EUR/USD, UD/NZD, CAD/GBP|
|Fixed Income||30 Year US Treasuries, Eurodollar||US, ED|
|Agriculture||Corn, Soybeans, Wheat, Lean Hogs||C, S, W, LH|
|Metals||Gold, Silver, Copper||GC, SI, HG|
|Softs||Cocoa, Sugar||CC, SB|
|Indexes||E-minis: S&P, Dow, Russell 2000, Nasdaq||ES, YM, RTY, NQ|
|Equities||Index ETFs||DIA, QQQ, SPY|
Chart Type, Timeframe, Session, Time Zone:
|Chart Type||Regular Candlestick||Charting is only useful for validating entry and exit signals|
|Timeframe / Interval(s)||Daily > all instruments |
60 minute > forex only
Phase 2: Build
5. Manual Test
Pass, using energies.
Not much to say here. Sintia provided very clear trading rules for me to follow.
7. Unit Test
Note: Unit test verifies that the system is executing the trading rules correctly. It is, essentially, quality control.
Phase 3: Test
Complete. I optimized these two parameters:
- ATR_Stop_Length : Range 1 to 20, step 1
- ATR_Stop_Multiplier : Range .5 to 4 , step .5
9. Walk-Forward Analysis
I performed WFA on all instruments and everything failed (equities and forex included), with a few exceptions. I have three instruments on the Daily timeframe that may work.
The big surprise here Sugar (SB). I have not yet developed a working ‘softs’ trading system, so maybe I have something here.
All other symbols, including Forex failed epically. I will advance these three to Monte Carlo simulation.
10. Monte Carlo Simulation
The exceptionally low Profit Factor for each system led me to believe that these would not pass Monte Carlo, but I still ran the test. I have been known to be wrong from time to time. 😀
They all failed. If you recall, one of my performance goals was Return/DD of greater than %2.0. This system did not meet the goal.
Even though Monte Carlo failed, I looked at the 6 month incubation for Crude (CL):
- +$377 – Closed Trades
- +$8,272 – Open P/L
- -$13,000 drawdown
The system picked up the recent rally in Crude but had a long list of stop losses. I will discuss in further detail in the Commentary section.
Phase 4: Deploy
Did not make it this far.
Notes and Commentary
This system had some promise but failed for most instruments during the optimization and walk-forward analysis steps. The remaining survivors did not do well. Why?
I suspect the exits are at fault. I optimized the ATR for stop loss, but in the case of the incubation period for Crude, the idea failed because the stop loss (optimized) was extremely tight for that walk-forward period. There was a string of 22 consecutive stop losses on Crude. Ouch.
The other exit problem is that Parabolic SAR (PSAR) exits are notorious for getting us out of a good trend too early. I have done a lot of work on Parabolic SAR and it is great for getting into a move early, but it generates a lot of false signals. In this case, I think an alternate type of exit, such as ATR trailing stop, may work.
I am not a huge fan of stop losses, except where they may mitigate a catastrophic loss. I am continuing to develop an understanding of ATR stops, which I will introduce in some of our future systems.
Areas for improvement and future trading system ideas:
- Do not optimize the ATR stop loss
- Replace PSAR stop with a trailing ATR stop
- Optimize SuperTrend’s ATR parameters
Trading System Result: FAIL
Did you like what you read? Do you want to see more? Subscribe now and receive our weekly email, with the Trading System Idea of the Week and other fun stuff.
Don’t worry, we will never, ever, ever sell, overuse, or donate your email address. Promise.