This week’s idea is taken directly from John Ehlers’ third book, Cybernetic Analysis for Stocks and Futures, the Instantaneous Trendline, or iTrend as a prefer to call it. In the simplest terms, the iTrend is supposed to be a zero-lag trend indicator, akin to a moving average. I will not get into the math, but Ehlers tried to create an indicator that accounts for both trend and cycle (or range/consolidation).
I have spent a considerable amount of time trying to find a use for the iTrend indicator. Several years ago, I attempted to use the iTrend in various systems, but my testing and optimization process was a hot mess, and everything failed. Ehlers did not include the iTrend in his later writings, so I am wondering if he abandoned this for some of his better methods of technical analysis.
Ehler’s provided the code for the trading system in this book, so I am going to take the exact system and put it through our testing process. This week’s idea is a little different in this respect, where the code is all but complete, so the build phase is done. I really want to see if this system still works and to see how well it aged. As my son says, “Let’s go!”.
Phase 1: Plan & Design
1. Trading Idea
We are using two values, which you can imagine as two moving average lines: Trigger and iTrend. The idea is simple:
- Buy when Trigger crosses over iTrend
- Sell short when iTrend crosses over Trigger
- If we somehow get on the wrong side of the move, reverse position if we are a certain percent (Reversal %) away from the Close price:
- If Long and Close < (Entry Price / Reversal % ), sell short
- If Short and Close > (Entry Price * Reversal %), buy
- Profit Target: None
- Stop Loss: Yes, Ehler’s suggested using one for currency futures, so I will use for all
2. System Definition
- Futures: 1 contract
- Equities: 1 share
I will use the following input parameters:
|alpha(.07)||double||Yes||I think this is a component of the digital signal filter, but I do not completely understand the mathematics behind it. I know this is used to calculate iTrend.|
|RngFrac(.35)||double||Yes||Used to calculate our entry price (limit order)|
|RevPct(1.015)||double||Yes||If price moves against us by this percentage, then reverse our position|
|StopLoss(100)||double||Yes||Optimization ranges will vary based on instrument.|
|iTrend||Double||0||This is the Instantaneous Trendline: (alpha – alpha * alpha / 4) * Price + .5 * alpha * alpha * Price – (alpha – .75 * alpha * alpha) * Price + 2 * (1 – alpha) * iTrend – (1 – alpha) * (1 – alpha) * iTrend Note: [N] = the value of the variable N-bars ago|
|Trigger||Double||0||2 * iTrend * iTrend|
- If Trigger crosses over iTrend then buy next bar at (Close – (RngFrac x (H – L)) limit
- If iTrend crosses over Trigger then sell short next bar at (Close + (RngFrac x (H – L)) limit
- If Long and Close < (Entry Price / RevPct) then sell short next bar at Open
- If Short and Close > (Entry Price * RevPct) then buy next bar at Open
- Exits only on stop loss
3. Performance Objectives
The system will meet the following objectives:
|Strategy Type (trend, mean-reversion, |
day, swing, etc.)
|Risk of ruin||0%|
|Profit Factor||>= 1.3|
|Win Percent||>= 30%|
|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
|Currency||Futures: Euro FX, Australian Dollars, Canadian Dollars, Japanese Yen, British Pound||EC, AD, CD, JY, BP|
|Agriculture||Soft Red Wheat, Oats, Lean Hogs, Soybeans, Corn||W, O, LH, S, C|
|Metals||Silver, Gold, Copper, Mini-Silver||SI, GC, HG, YI|
|Equities||Random S&P components||AEP,BA,BEN,BHGE, BMY,CAT,CL,COP, CPB,CVS,CVX,D,DE, DHI,DHR,DRI,DTE, EBAY,ED||All of these were on the S&P 500 January 1, 2010 until now.|
|Fixed Income||U.S. 30-Year Treasury Bonds||US|
|Index Futures||E-minis: S&P, Dow, Russell 2000, Nasdaq Micro E-Minis: S&P, Dow, Russell 2000, Nasdaq||ES, YM, RTY, NQ MES, MYM, M2K, MNQ|
Chart Type, Timeframe, Session, Time Zone:
|Chart Type||Regular Candlestick||Charting is only useful for validating entry and exit signals|
|Timeframe / Interval(s)||Daily, 720 minutes, 360 minutes|
Phase 2: Build
5. Manual Test
It works. The main attraction with iTrend is how long it will stay in a trend. The big problem is with consolidation and ranges. Again…it works, but is it profitable?
7. Unit Test
Note: Unit Test verifies that the system is executing the trading rules correctly. It is, essentially, quality control.
Phase 3: Test
Before getting to the results, I want to mention trading costs and the importance of these to optimization and walk-forward analysis. Trading costs, specifically commission and slippage, are always added to my systems during optimization and walk-forward analysis. The reason is simple: there are costs to trading and there is no such thing as ‘zero commission’ (you too, Robinhood clients). Your broker will make money on your transaction, one way or another.
For futures, trading costs include NFA fees, exchange fees, and commission. For example, a $.99 broker’s commission might really be $2.55 for one contract and one trade when you account for all of the fees involved. Continuing this example, one futures contract bought and then sold will cost is $5.10. Slippage is another component. If you are using market orders, your slippage is automatically the difference between the ask and bid price. I will not get into the nitty gritty, but slippage is generally several ticks per trade. The whole point is that I add generous amounts of slippage and commission to all my testing, regardless of futures, equities, or forex trades. This allows me to have more accurate results if a system passes all tests. This also explains why some of my systems do not pass. Rest assured, the results I show always have commission and slippage built into them. I do not see any point in presenting any hypothetical trading results without them included
For this system, I optimized these parameters:
- alpha: Range .03 – .11, step .01
- RngFrac: Range .15 – .35, step .05
- RevPct: Range 1.005 – 1.03, step .005
- Futures: Range 500 – 3,000, step 500
- Equities: Range 100 – 1,000, step 100
9. Walk-Forward Analysis
Not much passed, but we have a few candidates for Monte Carlo:
- NQ Daily,
- MNQ Daily,
- YI Daily
I have serious doubts that these will work, mainly because I have been working with iTrend for a few years, so this is what I have come to expect. It will be interesting to see if E-mini Silver (YI) works, given that regular SI (silver) did not make it this far.
I want to show you one thing that I look at: Winning vs. Losing Series. Below is MNQ.
For a trend following strategy, a lot of losses can be expected, which is why the win rate is so low (31% in this case). However, I like to have as few losing series as possible. Can I psychologically handle 20 losses in a row without pulling the plug? This is generally an early warning sign that things may go haywire.
On that happy note, all other instruments failed.
10. Monte Carlo Simulation
|Instrument||Equity Required||Ruin||Median Drawdown||Median $ profitability||Median Return||Return/DD||Probability of Profit in 1 year > 0|
My goal of Return/Drawdown ratio >-= 2 was not met for any of these instruments.
We did not make it this far.
Phase 4: Deploy
We did not make it this far.
Notes and Commentary
I give up.
I have tried using Ehler’s Instantaneous Trend, the iTrend, to no avail. This is the first time I put it through my standard process of system development. John Ehler’s wrote some additional articles on iTrend (see Sources/References), but his later writings and his last workshop in late 2020 make no reference to it. I did some additional analysis, not included here, on forex, but with nearly identical results.
One litmus test I use for ideas based on technical analysis methods is Perry Kaufman’s Trading Systems and Methods, 6th Edition. If an idea is not in the book, it probably did not perform well enough for inclusion. There are plenty of Ehler’s technical analysis tools in it and I know Kaufman has a deep respect for Ehler’s contributions to technical analysis, as I do. I think I will reference this book when utilizing more obscure technical analysis methods (and I will be presenting some of those obscurities).
I had to get this iTrend monkey off my back, once and for all. I do not think I will look at this indicator again. It has been like a love-interest who breaks your heart, again and again and yet again. She is alluring, but it is a dead-end relationship. I thank you for sticking with me through to the end here.
Trading System Result: FAIL
- Cybernetic Analysis for Stocks and Futures, John Ehlers
- Modeling the Market = Building Trading Strategies, John Ehlers, Stocks & Commodities (TASC) Magazine, v24:8
- Trading Systems and Methods, 6th Edition , Perry Kaufman
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.