Anti Strategy - Linda Bradford-Rasche
|June 1st, 2013, 05:41 PM||#11 (permalink)|
Futures Experience: Advanced
Platform: NinjaTrader, MultiCharts
Broker/Data: Interactive Brokers
Favorite Futures: Keyboard
Posts: 9,525 since Mar 2010
Thanks: 4,213 given, 24,787 received
This does not look like the original Anti setup.
In her book Street Smarts Linda uses a Stochastic with a period_K = 7, a smoothing period of 4 and a period_D = 10. In her LBR trading manual she uses a period_D = 12 instead of 10.
The main idea behind the setup is to use the D-line as a trend indication. The established trend should cover at least 4 bars. Then the faster line, which is the smoothed K-line should move against the trend for at least 3 bars. Then the smoothed K-line should turn and align with the higher timeframe trend as represented by the D-line.
The idea behind this concept is to find and early entry after a failed countertrend move, or otherwise put, re-enter in the direction of the trend after a retracement.
The two entries shown on the chart are not "Anti" setups
The two entries do not match the requirements for an entry setup. For both entries the D-line does not support the entry
- first hook: the D-line is in a downtrend and thus in contradiction with a long entry
- second hook: the D-line is basically flat and does not support the long entry either
Weaknesses of the concept
The concept has a few weak points, in fact there are three of them.
(1) As Elliot disciples know, and as Al Brooks repeats all the time, corrections often have two legs. The Anti setup often will get you into a position too early - that is after the first leg has formed. There is a high probability that you will get stopped out during the second leg. Therefore it is recommended by Linda to exit the position quickly. Applied to intraday trades this means that the "Anti" setup is only good for scalping.
(2) Another weakness is that the trend definition only relies on 1 timeframe. I believe that there is a higher probability for the setups, if two higher timeframes are used. For example you could use two Stochastic D-lines rather than one to build the trend filter.
(3) The Stochastics itself is not a good trend indicator, if it is only judged by slope. A weakening downtrend - which can also be detected by a rising momentum or a divergence - will produce a rising Stochastics. It will therefore invite you to trade "Anti" setups too early, before the trend has changed. This is an old problem of trading, when trend, which is the first derivative of price is confused with slope of momentum, which is the second derivative of price. The second derivative of price is meaningless, if you do not take into account the first derivative.
This is one of the trading fallacies:
- a divergence means that a trend slows down, this does not mean that price will reverse
- when the MACD crosses the signal line, this does not indicate a trend change, the absolute value of the MACD
represents the first derivative of price, also needs to be taken into account
The "Anti" setup is therefore incomplete, because the first derivative of price has been omitted.
The Stochastics and the RSI are pretty similar as oscillators. Both of them have a fast and a slow line, so they can often be exchanged. For the RSI there are two oscillators available that come with two signal lines. This allows to take into account two higher timeframes rather than a single one.
One of those indicators is the Trader's Dynamic Index by Dean Malone, the other one is the CFG Momentum Oscillator developed by Andrew Cardwell (also know as Composite Index as described by Constance Brown). Both of them can be used to identify Anti-setups, if you change the parameters.
I use a modified normalized MACD with an additional trendfilter to find similar setups.