<--- GO BACK TO THE BITS TIMERS FAMILY MAIN PAGE
Before Gerald Appel worked on his Breadth Impulse theories, Martin Zweig had noted the use of the EMA of A/(A+D)*100 signal line in what later became to be known as the "Zweig Breadth Thrust" (ZBT) in quantitative finance circles.
A ZBT occurs when the signal line rises from below 40 (oversold) and reaches 61.5 (overbought) within a 10 trading day window. This is a rare market momentum turning point with deadly accuracy. According to Dr. Zweig, there have only been fourteen Breadth Thrusts on the US markets since 1945. The average gain following these fourteen Thrusts was 24.6% in an average time-frame of eleven months. Dr. Zweig also points out that most bull markets begin with a ZBT Thrust.
Quantitative finance circles like to refer to Appels' BITS signals (when the signal reaches 60 from anywhere) as "Breadth Continuation signals" and Zweigs' signals (the rise from 40 to 61.5) as "Breadth Thrusts".
When researching the nature of ZBT signals for the JSE we noted that periods when the BITS signal dropped below 44 and then rose within 18 trading days to 54 did a pretty damn good job of picking up market troughs over the last 15 years, as shown in the chart below:
We hadn't seen anything this good since Fosback or Big Dipper, the important observation being that bull-traps in all significant bear markets were deftly avoided whilst the system had a high frequency of signals.
We then sought to replicate the trading system development we adopted for our original 4 BITS timers, but this time using these trough signals as entries as opposed to waiting for the signal to reach 60 before opening a trade. The logic was these trough entries would be superior entries to the original BITS entries which were entries on extraordinary strength (which could sometimes be quite a distance from a troughs' ground zero). For the EXIT signal, as with the BITS family of timers, we would seek out a threshold below which the signal line falling would trigger a close of the long trade. This would give us a trading system with clear, unambiguous, mechanical entries and exits.
We discovered not one, but TWO highly successful trading strategies deploying this technique, which we shall dub ZWEIG BITS ACTIVE and ZWEIG BITS LAZY-BOY respectively. The trade statistics from 5th Dec 1997 to 25th March 2011 appear below and including interest at prime less 4% during non-vested periods. When compared to those trade stats we discussed in the original BITS research note, they are simply mind boggling:
The ACTIV system uses a 18-day rolling window to determine swings in the BITS signal between its low and high thresholds and the LAZY system uses a shorter, stricter 12 days. Both Zweig Thrust derivatives of the BITS timers achieve a 24% compound per annum growth over 14 years versus the JSE's 12.35% per annum. Their difference is how they get there, with "ACTIVE" conducting far more frequency of trades (39 versus 22) and being vested for only 55% of the time (versus 69% for the LAZYBOY system). Whilst both have respectable gain/loss ratios the LAZYBOY at 16.88 is more along the lines of the original BITS timers, obviously due to the higher win ratio of 81.8% as a result of the longer trades.
Both systems use 44.41 for the "arming signal" (the lower threshold depicting oversold market) and have slightly different figures for the BUY trigger that makes up the swing threshold. They differ in the EXIT signal though with the ACTIV system quitting when the signal gets to 44.41 during an open trade and the LAZY system quitting when the signal gets to 40.77 (allowing for longer trades and bigger draw downs).
Both these systems achieved Total Returns (TRI) of above 20 (2,000% percent gain) versus the best original BITS signal (BITS-MT) that achieved a TRI of 13 in the old research note! When taking this TRI and accounting for the amount of time each system was vested in the market to achieve it, we get a "power" score of 9.0 to 1 for the ACTIV system meaning its earning power is 9 times that of the JSE per unit of time in the market! The charts to the left of the stats tables above show how the returns of the two systems utterly out-pace a JSE buy+hold strategy, both achieving over 2,000% return versus the JSE's 415% return.
Below is a chart showing the vested history (periods it was long) of the LAZYBOY system. We have only seen something this good with HedgeTrader HT-1 and Turtle-S3, making LAZYBOY an excellent long term trading signal. As with those other two systems, when LAZYBOY is out the market you need to be concerned and could quite possibly even short the market quite successfully. The vested chart for ACTIV will appear similar except with many more trades. It is less successful with shorting though.
The 22 trades executed by LAZYBOY are tabulated below:
CDAYS is calander days, TDAYS is trading (work) days, GAIN is % profit on long trades, CAGR is compound annual growth rate of the profit, G/L is points up divided by points down (gain/loss ratio). We track cumulative stats for the long as well as short trades executed on the market. The short depicted in a row runs from the TO date when the long was closed to the FROM date in the next line when a new long is opened.
We note that fairly consistant, large returns are logged on the long trades, but the short trades, whilst profitable over 15 years, have far more mixed results. About 50% of the short trades are losses from intra-bull market pauses, but every now and again you latch onto the "big makuna" bear markets associated with economic recession or bubbles bursting and make an average 25% gain on your short. This happened 3 times in the last 15 years.
But the fact remains the long trades are far more predicatable and consistant than the short trades. You are probably better off just sticking your funds into money markets or call deposits during the shorts to get better risk-adjusted returns.
Note how the short-duration long trades, with 5-9% gains have very high compound annual growth rates (CAGR) - meaning you get bang for your buck even though you are in the market for a short space of time. In fact we would say the most impressive characteristic of this system is the CAGR column showing you the rate of annual returns per trade.
Note also how the losing trades are generally cut short and dont' last very long.
Even if you do not trade using this system, just as with HedgeTrader and Turtle-S3, you need to be wary when this system is out the market as it is a warning the market is vulnerble. Now a characteristic of the system during a bull market is to ride wonderfully long trades punctuated now and again by an exit from a particularly large (but healthy) correction. Then a few days, at most weeks thereafter the system goes long again as the bull market resumes. This may appear innefective or innefficient but the fact of the matter is this pulling out the market when breadth gets particulalrly weak is what SAVES YOU FROM THE CRIPPLING BEAR MARKETS.
It is a small price to pay for peace of mind you should never get caught up in a correction that can destroy up to 50% of your capital. When the system is long or has just gone long, it gives you comfort you can conduct all your stock market operations (be it investing, trading etc) with the wind at your back. The longer the system is OUT the market, the more ominious the signs of the footsteps of the approaching bear.
Two new trade charts appear in the JSE TIMERS sheet in your JBAR reports from 28 March 2011, right under the 4 original BITS timers (since ZBITS is part of the BITS family!) As at 28 March 2011 they appear as shown below (the charts show activity since this bull market commenced in March 2009).
The yellow line is the BITS signal we all know and love from the old BITS charts we still use. The horizontal dotted orange line is the lower threshold to "arm" or "prime" the system. When the BITS signal drops below this line then we have 12 trading days for it to rise to the horizontal green dotted line (the BUY trigger). If this is reached we BUY the ALSH.
As you can see the system has JUST achieved this on the above chart on the right and the Status heading has changed from SHORT to LONG. The system stays in the JSE for as long as the yellow signal line remains above the red dotted line at the bottom of the EXIT ZONE. Very similar to BIT-41 and BITS-45 except we use different triggers and have now incorporated the use of an arming signal of 44.41 as prerequisite to the BUY trigger being reached within x amount of days.
Let's look at the trade chart for the ACTIV strategy. It also has just fired a signal on the right of the below chart. Clearly this system is far more active than the lazy system having made 8 trades versus 2 in the last 2 years. This system is more geared to SATRIX CDF's or ALSI futures given the smaller draw downs, whereas the LAZY system is best for ungeared or low-geared trades:
With the ACTIV system, the exit threshold is identical to the arming signal, namely 44.41, so when the system EXITS the market it AUTOMATICALLY arms itself for the next signal. However with this system we use a longer 18-day rolling window to allow the swing from low to high thresholds to occur.
There is also a "90% ARMING ZONE" below the 40 threshold which shows "great troughs" in the making. We treat BUY signals originating from this zone as the JSE's "Zweig Breadth Thrusts (ZBT's)." There have only been 20 of them since 1997 and if one considers a 30 trading-day holding period, then BUY signals flagged from below 40 have a 90% confidence level. Those buy signals flagged from below the 44.41 (exit zone) but above the 40 level, have an 80% confidence level.
These two timers give us the frequency of trades lacking in the reliable BITS timers and serve as useful additions to system diversification. We recommend you trade the ZBITS signals whenever you can and ALSO add to your positions when you see the standard BITS signals (as they will effectively serve as PYRAMIDING "continuation" signals for the ZBITS signals.)
The best way to play the LAZYBOY signal is to buy plain old boring SATRIX-40 shares. You dont' have to gear, just buy the ETF's. Each trade has a target of at least 20% gains, but the main objective is not the 20% but the CAGR rate you will achieve the 20%. As you will be swithcing in and out the markets occasionally there will be tax considerations. we maintain tax consierations are a nice problem to have and you should never let them dictate your actions on the markets. Buying and then holding and hoping for long periods can mean you might avoid SARS but a bear market will then inevitably exact its 40-50% tax on you so you land up paying the piper anyway!
You can DOUBLE the performance of this trading system by playing it with SATRIX-RESI, due to its higher beta. The volatility is far higher but so are the gains. If you can stomach the intra-trade RESI volatility you will be handsomely rewarded.
If you have nerves of steel you can try your luck with gearing with LAZYBOY. But remember that intra-trade there are large drawdowns you need to cater for.
<--- GO BACK TO THE BITS TIMERS FAMILY MAIN PAGE