RELATED TOPICS --> | Trading Tools Guide | Trading individual shares using ShareTrader |

Traditional wisdom dictated that “time in the market” and not “timing the market” was your route to stock market success. “Buy and hold” investors would be better off than those that attempted to jump in and out the markets trying to “buy low and sell high”.

But this traditional wisdom is coming under fire, especially in light of the latest stock market crash having wiped 45% of investments value. Surely there must be a way to get into the market when it is “less risky” and get out into safe fixed-deposits or money market accounts when it is “more risky”? Surely such an investment scheme would produce superior returns to a buy and hold strategy? At PowerStocks Research, we set about a 24 month journey to discover the answers to the above questions, and with very significant and surprising results indeed.

In Michael Covels' international smash-hit "Trend Following" he showed how hedge funds and famous private investors have consistently made millions and billions of dollars using 100% systematic trading systems that follow trends and use quantitative models. These traders consistently beat the pants of mutual funds and other actively managed funds that use "fundamental research".

The table below shows the 32 year returns on an initial R1.00 investment from various long-term timing strategies we have developed. Apart from the amazing returns from only 1 trade per 24 months on average, these strategies only exposed you to stock market risk between 39.5% and 58.5% of the time, the balance of the time your funds were safely tucked away in fixed deposits earning prime less 4%. The SUPERModel strategy, delivered over 10 times the buy-and-hold performance by exposing you to half the risk!

This is not something we concocted up, it is merely an implementation of an investment timing model tailored for the JSE and SA economy, based on similar work done by investing greats such as Dr. Martin Zweig and Gerald Appel for the U.S markets. These timing concepts are sound and they work.

With the advent of cost effective Exchange Traded Funds (ETF’s) such as SATRIX, and the ease of use of discount on-line brokerage accounts for the man in the street, coupled with a rational system for entering and exiting the markets, the execution of such a strategy by the private investor would be a rather simple affair.

We believe that it is not in the interests of the fund management and brokerage industry for private investors to time their own investments - after all, if this were possible then who would need them and pay their expensive management fees and commissions? And it would be rather embarrassing for the industry if private investors could significantly outperform the large funds now wouldn't it? Actually 80% of fund managers worldwide struggle to beat the market indices, and because they run such large books, they are actually precluded from executing market timing! Think about it - they can't go and dump their entire book overnight when they see a sell signal - they are forced to short the market with derivatives to hedge against their losses! But the man in the street is hardly likey to have a large enough line to impact prices when he switches from equity to cash - so this really should have you thinking about doing some of your own "fund management"!

To summarize, anyone who says you cannot time the markets is talking bunk. Whilst you might not be able to perfect such timing (but some of our systems come very close) you can certainly deploy it effectively to significantly reduce your risk exposure and enhance your returns over a period of various time horizons. As these system stack the odds in your favor, the longer they are used the more success you are likely to achieve.

We have researched, back-tested and developed highly successful long-term (multi-year) investment timing signals with our SuperModel and Long-Bond Yield Curve Investing Timing Systems. Then we have medium-term (multi-month) timing signals with our JSE SwissClock, Breadth Impulse Trading System (BITS) and others. Finally, we have  SwingTrader-III and Sincerity-Trader for short-term (multi-week and sometimes multi-day) timing signals.

All three category of timing signals (long, medium and short) offer significant risk reduction and out-performance of the ALSH index. The purpose of this note is not to delve into the mechanics of these timing signals but to offer an example of a diversified deployment technique as used by us for our own brokerage accounts. To dive into the technical and performance details of our long, medium and short term JSE timing systems, go to the main MARKET TIMING page on our website or view our Trading Tools Guide. The rest of this page focuses on an introduction and technique of deployment.

Our approach has been to build and trade/invest with a host of long, medium and short-term trading and timing systems. Each system is unique in its approach and quantitative theory which offers us Strategy Diversification. Specific systems have been built to time uncorrelated markets such as GOLD, PLATINUM, FINANCIALS, LISTED PROPERTY etc. which gives us Market Diversification. Differing systems are suited to operate over specific time horizons, which offers us Time Diversification. This diversification is important and it is well documented that most successful quantitative hedge funds operate with high levels of this type of diversification.

The catalog of our various proprietary systems appears below. Although the systems are proprietary you will note from their technical descriptions that they are founded on well-known quantitative theory and optimized for operation of the JSE. Although it seems that there are a lot of systems, it is important to note that our approach has been to develop a range of high-confidence, non-overlapping (uncorrelated) systems with high gain/loss ratios. With this business, the downside to high-confidence is low frequency of signals. It is really no use having a system that is 90% accurate but only presents an opportunity whenever Halley's Comet decides to swing around Earth! But because we have a wide range of non-overlapping high confidence systems, spanning each time horizon, we land up with enough signals to keep subscribers busy - and by this we mean an average of 2 signals per month.

All the below systems have very predefined ENTRY signals. Except for TroughFinder-III, all the systems also have very predefined stop-loss and EXIT rules. TroughFinder-III is our oldest system and provides fantastic risk-adjusted early entries, but we have left the exits for the "traders gut-feel". However, you can revert to SwingTrader-III or Sincerity-Trader exit rules once opening a TroughFinder-III trade (provided they have opened trades shortly after the TroughFinder entry.) You can also use the SWENLIN reversal signal to time an exit for TroughFinder-III.

1.6 BULLISH PERCENT BPI (very rare, very short trades)

All the below also have very specific ENTRY, STOP-LOSS and EXIT rules. As the trade horizons are longer, you will need to stomach higher volatility and draw-downs during the life of the trade, so you may want to use less leveraging (margin or gearing) or simply play vanilla un-geared ETF's with these systems.

2.1 HEDGE-TRADER HT-1 A (aggressive) & B (conservative)

These systems are vested 70-75% of the time in the JSE, but are not investment systems as they are still quite actively in and out the market. Good for shorting when out the market.


You can use these to trade with but most people use them as "backdrops" against which they validate their various JSE operations, as they tell you when we are in bull or bear markets and give guidance as to relative bullishness (expected returns.) Whenever these systems signal BUY's it means it is time to start playing in the markets again and move from cash to equities, regardless of if you view yourself as a trader or a long term investor. When this happens, we invest R1M into various Exchange Traded Funds (ETF's) and forget about them for 2-3 years until we see the SELL signal. They are very low frequency. You are advised to only trade in individual shares when these systems are very bullish (ie SuperModel signal line is at least two or the LBYC yield curve spread is 2 or more.) and for the other times stick to ETF's.


These are not trading systems per-se but identify very favorable JSE entry points for long-term investors (holding periods of 12-36 months) or for Rand-Cost-Averaging investors (subscribers or fund managers that put regular payments into the JSE on a monthly or other basis.) They are also used by investors which have identified a share they like and are waiting for the best time to purchase it (in terms of reducing their risk of it going down after they purchased it.)


Play up to 5 trading systems concurrently, say 2 from short term trades (whichever 2 come first) and 3 from medium-term trades. Play each trade according to the rules of the system you used to open it and don't mix and match signals. Keep high leverage to the shorter term trades unless you have the experience, skill and psychology to handle large draw-downs and volatility. Don't just trade the JSE, trade its various sectors as well with our SECTOR ETF TIMERS as this diversifies your exposure to markets and could bring more opportunities to the fore. For short and medium term trades don't expose more than 2-5% of your capital to the trade or if leveraged, the trade margin. The only reliable shorting timing mechanism we have is when either HT-1A or Turtle-Trader-3 are out the market. For the rest of your market "idle-time" stick to cash or short term money market accounts or rotate into accumulation of inflation-protected bonds or gold (JSE peaks are the best times to be buying these assets.) We cover more detailed techniques lower down on this page.

The use of our timing signals is effectively a risk-reduction exercise. Our systems ensure you only deploy funds into the JSE when the odds are stacked in your favor. Contrary to popular belief, the JSE is only "safe" about 50-55% of the time! Now in our "Is the JSE Safe?" research paper we showed that the long-term upward bias of the JSE coupled with "time in the market" is indeed an effective way to neutralize JSE risk. It is a fact that the longer you hold your investments, the better they will perform. But what if we could stay out the JSE that 45%-50% of the time it is "unsafe"? Surely this will reduce risk and enhance returns even further?

Let us assume you have read the research behind all our timing models and have now seen the light and want to deploy timing as your JSE strategy. How do you do this? The techniques you deploy will ultimately depend on your investment style (active, passive, both) , risk appetite and resources (funds available to invest) but below we offer a technique we successfully use ourselves to get you thinking.

1. Multiple Strategies and Brokerage Accounts
We recommend you execute both long-term and medium-term timing strategies. If you have the time and appetite we also recommend you execute short term timing strategies. We recommend you open separate brokerage sub-accounts with your on-line broker for this purpose. Standard Bank (who we use) allow you to open multiple sub-accounts within your main brokerage account at no extra cost. We have three accounts, namely SUPERMODEL, MEDIUM_TERM and SWINGTRADER

The separation of these accounts is useful for tax purposes since trading frequency on each account will vary significantly. On SUPERMODEL, you will make, on average 1 trade per 2-3 years for each underlying share in the account. With MEDIUM_TERM you will be making one trade per 4-6 months on average. With SWINGTRADER you will be making one trade per 3-6 weeks. Keeping the accounts separate assists you with tax calculations and redemption's and also will assist with the measurement of effectiveness of your trading with respect to profits, brokerage fees etc.

2. Funds Allocation
The amount of funds you allocate for deployment to each account will vary. At PowerStocks we deploy 45% of our funds pool to SUPERMODEL, 25% of our funds to MEDIUM_TERM and 15% of our funds to SWING_TRADER. The balance of 15% we keep for "special opportunities" - little gifts from the stock market gods that arrive in HeadsUP! Alerts about 4-8 times per year.

Note this is merely funds allocation to a strategy and does not mean we blow the whole allocation on a signal from day one!

3. Phasing in of funds with SUPERModel
With SUPERMODEL, we phase our funds pool into (and out of)  the JSE gradually depending on what SUPERModel Signal is showing. Let us assume we are dormant, with no SUPERModel BUY signal showing and our funds stuck in a money market account. This will occur during a bear market or during a period when JSE returns have been rather lackluster, such as in a sideways market. The SUPERModel signal is usually reading "-2" at this stage (a signal that tells us its dangerous or not worth our while to be in the JSE.)

One day you look in HeadsUP! or read your Weekly JSE Pulse (WJP) and you see that SUPERModel has awoken from hibernation and raised its signal from -2 (Bearish Zone) to -1 (Neutral zone) as shown below the last time it came out of hibernation on 12 December 2008:

This means one of the three SUPERModel timing sub-models has started showing a signal. Each of these sub-models in their own right are excellent timing strategies (as shown in that table right at the top of the page), so we respond by phasing in 15% of our SUPERMODEL funds pool into the market. When the signal gets raised from -1, to 0 we phase in an additional 20%  and when it goes from 0 to +1 we place a further 25% and finally if it goes from +1 to +2 we place another 25% of funds into the market.  At this stage we have 85% of our SUPERMODEL funds pool in the JSE.

You can see this consecutive phasing in action on 18th May 2009 when the SUPERModel signal made its second move,  a large and sudden jump from -1 (Neutral zone) to +2 (very Bullish) meaning on that date we placed a further 20%+25%+25%=70% of our funds into the market, to be 15%+70%= 85% vested. As at 19 October 2008 the JSE has grown 18.5% (52% CAGR) since then.

From time to time, the signal will make a rare move from +2 to +3 (Turbo Zone) and this is when we throw in our final 15% to be 100% vested in the market. The signal only stays on +3 for about 2-6 months, so when it drops back to +2 we sell 15% of our total SUPERMODEL holdings again. Because the jumps from +2 to +3 and back are more frequent than the periods when the signal remains on +1 or +2 (sometimes for 2 years!) we recommend you use the MEDIUM-TERM account to effect these trades to separate out your "trading" activities from your "investing" activities.

Over time, the SUPERModel signal will steadily start deteriorating as the bull market approaches its sunset (takes from 1-3 years) and for each 1 point the signal drops we reverse the phasing of our holdings. You can see from the above chart that in the run-up to the great 2008 crash, that by June 2006 the signal was at +1 and we had already disposed of 25% of our holdings into the money market accounts. On August 2006 the SUPERModel signal dropped a massive 2 points from +1 to -1 and we moved a further 45% of our holdings into the money markets, which by this time were not earning bad interest rates at all since we were way into the interest rate tightening cycle.

At this stage we had our last 15% vested in the JSE until September 2007 when the signal collapsed to -1 and that was it - we were getting the hell out of our long term investments. Although we only had 15% of our funds committed during this "last gasp" of the bull market, you can see the JSE rose from 21,500 to 30,000 during this period, a meaty 39.5% growth in 12 months! We got out just in time too -as five weeks later a massive correction (no, call it what it was, a crash) ensued.

From September 2007 through to December 2008 (14 months), all the way through the great 2008 crash, our funds were socked away in fixed deposit and money market accounts earning fat interest. In fact the Reserve bank was hiking interest rates all the way through 2008 and we were just loving it. That is the beauty of the SUPERModel - get into the JSE when it is safe and interest rates are low, and get out when it is unsafe and interest rates are high! You just can't lose with this strategy, all you need is patience.

This phasing-out of money as the bull market progresses and the JSE becomes more risky is nothing new, it is sound risk management. Having 15% of our funds committed during the "blow off" phase of the market was also prudent, we didn't expose our hard earned capital to risk and caught onto a healthy 39% growth. If you would like to learn further about the various stages of the stock market cycle, and how to adjust your exposure during each stage (as we did with SUPERModel) then read our educational note "The importance of understanding Stock market Cycles"

Another interesting thing - many of our subscriber play SUPERModel on the 0 "Buy/Sell" line and will not be vested in the JSE unless the SUPERModel signal is zero or more. This is a more conservative strategy, and the one we have back-tested to achieve in excess of 117,000% return over the last 31 years versus the ALSH buy-and hold strategies' 7,878% return. In this case they would modify the percentage funds allocations accordingly to cater for the 4 steps from 0 to +3 using say 25% allocation each time the signal goes up by 1 as an example.

A final note - sometimes it pays you to wait for a TroughFinder signal before phasing in your additional funds. This is called "loading up on the dips" and can have a huge leverage effect on your profits since you are always adding to your portfolio at the cheapest price of the time. It also lowers your risk since when TroughFinder fires a signal you are at a trough and the only way forward is up. There is nothing worse than investing in a share only to see it drop shortly afterward, and the JSE has an uncanny knack of doing this to a person! A rising tide floats all boats so buying your shares on TroughFinder signals minimizes the chances the share you bought will drop (unless you bought a dog of course, but more on stopping that later!)

4. Playing the Medium-Term Strategies in tandem
Note that we still played the JSE after liquidating our long-term SUPERModel investments in September 2007. In fact the volatility that commenced with an uncertain and wobbly market after October 2007 offered great medium and short term trading opportunities with our other timing signals. When TroughFinder starts showing an A-signal greater than 5 though we stay out the market completely as we are in a "crash class event" and all the rules go out the window.

The medium-term strategies we play in tandem with the SUPERModel. We split the funds allocated to this strategy 40% to SwissClock, 35% to BPI and 25% to BITS. When the signals show, we vest 100% of allocated proportion of funds (we don't phase in as with the SUPERModel). We play multiple medium-term strategies since they start and stop at different times, and sometimes some are more successful than others. Playing multiple strategies just diversifies your approach and even if one strategy leads to mediocre returns it will probably be made up by another. All three medium term strategies use very different unrelated mechanics to generate signals which provides for proper diversification. Bear in mind though that for a great period of time these strategies are in play, they will have overlapping signals.

5. Playing the Short-Term Strategies in tandem
With the short-term trading signals we play SwingTrader provided no major crash is on the go. A side-note on short-term trading - this is obviously more risky and for that reason very different money management and stop-loss rules apply and as an additional precaution we only deploy up to a maximum of 30% of our capital allocated to short term trading on any one trade. If through a rare misfortune one of our leveraged trades gets wiped out we still have funds left to play with to make it up.

Do not be put off by "leverage", "futures", warrants and other complex jargon and horror stories you may have heard about more sophisticated investing instruments. Yes, trading in these is speculation (gambling) but think of this : timing these trades with our SwingTrader and TroughFinder signal systems significantly reduces your risk and give you the edge. Think of a card counter that (usually) makes a lot of money at the casino (if he hasn't been banned from it yet.) He is still gambling when he manages to sneak into the casino in his disguise, but by card counting he virtually overturns the odds the house has stacked against him. Is this still gambling? Probably not! If you have the time, attend your on-line brokerage houses' free courses and education on warrants and SSF's and "paper trade" a few of our SwingTrader and TroughFinder signals to build up your confidence and you will soon be learning to count cards - the bonus is you won't get thrown out the JSE for consistently winning!

If you want to see a detailed trading history of how we earned 15,000% compound growth per annum in the last 5 months with SwingTrader, by putting very few of our funds at risk, go to the SwingTrader page.

6. Underlying investment instruments
So you have great entry/exit timing signals. The next question that bears answering is "What shares do we buy when signals flash BUY?"

If we have just exited from a major crash or correction exceeding 20%, we use our Crash-peak Recovery Strategy taken from our Anatomy of a Crash research, to select individual shares. You only have one crack at this and this type of opportunity does not come along often.

For the short term trading signals this is easy - we only use leveraged products and this is SATRIX40 and SATRIX RESI Single Stock Futures (SSF's) and Standard Bank TOP40 Knock-out Warrants (which are geared anywhere from 5 to 10 times). Any instrument you feel will follow the JSE up and down in very close correlation will do, but as the trades are very short, you will need to deploy leverage to ensure brokerage costs don't chew up your profits. We like knock-out barrier warrants as they are highly geared, have low brokerage and the most you can lose is the initial investment.

For Medium Term trading signals we play standard vanilla un-geared SATRIX40 and SATRIX RESI ETFs. There is some volatility over the trade period so if you have the stomach for being underwater up to 10% on occasion, you may take a punt on a ALSI or TOP40 futures contract or an ETF SSF.

For long-term investing via SUPERModel, we split our investments 40% to ETF's (such as SATRIX40 and SATRIX RESI) and 60% to individual share purchases through various of our mechanical share picking strategies and screens. If you are a new investor and/or with limited funds we suggest strongly that you start by playing SUPERModel and/or SwissClock with an ETF of your choice to mirror the market and to ensure brokerage fees don't eat into your returns.  Correct selection of ETFs with SUPERModel can significantly enhance returns, see our paper "Market Timing and Sector Rotation"

When SUPERModel deteriorates to a  signal less than 1 we cease purchases of individual shares since the market is weakening and you only should be purchasing individual shares to buy into dividend streams (at the beginning of the bull market) and this would be pointless at this stage. Additionally if the signal should flag a "get out the market" you don't want to be hassling with offloading 10 different shares under pressure or get caught with your pants down, especially with some illiquid issues. At least if you are just sitting with a liquid ETF, its one single trade to dispose of your holdings.

Remember, if you invest less than R10,000 per share then chances are you will be paying too much for brokerage costs (as a percentage of your trade) since these are normally set at minimums of R70 to R100 on the way in AND on the way out. So if you have limited funds, stick to ETF's with the timing models.

To ensure we are not pumping up the ETF's too much, you need to consider that there is no doubt that portfolios of carefully and properly selected shares (such as those from various of our strategies) will SIGNIFICANTLY outperform any vanilla ETF over a trough-to-peak bull market period, especially when dividends are taken into account and especially if incepted straight on the back of TroughFinder signals. Just go to our STRATEGIES SCOREBOARD page to see this in action.

7. Avoiding confusion with seemingly conflicting signals
Our long, medium and short term timing signals all correspond to different investing horizons, risk profiles and return characteristics. For this reason one can be showing a BUY while another is showing a SELL and you can be forgiven for being confused:

The SUPERModel is a long-term investing signal (18 months to 3 years) designed to get you into secular bull markets as close as possible to their beginning and exit before the bear markets set in. This shows a measurable signal from -2 (bearish) to +3 (turbo charged). It assesses monetary conditions(interest rates), the cyclical turning points of the economy, seasonality and JSE rate of change (momentum) to determine if we are in a secular bull market and have high probabilities of favorable returns. It is not concerned with consolidations and pull-backs of up to 10-15%, that is the prerogative of the medium term timers. Whilst the latest posting to Dave’s Diary warns of a secondary correction imminent due to various divergences and efficiency readings in the JSE, the SUPERModel does not get concerned with such detail. The only time we over-ride the SUPERModel signal is when there is a CRASH ALERT on PeakFinder.

It is quite feasible that SUPERModel is showing a BULLISH signal but SwissClock is BEARISH. There are 2-6 SwissClock cycles (trades) per SUPERModel cycle on average so this should happen at least once during a SUPERModel signal. When SwissClock is EXITING you probably shouldn't be disturbing your long term investments for tax purposes and besides you bought into a dividend yield right? If you play our system then chances are you would have only purchased your long term investments in a TROUGH so should have a nice buffer to survive the inevitable  SwissClock down-cycle.

One tip though : if SwissClock EXITS the JSE and you're sitting with shares showing less than 5% appreciation you MIGHT want to consider pruning these from your portfolio. Remember the curve of increasing pain discussed in Dave's Diary showing what recovery is needed to make up for a respective loss. If the correction is 15% then you land up being 20% down and that needs 35% gains to recoup your losses.

Just as there are multiple medium-term cycles with each long-term cycle, there are multiple short term cycles per medium term cycle. So when viewing a BUY or SELL or BULLISH or BEARISH signal, always remember to consider where its coming from (SUPERModel, SwissClock or SwingTrader) and what time-span were talking about.

8. Overlays : TroughFinder and PeakFinder
These two highly reliable (but infrequent) signal systems are not market-timing strategies per se, but are purpose-built to identify stock market troughs (valleys or bottoms) and reversals as well as large stock market tops that are forming (peaks). When these systems flash signals on HeadsUP! we ALWAYS take notice.

We have already discussed how TroughFinder is an excellent tool when combined with SUPERModel to "load up our portfolio on the dips". But we almost always enter a Warrant or SSF trade when we see a TroughFinder signal of strength 3 or more, even if no SwingTrader or SwissClock BUY cycles are in progress.

That is why it is important to keep 15% of your funds lying about. We call it the "Opportunity Funds" to allow us to have cash to deploy when high-probability events occur on the JSE that we just know are going to make a profit. It would be a shame when these events come along and all our cash is tied up in stocks.

When PeakFinder gives off strong signals, we don't care what investment models we are running we always sit up and take notice, especially if its during periods where the long or medium term signals are showing some vulnerability. We assess the situation for our long, medium and short term trades and take action. We most certainly are likely to liquidate any short term positions we have open and tighten stops and observe closely what is happening with the medium term trades and the long term trades.

RELATED TOPICS --> | SuperModel | SwissClock | BPI | SwingTrader | TroughFinder |
Make a Free Website with Yola.