An Approach to Trading Shares
Issued: 26 June 2005
© Copyright 2003, 2005 by Udo Stegen, Rettmer Trading Trust
A short PowerPoint presentation is also available on this subject. Click here to download and view.
1. Basic Prerequisites
Throughout this discussion, I assume the reader has a fair understanding of -
· The 4-stage concept of a share's life cycle: This is best explained in Warren Magness' video “The Share Market Life Cycle”.
· The Market Analyser and its key indicators: This program is developed and distributed by MDSnews.com.
· Fibonacci Numbers: The series where each number is the sum of its two predecessors. 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, etc.
Long and Short trading: This is something you need to study for
Unless you can afford to lose unlimited funds, paper-trade until you know the ropes.
None of the above subjects will be explained in any detail.
Several years ago, I designed my first set of templates that was meant to show me at a glance where a stock was sitting in its life cycle. It was based on the famous Four Stages, made famous by Warren’s video.
For the tools then available (MA2000 was just being built) it was a simple approach - some people would even say "simplistic". But it worked - more often than not. It was those "when not" that hurt, and I discussed the problem with various friends – e.g. A,B,C,D: Arthur, Bart, Cool Hippo, David. As a result, three main groups of indicators emerged:
1. looking at price and volume
2. looking at momentum, ie strength of price change
3. looking at the stock's history.
At one stage, there were three sets of threes - hence the term "Trinity" was coined. Although much has changed with time, the name stuck. Meanwhile, it's become well enough known as a concept, so we seem to be stuck with it. If others have their Holy Grail, why shouldn't we consult the Trinity!
For any trading, the basic premise is to "Buy low, Sell high." Applied to objects (shares) with oscillating price charts, we need to find the points where the price chart turns up (to Buy) or down (to Sell). If we bought "in between", we would simply forego too much of the profit margin. Likewise, there is no point in selling too late when our profit margin has already been half eroded.
Depending on our Trading Horizon, we can use different periods. Short periods have the advantage that predictions about a stock's behaviour are more likely to be accurate for the next few days. However, the further out our plan, the less certain the prediction becomes. On the other hand, the price usually only moves so much in a day, so the longer a trend can run, the better for our profit. We are therefore looking for a compromise that balances the two opposing influences.
For Trinity, we use Daily charts and a main period of 34 days. Its other key numbers are 3, 7, and 13.
Note: 7 is the "odd number out" inasmuch that it is not a Fibonacci number. For some stocks, 5 or 8 seem to work just as well or better. This is something we have to test and verify: Find out for an equity, which set of parameters works best. I describe such a set as a stock’s “DNA”.
3. Trinity’s Three Components
After a stock has ended its downtrend, it usually enters a consolidation phase, ie it moves sideways in a narrow channel. At some time, it leaves this channel. Depending on prevailing market forces, it can break down through the lower boundary, thereby continuing its downtrend, or it breaks to the upside. It is this breakout that we're looking for. The tool we're using is the Market Analyser: in particular, a set of Indicators that have the potential to show likely entries. Here are the key components:
1. Identify the potential end of a down trend. We use A(34), the 34-day Accumulation Indicator with a 3% price drop. If I scan the entire market, I ask for high volume: at least a factor of 1.5. For the Top 50-100 stocks, 50% volume spikes are much rarer, and I am satisfied with a factor of 1.
2. Identify the significant Turning Point. Trinity looks for a TP34. However, waiting for a TP34 poses one big problem. By the time it has locked in as a confirmed TP34, it is useless for a trade. By definition, we may have to wait 34 days to know that a TP is actually the lowest price within a 34-day period. Therefore, we have to look for something that usually "goes with" a TP34.
3. Verify the Turning Point. Experience has shown that such a co-indicator exists. The classic Trinity method uses a combination of Stochastic and MACD. The former is triggered when the fast line comes out of Oversold, ie moves above 20%. MACD confirms the TP when it crosses zero. (For some stocks, it is sufficient to see MACD cross the signal line. Note under “DNA”.)
Picture 1: Steps 1 to 3: Example BHP, May 2004
4. Variations on a Theme
Sometimes, the reversal after an A34 is only short-lived. The price hits resistance and fails, thus establishing a consolidation zone. According to Warren Magness, the first is likely to hit Primary Resistance, from where it can bounce back in a move dubbed "First Test, First Failure". The level where this happens determines the upper level of the consolidation zone (aka “Stage 2”).
Quite often, we find that the Lows in MACD and the corresponding Lows of the price chart, trend in opposite direction. This is known as “MACD Divergence”, and I find it a reliable hint that the next Low may indeed be the TP we are looking for.
It is up to each individual and their risk tolerance, whether they are prepared to enter early, or prefer to wait patiently until the level of Primary Resistance is broken in the next “test” that does not fail. A quick scroll back into history can provide an indication of the likely profit margin between an early entry and the FTFF level. Such scrolling back is always recommended to become familiar with a stock and its "DNA".
Picture 2: FTFF and Bullish Divergence: Example WPL, May 2005
With the MA Pro, it is possible to write entry scans by either combining the 3 steps into a Wizard creation, or using Pascal to program a dedicated script. Best results are achieved by fine-tuning the various events and their sequence in time; therefore, I opted for the latter. I have also written “inverted” scripts that suggest entries for Short trades.
Apart from scanning, entry formulae can also be added to charts, and in the above pictures I use that feature (see “Trinity signal”)
Lastly, the scripts can be used to build Trading Systems in MA Pro, which can be back-tested for various watchlists of stocks that share similar DNA. Trading with the overall market, Long entries have consistently returned profitable results. Win:Loss ratios of 5:1 over 5 years have been demonstrated.
Picture 3: Profit Chart: Example DNA-keyed Long trades July 1999 to June 2004
For a quick scan, I am using the Daily template for all periods. For in-depth studies, I have adapted the key parameters to suit Weekly, Daily, Hourly charts. Naturally, the shorter periods will work only with highly liquid stocks. A larger ATR is also required, or the Intraday moves are too small to cover risk, spread, and brokerage.
These are my setups:
· A34 and D55. When I trade large-cap stocks, I use volume 1, price drop or rise 3%
· C 6, used only for clarification, not for scanning.
· EMA 3, 7, 13, 55, 233 - the longer ones only to visualise trends and possible support/ resistance zones
· Volume with SMA34
· Stochastic 13, 2, 3 with Weighted average for %D, and an EMA34 as average trend indication
· MACD 3, 13, 7, also with an EMA34
I establish key levels by using Fibonacci Retracement with Logarithmic (Natural) divisions. However, if there are obvious S/R lines, maybe even converging with EMA55 or EMA233, those will always take priority.
Disclaimer: Analyses over historic data need not be repeatable under future conditions. I do not claim any scientific rationale for the above, nor represent that any material or programs be free of errors. An automated trading system is never better than the person running it is disciplined.
The only person responsible for your actions and decisions is YOU.
© Copyright 2003, 2005 by Udo Stegen, Rettmer Trading Trust