This week we discuss price patterns.
Q: What are price patterns?
Price patterns are recurring price behaviour that are observed visually (and can be modelled by computers). This key here is ‘recurring’. If a price pattern is not repeating, then it is just random noise.
The act of finding price patterns is called “pattern recognition”.
Q: How can price patterns be used in market speculation and investing?
When a pattern is repeating, it can be predictive and harnessed for profitable trading activities. To do this, the pattern must be (a) identified, (b) measured, and (c) traded upon profitably.
However, there are times when (a) and (b) are satisfied, but not (c). A pattern may exist, but it still can not be traded profitably because of, say, high transaction costs.
Q: What is the best place to start looking for price patterns?
I would say charts. A lot of price patterns are found visually. You can use line, bar or candlestick charts. The point-and-figure chart is also useful.
Next, you might want to vary the frequency of the charts. Some patterns could be more obvious if you switch to Weekly Bar charts, for example. Some are more applicable on Daily Candlestick. You can split pattern by frequency:
- Daily Pattern – Pattern 1, Pattern 2,… etc
- Weekly Pattern – Pattern 2, Pattern 2,… etc
Q: How do you determine if a price pattern has emerged?
To check if you have found a pattern, you need to ask:
- Have you seen this before? Where?
- What happened then? Describe it.
- What were the actions preceding the pattern? This could be important as you’re determining the factors leading to the pattern.
Once you answer some of these basic questions, you will know quickly if the pattern is indeed a pattern.
Q: Once you have identified a pattern, what is the next step?
Verify the pattern. Is the pattern random or regular? How big are the price returns pre- and post-pattern? Is the pattern long-, medium-, or short-term (duration)? Can you act fast enough to capture it? Is the pattern profitable once transaction, spreads, and other costs are factored in?
To answer these questions, you will need to be able to describe the idea, then backtest the idea. The action flow is something like this:
Observe -> Pattern Found -> Description of Pattern (Models) -> Backtest -> Results (Analysis) -> Trade
Q: What sort of price patterns are there?
There are many hundreds of price patterns used by traders, funds, or investors. For obvious reasons, the more profitable ones are probably kept hidden from the public. In general,
- Calendar Based – eg, Month of the Year or January Effect
- Sequence Based – eg, Runs Analysis
- Trend Based – e.g., Head-and-Shoulders
- Return Based – e.g., Daily/Weekly Returns, 52-Week Returns
- Short-Term Pattern – e.g., Price Gaps, Reversal Day
- Plot Based – e.g., Point-And-Figure
- Cycle Based – e.g., early/mature bull markets
etc. One of the simplest calendar pattern I can think of is the Quarterly Effect in equity indices. For example, the ‘Santa Claus’ rally is evident in the DJ Industrial Index, whereby the last quarter of the year generates, on average, the highest return compared to other quarters.
Q: How do we choose which patterns to incorporate into our trading plan?
One important consideration is consistency. Is the pattern reliable?
Reliability here refers to behaviour of returns. If a pattern generates lots of small profitable trades and then proceed to lose big in one trade, then one has to be really careful about using it. If a pattern has lots of small unprofitable trades and win big in a few signals, it might still be viable.
The next issue is its scalability. Is the pattern applicable to other markets, sectors, or asset classes? Can it be scaled up with more capital? This issue has to be researched and tested rigorously – simply because it is dependent on the actual pattern, depth of markets, and capital deployed.
The last issue is shelf-life. How long has the pattern been observed? Is there any studies on it? Is it widely known? Generally, the more capital chasing the pattern, the shorter its shelf-life will be. This is due to a principle of capitalism: Returns fall as capital employed rises.
Q: What about combining patterns?
As a rule of thumb, the simpler the pattern the better. A complex pattern will probably contain too many variables and this may impact its effectiveness. For example, is a complex head-and-shoulders really better than a normal head-and-shoulders? I doubt so.
In the realm of market analysis, price patterns are an important part of it. But not all patterns are equal. Some are profitable; some are money-losing. Some work for a time, then fade away. Some work in stocks, but not in others. So a lot of research is required to extract good trading/investing ideas.
Lastly, if you found one good pattern, try to develop little trades around this pattern to maximise its usefulness. In my experience, it is better to focus on a few good patterns rather than a lot of marginal ones. Diversify, but only to a point.
Ten things to understand before you start trading…
1. How to use support and resistance levels in trading
2. Using Moving Averages Effectively – Part 1
2.5 Using Moving Averages Effectively – Part 2
3. Momentum indicators and trends change
4. Understanding Price Breakouts and its Significance
5. Q&A On Price Patterns With Jackson Wong PhD
6. The Importance of Group Analysis
7. Three Chart Characteristics That Precede A Trend Change
8. Thoughts on trading the market via Breadth
9. Six Market Trends To Look For Outside Individual Price Action
10. The Key To Long-Term Investment Success – Know Yourself…
Compare Vetted Investing, Trading & Currency Accounts
|Investing Accounts||Trading Platforms||Currency Transfers|
Jackson has over 15 years experience as a financial analyst. Previously a director of Stockcube Research as head of Investors Intelligence providing market timing advice and research to some of the world largest institutions and hedge funds.
Expertise: Global macroeconomic investment strategy, statistical backtesting, asset allocation, and cross-asset research.
Jackson has a PhD in Finance from Durham University.