The Science of Technical Analysis

Relax. Nothing is especially scientific or particularly complicated about technical analysis. Many in the market use the term science to describe the mechanics of various technical tools, but in my opinion technical analysis is far more art than science.

Each tool in technical analysis has a number of concrete elements that we need to outline before you can start interpreting what they mean. Unless you’re developing your own systematic trading model, you don’t need to get too caught up in the math or the calculations behind various indicators. Far more important is understanding what the indicators are measuring and what their signals mean and don’t mean.

Momentum oscillators and studies

Momentum refers to the speed at which prices are moving, either up or down. Momentum is an important technical measurement of the strength of the buying or selling interest behind a movement in prices. The higher the momentum in a down move, for example, the greater the selling interest is thought to be. The slower the momentum, the weaker the selling interest.

Currency traders use momentum indicators to gauge whether a price movement will be sustained, potentially developing into a trend, or whether a directional move has run its course and is now more likely to reverse direction. If momentum is positive and rising, it means prices are advancing, suggesting that active buying is taking place. If momentum begins to slow, it means prices are advancing more slowly, suggesting that buying interest is beginning to weaken. If buying interest is drying up, selling interest may increase.

Momentum takes on added significance in currencies because no viable way of assessing trading volume on a real-time basis exists. In equity and futures markets, volume data is an important indicator of the significance of a price move. For example, a sharp price movement on high volume is considered legitimate and likely to be sustained while a similarly sharp move on low volume is discounted and viewed as more likely to reverse.

Momentum indicators fall into a group of technical studies known as oscillators, because the mathematical representations of momentum are plotted on a scale that sees momentum rise and fall, or oscillate, depending on the relative speed of the price movements. A variety of different momentum oscillators exist, each calculated by various formulas, but they’re all based on the relationship of the current price to preceding prices over a defined period of time.

Momentum oscillators are typically displayed in a small window at the bottom of charting systems, with the price chart displayed above, so you can readily compare the price action with its underlying momentum.

Overbought and oversold

Momentum oscillators have extreme levels at the upper and lower ends of the oscillator’s scale, where the upper level is referred to as overbought and the lower level is referred to oversold. No hard definitions of overbought and oversold exist, because they’re relative terms describing how fast prices have changed relative to prior price changes. The best way to think of overbought and oversold is that prices have gone up or down too fast relative to prior periods.

Many momentum indicators suggest trading rules based on the indicator reaching overbought or oversold levels. For example, if a momentum study enters overbought or oversold territory, and subsequently turns down or up and moves out of the overbought or oversold zone, it may be considered a sell or buy signal.

Just because a momentum indicator has reached an overbought or oversold level does not mean that prices have to reverse direction. After all, the essence of a trend is a sustained directional price movement, which could see momentum remain in overbought or oversold territory for a long period of time as prices continue to advance or decline in the trend. Momentum is only an indicator. The key is to wait for confirmation from prices that the prior direction or trend has, in fact, changed.

Divergences between price and momentum

Another useful way to interpret momentum indicators is by comparing them to corresponding price changes. In most cases, momentum studies and price changes should move in the same direction. If prices are rising, for example, you would expect to see momentum rising as well. By the same token, if momentum begins to stall and eventually turn down, you would expect to see prices turn lower, too. But relatively frequently, especially in shorter, intraday timeframes (15 minutes, 1 hour, or 4 hours), prices diverge from momentum (meaning, prices may continue to rise even though momentum has started to move lower).

When prices move in the opposite direction of momentum, it’s called a divergence. Divergences are relatively easy to spot - new price highs are not matched by new highs in the momentum indicator, or new price lows are not matched by new lows in the momentum study. When a new price high or low is made, and momentum fails to make a similar new high or low, the price action is not confirmed by the momentum, suggesting that the price move is false and will not be sustained. The expectation, then, is that the price will change direction and eventually follow the momentum.

When prices make new highs, and momentum is falling or not making new highs, it’s called a bearish divergence (meaning, prices are expected to shift lower - move bearishly - in line with the underlying momentum). When prices are making new lows, but momentum is rising or not making new lows, it’s called a bullish divergence (meaning, prices are expected to turn higher - bullish - in line with momentum).

Divergences are great alerts that something may be out of kilter between prices and the underlying strength or momentum of the price move. Whenever you spot a divergence between price and momentum, you should start looking more closely at what's happening to prices. Are stop-loss levels being run in thin liquidity conditions? Or has some important news just come out that has sent prices moving sharply, and momentum will eventually catch up?

In a trending environment, prices may continue to move in the direction of the trend (that’s what a trend is), but at a slower pace, causing momentum to diverge. To know for certain, you need to wait for confirmation from prices before you enter a trade based on a divergence.

Using momentum in ranges and trends

Momentum indicators work best in range environments, where price movements are relatively constrained. As buying drives prices toward the upper end of a range, for example, selling interest comes in, slowing the price advance and turning momentum lower. As the buyers turn around, the selling interest increases and momentum begins to accelerate lower, confirming the change in direction. At the bottom of the range, the same thing happens, but in the opposite direction.

Momentum studies frequently give off incorrect signals during breakouts and trending markets. This is especially the case when using shorter timeframes, such as hourly and shorter study periods. The key to understanding why this happens is to recognize that momentum studies are backward-looking indicators. All they can do is quantify the change in current prices relative to what has come before. They have little predictive capacity, which is why you always need to wait for confirmation from prices before trading based on a momentum signal.

Some of the most extreme price moves typically occur when momentum readings are in overbought or oversold territory. Divergences in shorter time frames also appear frequently, especially during breakouts, where rapid price moves are not reflected quickly enough in momentum studies. By the time the momentum indicator has caught up with the price breakout, prices may already have peaked or bottomed, again causing momentum to signal a divergence. Just because momentum is overbought or oversold doesn’t mean prices can’t continue to move higher or lower.

Most technical studies analyze prior price action over specified time periods. The period refers to the time frame in which you’re viewing the technical study, such as 15 minutes, an hour, or days. If you’re looking at a technical study in the hourly time frame, for example, and the study is using 9 and 14 as the two time periods, it means the study is looking at the price action over the last 9- and 14-hour periods for its price data. If you switch to a 15-minute timeframe, the data is drawn from the last nine and fourteen 15-minute periods. The shorter the period (the lower the number), the more sensitive the indicator will be to current prices. The longer the period (the higher the number), the more slowly the study will respond to current price moves. As a result, shorter-period studies tend to be more volatile and generate more signals than longer-period studies. Charting systems typically supply preset default periods, or parameters, for each study, based on the developer of the study or common market usage.

Here are the main momentum oscillators used by currency traders: -

ü  Relative Strength Index (RS1): A single-line oscillator plotted on a scale from 0 to 100, based on closing prices over a user-defined period. Common RSI periods are 9, 14, and 21. RSI compares the strength of up periods to the weakness of down periods - hence, the label relative strength. RSI readings over 75 are considered overbought; readings below 25 are considered oversold. RSI signals are given when the indicator leaves overbought or oversold territory and on divergences with price.

ü  Stochastic: A two-line oscillator plotted on a scale oi 0 to 100. The two lines are known as %K (fast stochastic) and %D (slow stochastic). Stochastics are also based on closing prices of prior periods. The basic theory behind stochastics is that the strength of a directional move can be measured by how near the close is to the extreme of a period. In an uptrend, a close near the highs for the period signifies strong momentum; a close in the middle or below signals that momentum is weakening. In a downtrend, the close of a period should be nearer to the lows for momentum to strengthen. As momentum shifts, the %K line will cross over the slower-moving %D line. Crossovers in overbought or oversold territory are considered sell or buy signals. Overbought is above 80, and oversold is below 20.

ü  Moving Average Convergence/Divergence (MACD): Not really a momentum oscillator, but a complex series of moving averages. (It functions very similarly to momentum studies, so I include it here.) MACD fluctuates on either side of a zero line and has no fixed scale, so overbought or oversold are judged relative to prior extremes. MACD also consists of two lines: the MACD line (based on two moving averages) and the signal line (a moving average of the MACD line). Trading signals are generated if the MACD line crosses up over the signal line while below the zero line (buy) or crosses clown below the signal line while above the zero line (sell). MACD tends to generate signals more slowly than RSI or stochastics due to the longer periods typically used and the slower nature of moving averages. The result is that it takes longer for MACD to crossover, generally preventing fewer false signals.

Trend-identifying indicators

One of the market’s favorite sayings is “The trend is your friend.” The idea is that if you trade in the direction of the prevailing trend, you're more likely to experience success than you are if you trade against the trend. Now, how can you argue with logic like that?

The hard part for us mortals is to determine whether there’s a trend in the first place. The question becomes more complex when you look at multiple time frames, because trends can exist in any timeframe. On a daily time frame, the market may be largely range bound. But in a shorter time frame, such as hourly or 30 minutes, there may be a trending movement that presents a trading opportunity.

Determining whether a trend is in place is also important when it comes to deciding whether to follow the signals given by momentum indicators. Momentum studies are great in relatively range-bound markets, but they tend to give off bad signals during trends and breakouts. The key is to determine whether a trend is in place. In the following, we look at a few technical studies you can use to identify whether a trend is in place and how strong it may be.

Directional Movement Indicator system

The Directional Movement indicator (DMI) system is a set of quantitative tools designed to determine whether a market is trending. The DMI was developed by J. Welles Wilder (author of New Concepts in Technical Trading Systems), who also developed the MACD indicator. Using the DMI removes the guesswork involved with spotting trends and can also provide confirmation of trends identified by trend-line analysis.

The DMI is based on the idea that when a market is trending, each period’s price extremes should exceed the prior price extremes in the direction of the trend. For example, in an uptrend, each successive high should be higher than the prior period‘s high. In a downtrend, the opposite is the case: Each new low should be lower than the prior period's low.

The DMI system is comprised of the ADX line (the average directional movement index) and the Dl+ and Dl- lines (which refer to the directional indicators for up periods [+] and down periods [-]). The ADX is used to determine whether a market is trending (regardless if it’s up or down), with a reading over 25 indicating a trending market and a reading below 20 indicating no trend. The ADX is also a measure of the strength of a trend - the higher the ADX, the stronger the trend. Using the ADX, traders can determine whether a trend is operative and decide whether to use a trend-following system or to rely on momentum oscillator signals.

As its name would suggest, the DMI system is best employed using both components. The Dl+ and Dl- lines are used as trade-entry signals. A buy signal is generated when the Dl+ line crosses up through the Dl- line; a sell signal is generated when the Dl- line crosses up through the Dl+ line. Wilder suggests using the extreme-point rule to govern the Dl+/Dl- crossover signal. The rule states that when the DI+/Dl- lines cross, you should note the extreme point for that period in the direction of the crossover (the high if Dl+ crosses up over Dl-; the low if Dl- crosses up over Dl+). If that extreme point is exceeded in the next period, the Dl+/DI- crossover is considered a valid trade signal. If the extreme point is not surpassed, the signal is not confirmed.

The ADX can also be used as an early indicator of the end or pause in a trend. When the ADX begins to move lower from its highest level, the trend is either pausing or ending, signalling that it’s time to exit the current position and wait for a fresh signal from the DI+/DI- crossover.

Moving averages

One of the more basic and widely used indicators in technical analysis, moving averages can verify existing trends, identify emerging trends, and generate trading signals. Moving averages are simply an average of prior prices over a user-defined time period displayed as a line overlaid on a price chart. There are two main types of moving averages:

ü  Simple moving average gives equal weight to each price point over the specified period.

ü  Exponential moving average gives greater weight to more recent price data, with the aim of capturing directional price changes more quickly than the simple moving average.

In terms of defining a trend, when prices are above the moving average, an uptrend is in place; when prices are below the moving average, a downtrend is in place.

Traders like to experiment with different periods for moving averages, but a few are more commonly used in the market than others, and they're worth keeping an eye on. The main moving average periods to focus on are 21, 55, 100, and 200. Shorter-term traders may consider looking at the 9- and 14- period moving averages.

Another way moving averages are used is by combining two or more moving averages and using the crossovers of the moving averages as buy or sell signals based on the direction of the crossover. For example, using a 9- and 2- period moving average, you would buy when the faster-moving 9-period average crosses up over the slower-moving 21-period average, and vice versa for a crossover to the downside.

Don’t follow any single technical study blindly. Get used to looking at several 41-different studies in conjunction with one another. Use them to assist you and to serve as early warning signals of potential changes in price direction, pending confirmation from prices, such as breaks of trend lines or spike reversals. Also, the time frame of your study directly relates to its timing implications for the future - a very short time frame carries meaning for the relatively brief future, while a daily study will carry implications for several days ahead.