Technical Indicators: How the Common Indicators Work and Combine

This guide sorts the common indicators into types, compares how they behaved on one model Nifty series, explains leading and lagging indicators, how to combine indicators without repeating the same information, the default settings, and how indicators are used for index options in India.

Key takeaways

• Indicators fall into five broad types: trend, momentum, volatility, volume and price levels. Each answers a different question about the market.
• Every indicator is calculated from past data, so every indicator lags. "Leading" usually means an oscillator that can warn of fading momentum before the price turns, which is not the same as forecasting.
• Combining two indicators of the same type adds little. A trend measure, a timing measure and a volatility measure for stops cover more ground than three momentum oscillators.
• Standard settings are a sensible start. Settings tuned until they fit the past usually fail on new data.
• On Indian index options, indicators are read on the Nifty or Bank Nifty index or futures, and the option is the instrument, not the chart.

Types of technical indicators

Some indicators sit across types. MACD is built from moving averages, a trend tool, but its histogram measures momentum. Bollinger Bands combine a trend line, the middle band, with a volatility measure, the band width. Knowing what each part measures is what keeps a set of indicators from repeating itself.

The model Nifty series

Every number on this page comes from one model series of 72 daily closes, a seeded random walk that is not market data. The same series is used in the individual guides to RSI, moving averages, MACD and Bollinger Bands.

What four indicators said on six days

The table shows how the types overlap. At the high and the low, RSI, the distance from the average and %B all said the same thing: the move was large and fast. On days 37 and 52 all three trend and momentum measures changed side together. The MACD histogram needs 34 sessions of data, so it had no reading at the high.

Leading and lagging indicators

Indicators are often divided into leading and lagging ones. Both kinds are calculated from past prices, so neither forecasts.

On the model series no indicator turned before the price. RSI(14) peaked on day 32 and bottomed on day 44, the same sessions as the price's high and low. The MACD line peaked two sessions after the high and bottomed four sessions after the low. What the "leading" indicators offer is not an earlier turn but a reading of how stretched a move is and, through divergence, a warning that the next push has less force.

When each kind of signal came

Short price-action rules signalled first, the momentum and trend indicators together a little later, and the slowest trend signal, MACD crossing zero, last. A faster rule also gives more false signals: the same 5-session rule broke a recent high or low nine times in the first 25 sessions, when the index ended only 120 points from where it started.

Moving averages

A moving average is the average close over a set number of periods. The simple moving average weights every close equally; the exponential moving average puts more weight on the latest closes and turns sooner. On the model series the 20-day EMA turned up 7 sessions after the low and the 20-day SMA 13 sessions after it. Moving averages are the base of many other indicators, including MACD, Bollinger Bands and Keltner Channels.

Traders use them for the direction of the trend, as zones where pullbacks stall, as filters for other signals and as trailing exits. The calculations, the lag of each type and the common periods are worked through in the guide to moving averages.

RSI

The Relative Strength Index compares average gains with average losses over 14 periods and puts the result on a 0 to 100 scale. Readings above 70 are called overbought and below 30 oversold, but in a trend RSI tends to stay in a range, about 40 to 80 in a rising market and 20 to 60 in a falling one. On the model series it stayed between 55 and 81 during the recovery.

RSI is used to time entries within a trend, to filter trades by the 50 line and in mean-reversion rules. Its formula, settings and the reasons it differs between platforms are covered in the guide to the RSI indicator.

MACD

MACD is the gap between a 12-period and a 26-period EMA, with a 9-period EMA of that gap as its signal line and the difference between the two drawn as a histogram. A cross above zero means the faster average has moved above the slower one; a cross of the signal line means the gap is changing direction. MACD is measured in points, so its values cannot be compared across instruments.

On the model series the signal-line crosses came 5 and 8 sessions after the turns, and the histogram turned one session after the low. The calculation, each part and the settings are in the guide to the MACD indicator.

Bollinger Bands

Bollinger Bands draw a band two standard deviations above and below a 20-period SMA. They widen when the market is volatile and narrow when it is quiet. %B places the close within the bands and bandwidth measures how far apart they are. A squeeze, the narrowest bands in a long while, shows unusually low volatility but not the direction of the next move.

On the model series bandwidth ran from 2.02% before the rally to 7.92% after the fall, and closes fell outside the bands on 9 of 53 sessions. The formula and the two opposite ways traders use the bands are explained in the guide to Bollinger Bands.

Stochastic oscillator

The stochastic oscillator, developed by George Lane, compares the close with the highest high and lowest low of the last 14 periods. %K = (close − lowest low) ÷ (highest high − lowest low) × 100, and %D is a 3-period average of %K. Readings above 80 are called overbought and below 20 oversold.

The stochastic reacts faster than RSI because it compares the close with the period's range rather than averaging gains and losses, so it reaches its extremes more often. It suits ranges and short-term timing. Using it alongside RSI repeats most of the same information, since both are momentum oscillators.

Average true range (ATR)

ATR, introduced by J. Welles Wilder, measures how much an instrument moves in a typical period. The true range of a period is the largest of the high minus the low, the high minus the previous close and the previous close minus the low, so it includes gaps. ATR is a smoothed average of the true range, usually over 14 periods.

ATR does not give signals. It sizes stops and positions to the market's current movement: a stop two or three ATR from the entry is wider when the market is volatile and tighter when it is quiet. It is also the band distance in Keltner Channels and the base of Supertrend.

ADX

The average directional index, also Wilder's, measures how strong a trend is, whatever its direction. It is built from two lines, +DI and −DI, that compare upward and downward moves. A common reading is that ADX above 25 shows a trend and below 20 shows a range.

ADX is mostly used as a regime filter: trend-following rules such as moving average crossovers only when ADX is high, mean-reversion rules only when it is low. It does not say which way the trend runs; +DI and −DI do.

Supertrend

Supertrend draws a line at a multiple of ATR below the price in an uptrend and above it in a downtrend, and flips sides when the price closes through it. The common setting is a 10-period ATR with a multiplier of 3. It is popular on Nifty and Bank Nifty charts because it gives one clear line for direction and a trailing stop at the same time. Like every trend tool it whipsaws in sideways markets.

VWAP

The volume-weighted average price is the average price of every trade in the session, weighted by volume, and it resets at the open. Intraday traders use it as the day's fair price: a price above VWAP means buyers have paid more than the day's average. The Nifty and Bank Nifty indices have no traded volume, so VWAP is read on the futures. Bands at one and two standard deviations around VWAP are used much as Bollinger Bands are, but within one session.

Volume and on-balance volume

Volume itself is the most direct participation measure: a breakout on heavy volume carries more conviction than one on light volume. On-balance volume, introduced by Joe Granville, adds each up-day's volume to a running total and subtracts each down-day's. A rising price with falling OBV suggests the rise has less participation behind it. On the index, use the futures' volume.

Pivot points and the central pivot range

Pivot points are levels calculated from the previous session's high, low and close. The pivot P = (high + low + close) ÷ 3. The first resistance R1 = 2 × P − low and the first support S1 = 2 × P − high. The central pivot range (CPR) adds a bottom level, (high + low) ÷ 2, and a top level, 2 × P minus the bottom level.

Pivots are levels, not signals. They are used for targets and stops, and as places where other signals carry more weight.

Parabolic SAR and Ichimoku

Two more trend tools appear on most platforms. Parabolic SAR, another of Wilder's, places a dot below the price in an uptrend and above it in a downtrend; the dots move closer to the price each period by an acceleration factor, so the indicator works mainly as a trailing stop that tightens as a trend matures. Ichimoku Kinko Hyo, developed by Goichi Hosoda, draws five lines from the midpoints of recent highs and lows, including a shaded cloud projected forward. It combines trend direction, momentum and support and resistance in one display, which makes it dense to read and easy to over-interpret.

Candlestick patterns

Candlestick patterns are not calculations but named shapes of one to three candles, such as the hammer, the engulfing pattern and the doji. They are the most common way of reading price action directly. Defined with numbers, for example a lower shadow at least twice the body, a pattern becomes as testable as an indicator. Patterns carry more meaning at a level, such as support or a pivot, than in the middle of a range. The shapes, their definitions and how to test them are in the guide to candlestick patterns.

Combining indicators

The aim of a combination is to cover different questions, not to ask the same one several times.

On the model series RSI, MACD and the 20-day SMA gave their signals on the same sessions. A rule that required all three would have traded on exactly the same days as a rule that used any one of them, with three times the settings to fit. Indicators of different types disagree more often, and that disagreement is the information a combination is for.

A practical limit is one indicator per role, two or three in all. Each extra condition removes trades, and a rule with many conditions often matches the past well because it was fitted to it.

A worked combination on the model series

Take one indicator for each role:

• Direction: the close above its 20-day EMA, which held from day 51.
• Timing: a pullback in which RSI(14) falls for two sessions and then turns up, while staying above 50.
• Risk: a stop just below the lowest close of the pullback.

In the recovery the first such pullback took RSI(14) from 66.09 on day 59 to 62.53 and then 60.06, with Nifty easing to 25,037.15 on day 61. On day 62 RSI rose to 65.28 and Nifty closed at 25,191.55: direction, timing and a defined stop all lined up. The stop below 25,037.15 put the risk at 154.40 points, or ₹10,036 on one lot of 65 in Nifty futures.

The example also shows the sizing problem every rule meets. With a ₹5,00,000 account and 1% risk per trade, the budget is ₹5,000, less than one futures lot's risk at that stop. The choice is to trade a smaller instrument, such as an option or a spread with a lower rupee risk, to accept a wider share of the account, or to skip the trade.

Choosing indicators by market condition

Default settings

The defaults are not optimal; they are common, which makes results comparable and keeps settings from being fitted to one stretch of data. If a rule only works with RSI(11) and not RSI(10) or RSI(12), the result is more likely a fit to the past than a feature of the market.

Indicators on different timeframes

The calculations are the same on any timeframe; only the time each period covers changes. The NSE equity and F&O session runs from 9:15 am to 3:30 pm, 375 minutes.

Shorter timeframes give more signals, more noise and higher costs a month. A common approach takes direction from a slower chart and timing from a faster one, such as a daily trend filter with entries on a 15-minute chart.

Technical indicators for options

Indicators are built for prices that move with supply and demand. An option's price also falls with time and moves with implied volatility, so indicators on an option's own chart mix three effects. The usual approach is:

1. Read the indicators on the underlying. Use the Nifty or Bank Nifty index, or the near-month futures, which carry volume for VWAP and OBV.
2. Turn the signal into a view. A bullish signal is a view that the index will rise, by roughly how much and over roughly how long.
3. Choose the option to fit the view. A bought call needs the move to come soon, because of time decay; a bull call spread costs less and decays more slowly; a bull put spread earns if the index simply stops falling.
4. Size and exit on the option. The stop and target can be set on the index level or on the option premium, but the position size comes from the rupee loss at the stop.

Indicators do not see implied volatility. A signal before a scheduled event, such as the RBI policy or a company's results, can be right on direction and still lose on a bought option if implied volatility falls after the event.

Indicators on Indian markets

• No volume on the index. Volume-based indicators need the futures or the constituent stocks.
• The opening gap. Nifty often opens away from the previous close, which moves every indicator in the first candle. Many intraday rules ignore the first 5 to 15 minutes.
• Futures rollover. A continuous futures chart jumps at each monthly expiry unless the data is adjusted, and the jump passes into every indicator.
• Expiry days. Index option expiries bring fast moves late in the session that ordinary intraday readings may not reflect well.
• Corporate actions. Splits and bonus issues change a stock's price without changing its value, so indicators need adjusted data.

Closed candles and repainting

An indicator's value changes while a candle is forming. A rule that acts on the live value can see a signal that disappears at the close, so most tested rules act only on closed candles. Some tools also repaint: a swing high is confirmed only after the next few candles, and an indicator that marks it on the earlier candle shows a signal that was not available at the time. A backtest that uses such values looks better than live trading can be.

The limits of technical indicators

• They lag. All are built from past data.
• They repeat each other. Indicators of one type carry largely the same information.
• They depend on the data. The starting point, the data feed and adjustments change the values.
• They work in some conditions and fail in others. Trend tools whipsaw in ranges; oscillators stay at extremes in trends.
• They do not forecast. An indicator describes what has happened in a fixed way, which is useful because it is consistent, not because it predicts.

Common mistakes

• Stacking similar indicators. RSI, stochastic and a momentum oscillator together are one opinion stated three times.
• Tuning settings to the past. Keep the defaults until a test on several years and instruments justifies a change.
• Ignoring the market condition. A good range rule loses in a trend, and the other way round.
• Reading indicators on an option's chart. Use the underlying.
• Acting on unclosed candles. Signals can disappear before the close.
• Leaving out costs. Fast indicators trade often, and charges and slippage can turn a small edge into a loss.

Indicators in rule-based trading

An indicator becomes a rule when each condition is written as a number: the 15-minute close above the 20-period EMA, RSI(14) above 55, ATR-based stop at twice the 14-period ATR. Written that way, a rule can be tested on past data with costs, run on live prices with virtual money, and then automated.

In QuantPower's option strategy builder, indicator conditions on the underlying can be used for entries and exits, together with price levels, time windows and conditions such as the index being above or below a moving average. Strategies can be tested in the backtesting engine on five years of 1-minute data and run in paper trading before they go live through your broker.

Frequently asked questions

Related guides

• Moving averages: SMA and EMA, their lag and their uses
• RSI indicator: the formula, settings and ranges of RSI
• MACD indicator: the lines, histogram and settings
• Bollinger Bands: %B, bandwidth and the squeeze
• Candlestick patterns: reading price action candle by candle
• Indicator vs price action: how the two methods compare
• Option strategy builder: indicator conditions in a rule-based strategy
• Backtesting: testing indicator rules on past data

Sources

• Wilder, J. W. (1978). New Concepts in Technical Trading Systems. Trend Research
• Murphy, J. J. (1999). Technical Analysis of the Financial Markets. New York Institute of Finance
• Bollinger, J. (2001). Bollinger on Bollinger Bands. McGraw-Hill
• Appel, G. (2005). Technical Analysis: Power Tools for Active Investors. FT Prentice Hall
• Granville, J. (1963). Granville's New Key to Stock Market Profits. Prentice-Hall
• NSE: trading hours for the equity and equity derivatives segments