Technical Indicators: Types and How They Work

Technical indicators calculate summaries of market data, such as price direction, recent gains and losses, traded volume or range. They help make an observation measurable. An indicator value is not automatically an entry signal, a forecast or a complete trading strategy.

Start with native LuxAlgo charts, choose a specific question and inspect the relevant calculation. Use Quant, our coding agent, when you want to express or change indicator or strategy logic. Review generated code and run it manually before relying on its results.

Four Main Types of Technical Indicators

Trend, momentum, volume and volatility are useful categories, but their boundaries overlap. MACD combines moving-average relationships with momentum interpretation; VWAP combines price and volume. Choose an indicator by its calculation and role rather than its category label alone.

Trend Indicators: Direction and Smoothed References

Moving Averages

A simple moving average, or SMA, is the mean of a specified set of observations. Closing prices of 100, 102 and 104 give a three-period SMA of 102. An exponential moving average, or EMA, applies a recursive weighting scheme, commonly using a factor of 2 ÷ (length + 1). State the input source, length and initialization.

Price above an average, an average rising, and a short average crossing above a long one are three different conditions. A commonly discussed golden cross uses a 50-day average crossing above a 200-day average; the opposite crossing is called a death cross. Neither event identifies a guaranteed turning point or supplies exits and costs.

Smoothing reduces some short-term variation but changes responsiveness. A longer length may react later, while a shorter length can change direction more often. Neither choice is universally more accurate, and a lookback is measured in chart bars rather than calendar days unless the chart is daily.

MACD

Conventional MACD uses EMA(12) minus EMA(26), with a nine-period EMA of that difference as its signal. The histogram equals MACD minus signal. Its units are price units, so an absolute reading of 2 does not carry the same relative meaning on a $20 and a $200 instrument.

A MACD signal crossover occurs when MACD moves from at or below the signal to above it, or the reverse. The histogram crossing zero describes that same event; counting both does not provide independent confirmation. A MACD-line zero crossing instead describes the fast average crossing the slow average. Check averaging types and settings in the implementation.

Parabolic SAR and ADX

Parabolic SAR produces a trailing reference whose path depends on price extremes and an acceleration factor. Dots switching sides indicate a reversal in its calculation, not proof that the market has reversed. Choppy price action can generate repeated switches. If using SAR for exits, define when its value becomes available and how an order is filled.

ADX summarizes directional-movement strength without giving the direction by itself. A high ADX can occur in either an advance or a decline. A threshold such as 25 or 30 is a candidate filter to test, not a universal guarantee of trend reliability.

Momentum Indicators: More Than Price Speed

Relative Strength Index

RSI compares smoothed recent gains with smoothed recent losses. In its conventional form, RSI = 100 − 100 ÷ (1 + RS), where RS is average gain divided by average loss. If those averages are 2 and 1, RS is 2 and RSI is approximately 66.67. The smoothing method, warm-up and zero-gain or zero-loss handling matter.

RSI normally lies between 0 and 100. Guides such as 70 and 30 describe the oscillator’s position, not a mandatory sell or buy instruction. It can remain elevated during an advance or depressed during a decline. Being below 30, rising while below 30, and crossing back above 30 are distinct conditions.

Stochastic Oscillator

The raw %K calculation places the close within a selected high-low range: 100 × (close − lowest low) ÷ (highest high − lowest low). A close of 108 in a lookback range from 100 to 110 gives 80. This measures location within the range, rather than literal price velocity or the probability of a decline.

Fast, slow and full Stochastic variants apply different smoothing to %K and its companion %D. Record the range lookback and both smoothing choices. If the high-low range is zero, the implementation needs a defined rule rather than an unhandled division by zero.

Divergence and Timing

Regular bullish divergence compares a lower price low with a higher indicator reading at matched observations; bearish divergence compares a higher price high with a lower indicator reading. Define the series, pivot selection and matching rule. If a pivot needs later bars, the divergence cannot be used at the earlier turning point.

Divergence can persist while a trend continues. Treat it as a condition within a complete hypothesis, with a later executable trigger and expiry, rather than a standalone instruction to trade against the trend.

Volume Indicators: Understand the Feed and Formula

Volume is recorded activity on the selected feed. Exchange-specific traded volume is not automatically consolidated market volume, and forex tick activity differs from traded quantity. Aggregate volume cannot identify institutional participation or establish that a move has been validated.

On-Balance Volume and Accumulation/Distribution

OBV adds the current volume when the close exceeds the previous close, subtracts it when the close is lower, and conventionally leaves the total unchanged on an equal close. A one-cent rise and a one-dollar rise receive the same volume sign. OBV is a cumulative transformation, not a count of net buyers.

The Accumulation/Distribution indicator instead weights volume by the close’s position inside the current high-low range. Its multiplier is ((close − low) − (high − close)) ÷ (high − low). Multiply by volume and add that amount to the running total; define handling for a zero-range bar.

For high 110, low 100, close 108 and volume 1,000, the multiplier is (8 − 2) ÷ 10 = 0.6, so the running total increases by 600. If the previous close was 112, OBV would decrease by 1,000. Their disagreement follows directly from different formulas: the current close is high within its own range but below the previous close.

The abbreviation A/D can also refer to an Advance-Decline breadth measure. That counts advancing and declining securities rather than weighting one security’s volume. Specify the full indicator name to avoid confusing the two.

Volume-Weighted Average Price

VWAP is the sum of price times volume divided by total volume over a specified anchor or session. Two illustrative transactions, 100 units at $10 and 300 at $12, give VWAP = (100 × $10 + 300 × $12) ÷ 400 = $11.50. This differs from the unweighted average of $11.

Chart implementations often use a bar price such as typical price instead of every transaction. Specify that source, the volume feed, reset or anchor and session. Price above VWAP means above that calculated average; it does not automatically indicate a profitable long entry. A session VWAP and an anchored VWAP answer different time-window questions.

Volatility Indicators: Range and Dispersion

Average True Range

True Range is the maximum of current high minus low, absolute(high − previous close), and absolute(low − previous close). With high 105, low 103 and previous close 100, True Range is 5 even though the current bar’s high-low range is only 2. This captures the gap in the measurement.

Wilder’s ATR method smooths True Range. For a 14-period ATR after initialization, the next value is (13 × prior ATR + current True Range) ÷ 14. With prior ATR 2 and current True Range 5, the result is 31 ÷ 14, approximately 2.2143.

ATR measures magnitude, not direction. An ATR of 2 is 2% of a price of 100 but 1% of a price of 200. Normalize deliberately when comparing instruments. A stop at a multiple of ATR is a strategy choice; it is not a statistically guaranteed loss boundary.

Bollinger Bands

Classic Bollinger Bands use a moving average with upper and lower bands at a chosen multiple of standard deviation, commonly a 20-period SMA and two deviations. State the source, length and deviation calculation. Band width describes dispersion relative to the middle average, not the next breakout direction.

A band touch does not by itself establish a reversal, and price can continue along a band in a trend. Narrow bands describe compression under the selected calculation. Do not assume two standard deviations mean market prices must remain inside the bands 95% of the time.

Combine Indicators by Role

There is no required count of two-to-four indicators, and one from every family is not automatically a better system. Start with a complete rule and add a study only when it answers a distinct question. More filters can reduce opportunities and change exposure as well as remove some losing trades.

Trading style does not dictate a single indicator family. An intraday strategy can use a slow average for context, and a longer-horizon strategy can use an oscillator. Keep symbol, session, timeframe, adjustment method and source price fixed when comparing variants. Use only higher-timeframe values that were available at the decision.

Turn an Indicator Condition into a Complete Strategy

Consider an illustrative long candidate: MACD crosses above its signal on a completed candle, while RSI is below 30 and higher than its previous completed value. RSI moving from 27 to 29 passes this particular condition. Moving from 29 to 31 does not, because the current reading is no longer below 30; that would instead qualify for a different threshold-crossing rule.

A mirrored short candidate uses a bearish MACD signal crossover with RSI above 70 and falling. Exiting an existing long and opening a short are separate actions: specify whether either, both or neither is allowed. Neither candidate is a recommendation or a complete tested system.

For a hypothetical $50 entry and $48 stop, a $100 risk allowance less $10 estimated total costs leaves $90 for price risk. At $1 per price point per unit, quantity = floor($90 ÷ $2) = 45 units and notional exposure is $2,250. If a gap produces an exit at $47, the loss is $135 before costs. Include contract point values, currency conversion and size increments where applicable.

An indicator helps define a rule; it does not supply the risk budget. Cost estimates that depend on quantity need to be included in the sizing calculation. Check available capital and concentration separately, and do not treat a stop trigger as a guaranteed exit price.

Build and Review in Native LuxAlgo

Open native LuxAlgo charts and use the Indicators picker. Basic studies include familiar calculations such as RSI, MACD, Bollinger Bands and moving averages; the LuxAlgo Library provides additional studies in the same picker. Inspect settings and values instead of relying only on plotted labels.

Ask Quant, our coding agent to implement a precise indicator or strategy. For example: “Create this completed-bar MACD/RSI rule with exposed inputs, next-eligible-price entry, explicit sizing, exits and costs. Do not use future higher-timeframe values.” Inspect the generated code and run it manually. Check individual trades and the time each condition became available.

After that review, use Inputs for exposed numerical parameters and Properties for simulation assumptions. Use Quant again when changing logic. Review results in the native strategy viewer; a rerun after adjusting an input is not a fresh independent evaluation sample.

An indicator plot is not a complete strategy, and a notification is not an execution. Quant helps implement rules you specify rather than automatically establishing an optimal or profitable system.

Video: RSI with MACD, Bands and Candles

The Smart Investor’s 3-minute, 25-second tutorial discusses RSI alongside MACD, Bollinger Bands and candlestick charts. Use its examples to identify the distinct calculation and timing of each condition. It is an educational overview rather than evidence that agreement between those tools guarantees a trade outcome.

Common Mistakes and a Better Review Process

• Confusing calculation with prediction: a high oscillator value or rising average describes data under a formula; its trading value needs separate evaluation.
• Using incomplete information: live-bar readings change, and a historical pivot can be labeled before it was actually detectable.
• Retuning after every loss: choose a review schedule and causal adaptation rules in advance, preserving all versions. Fixed rules support reproducibility; adaptability is not a license to revise history.
• Adding redundant filters: compare each addition on common dates and include changes in trade count, costs, holding time and exposure.
• Selecting only favorable tests: keep a baseline, record the full parameter search, freeze rules for later data and investigate implementation errors separately from ordinary losses.

Distinguish win rate from profitability. Sixty wins averaging $10 and forty losses averaging $20 produce a gross loss of $200 across 100 trades, despite a 60% win rate. Costs make the result worse. Review payoff sizes, drawdown and net performance rather than judging an indicator by the percentage of winning trades alone.

Paper trading and a journal can expose timing or process differences, although simulated fills may differ from live execution. Record the signal timestamp, actual inputs, planned and observed entry, exit, costs and reason for any deviation. Keep no-trade decisions and failed setups as well as winners.

For further calculation references, StockCharts ChartSchool provides indicator explanations. Use current product pages for any pricing or platform limits, and compare implementations on the same data before assuming their values should match.

Frequently Asked Questions

What are the four main types of technical indicators?

Trend, momentum, volume and volatility are common categories. They overlap: for example, MACD uses moving averages and also supports momentum interpretation. Choose a tool by its calculation and intended role.

Does an RSI reading below 30 mean I should buy?

No. It identifies an oscillator condition that can persist during a decline. Being below 30, rising below 30 and crossing above 30 are different rules, each requiring context, execution assumptions and risk controls.

Can volume indicators identify institutional traders?

Aggregate volume transformations such as OBV and Accumulation/Distribution do not identify participants by themselves. Their meaning also depends on the venue and volume feed.

How many indicators should I use?

There is no universal required count. Start with a complete strategy and add a study only when it answers a distinct question. Evaluate its incremental effect rather than assuming more confirmation is better.

Does ATR indicate trend direction?

No. It measures smoothed True Range, including gaps relative to the previous close. Its value is in price units; normalize deliberately for comparisons and do not treat an ATR-based stop as a guaranteed loss boundary.

How can I test indicator rules in LuxAlgo?

Inspect studies in native LuxAlgo charts, ask Quant, our coding agent, to implement complete strategy logic, review the code and run it manually. Check trade timing, sizing and costs, then evaluate frozen rules on later data.