10 Trading Indicators Beginners Should Learn

Direct answer: the most useful beginner indicators are moving averages, RSI, MACD, Bollinger Bands, the Stochastic Oscillator, ATR, ADX, VWAP, Fibonacci retracement and volume. Each is a historical calculation that can describe trend, momentum, volatility, relative price location or activity. None can establish what a price must do next.

Indicators are easiest to learn when treated as labels for evidence, not instructions. First identify the chart timeframe, price input, settings and data source. Then ask what the calculation includes—and what it leaves out, such as news, liquidity, gaps, fundamentals, fees and execution. For a risk-first companion, read 7 Common Trading Mistakes to Avoid. For checking claims and methods, see How to Verify Market Claims in India and Market Data Method.

This guide is part of Trading Guides. It follows the standards described in the editorial policy; read the site disclaimer for the scope of this educational content.

Educational only, not financial advice. Examples below are fictional arithmetic illustrations, not market data, forecasts or recommendations. Product rules, margin terms and short-selling rules vary; check current applicable documents.

At-a-glance summary

1. Moving Averages (SMA and EMA)

What it measures

A moving average is a rolling summary of a series, usually closing prices. It smooths short-term variation so a broad tendency and its pace are easier to see. A simple moving average (SMA) weights each included observation equally; an exponential moving average (EMA) gives newer prices greater weight, so it usually follows price more closely. StockCharts’ moving-average guide and Fidelity’s SMA guide describe these as price summaries, not exact top or bottom finders.

How the calculation works

For an N-period SMA, add the last N closes and divide by N:

SMA_t = (C_t + C_t-1 + … + C_t-N+1) / N

Each new bar replaces the oldest included close. For an EMA, a common smoothing constant is K = 2 / (N + 1), then:

EMA_t = K × (C_t − EMA_t-1) + EMA_t-1

A conventional first EMA is the first N-period SMA. Common examples are 5–20 bars for shorter context, 20–60 for medium context and 100+ for longer context; 50- and 200-day averages are widely watched examples, not universal settings. A “period” means chart bars: 50 daily bars and 50 hourly bars are different inputs.

Fictional example and limits

With fictional closes of ₹102, ₹101 and ₹105, a 3-period SMA is (102 + 101 + 105) / 3 = ₹102.666…, or ₹102.67 when displayed. Keep the unrounded value as the first EMA seed. With K = 2/(3+1) = 0.5 and a next fictional close of ₹107, EMA = 0.5 × 107 + 0.5 × (308/3) = ₹104.833…, or ₹104.83 displayed. Rounding only the displayed result avoids accumulating rounding errors. This shows responsiveness, not a forecast.

All moving averages lag. Shorter lengths react sooner but can be noisier; longer lengths smooth more but react later. EMA values can differ early on because seed, warm-up history, input field, session treatment, adjustments and rounding vary across platforms. A crossing or slope can describe changing context, but in a sideways market it can reverse quickly. Do not treat proximity to an average as support or resistance that must hold.

2. Relative Strength Index (RSI)

What it measures

RSI is a 0–100 momentum oscillator developed by J. Welles Wilder. It compares the size and persistence of recent upward and downward price changes. A higher reading means recent upward changes have been larger relative to downward changes; it does not mean price must rise next. StockCharts and Fidelity document the conventional 14-period approach.

How the calculation works

RSI = 100 − 100 / (1 + RS), where RS = average gain / average loss; losses are positive magnitudes in the ratio. Wilder’s common default is 14 periods. The first averages are 14-period sums divided by 14; later averages are smoothed as [(prior average × 13) + current gain or loss] / 14. Traditional reference zones are above 70 and below 30, but the lookback and thresholds can change. The selected timeframe defines a period.

Fictional example and limits

For seven fictional consecutive close-to-close changes of +₹5, −₹4, +₹4, −₹3, +₹3, −₹4 and +₹3, gains total ₹15 and loss magnitudes total ₹11. Using the first simple seven-period averages solely to show arithmetic, RS = (15/7)/(11/7) = 15/11, and RSI = 100 − 100/(1 + 15/11) = 57.69, displayed to two decimals. This is not the usual 14-period default or an observed reading; subsequent readings would require the chosen smoothing method.

Above 70 and below 30 are descriptive conventions, not automatic reversal points. A strong trend can remain elevated or depressed for a long time; changing input price, interval, warm-up history or smoothing can change the displayed value. Divergence—price making a new extreme not confirmed by RSI—can be useful context but can also persist. RSI measures price momentum, not value, liquidity or future return.

3. MACD (Moving Average Convergence/Divergence)

What it measures

MACD is an unbounded momentum oscillator built from two EMAs. It combines a trend comparison—whether the faster EMA is above the slower one—with a measure of their separation. StockCharts’ MACD reference uses closing prices and documents its common settings; Fidelity’s guide notes its tendency to whipsaw in trading ranges.

How the calculation works

The usual structure is:

• MACD line = EMA(12) − EMA(26)
• Signal line = EMA(9) of the MACD line
• Histogram = MACD line − Signal line

Positive MACD means the faster EMA is above the slower EMA; negative means it is below. A signal-line crossover describes MACD moving above or below its signal line. A centerline crossover describes the faster EMA moving above or below the slower one. The 12/26/9 values are typical defaults, not requirements; on a daily chart they mean bars of days, and on another timeframe they mean bars of that timeframe.

Fictional example and limits

If a fictional 12-period EMA is ₹105 and a 26-period EMA is ₹102, MACD is ₹3. If the fictional 9-period signal EMA is ₹2.40, the histogram is ₹0.60. That states only the gap between averages and the gap from the signal line.

MACD is lagging because its inputs are moving averages. Crossovers can be frequent and misleading in a range, while divergences can appear during a continuing strong move. It has no fixed upper/lower bounds, and absolute values are price-level-dependent, so they are not directly comparable across differently priced securities. EMA seeds, price adjustments, missing bars and rounding can affect reproduced values.

4. Bollinger Bands

What it measures

Bollinger Bands are a three-line price overlay: a middle moving average with upper and lower volatility envelopes. Standard deviation causes bands to widen as recent variation rises and contract as it falls. StockCharts explicitly notes that high or low is not automatically a sell or buy conclusion; Fidelity likewise describes tightening and separation as volatility context.

How the calculation works

Using a chosen series—commonly closing price—and n bars:

• Middle band = n-period SMA
• Upper band = SMA + k × standard deviation
• Lower band = SMA − k × standard deviation

A common default is 20 periods and 2 standard deviations. Fidelity lists 10/1.5, 20/2 and 50/2.5 as illustrative combinations, not personal settings. A platform can differ in population versus sample standard deviation, price input, handling of incomplete bars and missing data.

Fictional example and limits

For fictional prices ₹100, ₹102, ₹98, ₹101 and ₹99, the 5-period SMA is ₹100.00. Using population standard deviation, SD is about ₹1.4142. With k = 2, bands are about ₹102.83 and ₹97.17. A hypothetical next close of ₹103 is high relative to this frozen example window, not proof of a reversal. A real rolling calculation would drop the oldest close and recalculate its bands when the next close enters the window.

A narrow band describes lower recent volatility, while widening describes higher recent volatility. A price can keep “walking” a band in a strong trend. A squeeze may precede a sharp move, but its direction is unknown and false breaks occur. The bands do not guarantee that future prices stay inside them.

5. Stochastic Oscillator

What it measures

The Stochastic Oscillator locates the latest close within a selected high–low range on a 0–100 scale. It is a momentum and price-location measure, not a statement about intrinsic value or future direction. StockCharts distinguishes fast, slow and full forms; Fidelity says it is most effective in broad ranges or slower trends.

How the calculation works

%K = 100 × (C − L_N) / (H_N − L_N), where C is the latest close and L_N/H_N are the lowest low/highest high in the last N bars. %D is commonly a 3-period SMA of %K. A 14-period lookback and 3-period smoothing are conventional examples. Slow and full versions add configurable smoothing. Above 50 means the close is in the upper half of the selected range; traditional reference zones are above 80 and below 20.

Fictional example and limits

With a fictional 14-bar lowest low of ₹140, highest high of ₹160 and latest close ₹155, %K = 100 × (155−140)/(160−140) = 75. If the previous two raw %K values were 65 and 70, the current three-period %D is (65 + 70 + 75) / 3 = 70. This fast-stochastic example says the close is relatively high in that chosen range—not that a price will change direction. A slow-stochastic version applies additional smoothing and can show different lines.

Extreme readings can persist in strong trends. Shorter lookbacks are choppier; longer lookbacks are smoother. If high equals low, the denominator is zero and a platform must define handling. Bar construction, smoothing, data history and the timeframe can change a reading. Treat zone moves, %K/%D comparisons and divergence as observations to test with other context, not automatic instructions.

6. Average True Range (ATR)

What it measures

ATR estimates typical absolute price movement per bar, including gaps. It is a volatility measure, not a directional measure: rising ATR means ranges are expanding, but either upward or downward movement may be involved. StockCharts’ ATR guide and Fidelity’s guide document this distinction.

How the calculation works

For each bar, true range (TR) is the greatest of:

• current high minus current low;
• the absolute difference between current high and previous close; or
• the absolute difference between current low and previous close.

A common 14-period ATR begins as the average of the first 14 TR values, then uses Wilder smoothing: ATR = [(prior ATR × 13) + current TR] / 14. The general form is [(prior ATR × (n−1)) + TR] / n. It can be applied to intraday, daily, weekly or monthly bars. Different chart histories, seeds and rounding can create modest differences.

Fictional example and limits

Suppose a fictional bar has previous close ₹102, high ₹110 and low ₹104. TR is the maximum of ₹6, ₹8 and ₹2, so ₹8. If prior 14-period ATR was ₹4, updated ATR is [(₹4 × 13) + ₹8] / 14 = ₹4.2857, rounded. It says the smoothed typical range rose; it does not state an up or down forecast.

ATR is in price units, so ₹5 of ATR is not directly comparable between differently priced securities; a normalised measure may be needed for comparison. Gaps, price limits, illiquidity, corporate actions and news can make a recent TR unusual. An ATR-derived level cannot guarantee execution at a chosen price. Position notional is also different from planned loss, and product leverage can change exposure materially.

7. ADX (Average Directional Index)

What it measures

ADX is Welles Wilder’s smoothed trend-strength gauge. It measures how strong a directional move is, not whether it is upward or downward. Companion +DI and −DI lines provide directional context. StockCharts’ ADX guide documents the component formulas and long warm-up need; Fidelity distinguishes ADX strength from DMI direction lines.

How the calculation works

For each bar, calculate TR and positive/negative directional movement from changes in highs and lows. Wilder-smooth +DM, −DM and TR, then calculate:

• +DI = 100 × smoothed +DM / smoothed TR
• −DI = 100 × smoothed −DM / smoothed TR
• DX = 100 × |+DI − −DI| / (+DI + −DI)

The first ADX is the average DX over the selected length; later ADX is [(prior ADX × (n−1)) + current DX] / n. A 14-period example is common. StockCharts notes roughly 150 periods may be needed for stable values.

Fictional example and limits

With fictional smoothed +DM = ₹6, −DM = ₹2 and TR = ₹20, +DI = 30 and −DI = 10; DX = 50. If 14 fictional DX values sum to 560, first ADX is 40. These are indicator units, not rupee returns.

Readings below about 20, above about 25 and the interval between them are common conventions—not guarantees. A high ADX with −DI above +DI describes stronger downward-direction context, not a bullish conclusion. Heavy smoothing creates lag; DI crossings can whipsaw during consolidation. Seeds, adjustments and bar definitions can differ by platform.

8. VWAP (Volume-Weighted Average Price)

What it measures

VWAP is the average traded price weighted by volume during a session. Price above it is relatively higher than that session’s volume-weighted average; price below it is relatively lower. It is a cumulative intraday reference, not a fair-value proof or forecast. StockCharts describes traditional session VWAP, while Charles Schwab notes its daily reset.

How the calculation works

For each intraday bar, calculate Typical Price = (High + Low + Close) / 3. Multiply it by volume, sum those products and divide by cumulative volume:

VWAP = Σ(TP × Volume) / ΣVolume

Traditional VWAP starts at the first trade or open and ends at the close. It has no rolling decay; later bars join a growing total, so it generally changes more slowly later in the session. StockCharts documents intraday periods such as 1, 5, 10, 15, 30 and 60 minutes. Anchored or range VWAP is a different convention and must be labelled as such.

Fictional example and limits

For three fictional bars (H, L, C, volume) of (₹101, ₹99, ₹100, 100), (₹104, ₹100, ₹102, 300) and (₹103, ₹101, ₹102, 600), typical prices are ₹100, ₹102 and ₹102. Σ(TP×volume) is ₹101,800 and total volume is 1,000, so VWAP is ₹101.80. A final close of ₹102 is ₹0.20 above that reference; it is a relative position, not an instruction.

VWAP normally resets for a new session. Platforms may use ticks or bars, differing sessions, adjusted volume or different anchors; compare settings before comparing values. Early-session VWAP is based on few observations, while later-session VWAP lags more. Volume weighting does not identify a winning side of a trade—every completed trade has both a buyer and seller.

9. Fibonacci Retracement

What it measures

Fibonacci retracement overlays percentage levels on a selected prior swing to mark areas where a pullback or counter-trend bounce might pause. It is a chart annotation tool, not a mathematical prediction. StockCharts calls levels alert zones rather than hard reversal points; Fidelity describes possible support/resistance areas. NSE India identifies the series and level formulas as course topics, not performance evidence.

How the calculation works

Choose a visible swing low and high on a stated timeframe. Common displayed levels are 0%, 23.6%, 38.2%, 50%, 61.8% and 100%. For an upward swing, a downward retracement level is:

high − (high − low) × ratio

The 23.6%, 38.2% and 61.8% figures arise from Fibonacci number relationships. The 50% level is a market convention, not a Fibonacci ratio. Anchor selection is subjective: another analyst, swing definition or timeframe can produce different lines.

Fictional example and limits

If a fictional share rises from ₹100 to ₹160, the range is ₹60. Downward levels are ₹145.84 at 23.6%, ₹137.08 at 38.2%, ₹130 at 50%, and ₹122.92 at 61.8%. These are calculated reference zones only. Price can pass through them, gap over them or change direction elsewhere.

The tool is retrospective and does not establish probability, causation or overall trend direction. A touch of a level does not confirm a reversal. Use a clearly stated timeframe and anchors, and do not convert a line into a target or certainty.

10. Volume

What it measures

Volume is the number of shares, contracts or other units traded in a selected bar—not simply the number of transactions. It is activity, not direction: every completed trade has a buyer and seller. StockCharts’ TA 101 material explains that the traded share counts from transactions are summed into volume bars; Fidelity’s Volume Oscillator guide reinforces that volume must be compared with a stated baseline.

How the calculation works

For a chosen timeframe, Volume_t is the sum of all units in transactions included in that bar. A simple baseline is:

Average Volume_t = (Volume_t + … + Volume_t-n+1) / n

The chart period may be intraday, daily, weekly or longer; there is no universal best setting. Compare like with like—same bar duration, comparable sessions and an understood data feed. Low activity can affect liquidity and price impact; high activity does not guarantee execution at the last price.

Fictional example and limits

Five fictional daily volumes of 1,000, 1,200, 800, 1,500 and 2,500 shares total 7,000 shares, or a five-day average of 1,400. The final day is above that baseline, describing higher activity in this example. If an illustrative price were ₹50, 7,000 × ₹50 = ₹350,000 is a simplified traded-share notional that ignores changing prices. Neither number proves a coming rise or fall.

Bar lengths, holidays, partial sessions, reporting coverage, corporate actions and venue conventions can distort comparison. Investor.gov’s order explainer notes that market orders do not guarantee an execution price. Volume can help describe participation and liquidity, but it cannot by itself identify cause, continuation or a reliable action.

Frequently asked questions

Are these indicators enough to decide on a trade?

No. They are price- or volume-derived descriptions with parameter choices and blind spots. They do not replace understanding the instrument, liquidity, costs, current rules or risk exposure.

Which timeframe is the right one for beginners?

There is no universally correct timeframe. State the timeframe because a 14-period or 20-period reading changes meaning when a bar changes. Learn one timeframe’s mechanics before comparing it with another.

Why can two platforms show different values?

They may use different data histories, EMA seeds, price fields, session definitions, adjustments, standard-deviation conventions, rounding or bar construction. Confirm settings before treating a difference as meaningful.

Does a crossover, band touch or extreme reading predict a reversal?

No. Such events describe the indicator under its chosen settings. Trends can persist, ranges can whipsaw and gaps or news can make past-price calculations stale.

How should I practise without assuming certainty?

Use fictional examples or a paper journal. Record the indicator, settings, timeframe, observation, alternative explanation and what could invalidate the interpretation. Review claims using the market-claim verification guide, not hindsight alone.

Before you practise: checklist

• Name the chart timeframe, input series, settings and source.
• State what the indicator measures in one sentence and what it cannot measure.
• Confirm whether the calculation is session-based, rolling or dependent on warm-up history.
• Keep position notional separate from any planned loss amount.
• Consider liquidity, gaps, costs, leverage and order execution rather than assuming an indicated level will hold.
• Write an observation and an alternative explanation; do not convert a chart label into certainty.

Conclusion

The best first use of an indicator is to make an observation more precise: trend smoothing, recent momentum, volatility, range position or traded activity. That precision is valuable only when its settings and limits remain visible. Start with the risk and process lessons in common trading mistakes, check how data were built in Market Data Method, and keep a written record of what each calculation did—and did not—say.

Educational only, not financial advice. Historical technical calculations cannot remove market, liquidity, execution, leverage or product risk.

Selected sources

• StockCharts ChartSchool: Moving Averages; RSI; MACD; Bollinger Bands; Stochastic Oscillator; ATR; ADX; VWAP; Fibonacci retracements; TA 101, Part 3.
• Fidelity Learning Center: SMA, EMA, RSI, MACD, Bollinger Bands, Fast Stochastic, ATR, Fibonacci retracement and Volume Oscillator.
• NSE India: Elliott Wave Principle & Fibonacci Retracement; Investor.gov: Types of Orders; Charles Schwab: Volume-weighted indicators.