Hook — a smooth line can hide sharp changes

A learner opens an imaginary chart containing five daily prices that jump up and down. A moving-average line is smoother than the underlying series, so it may appear easier to understand. But smoothing does not make tomorrow's value known. The simple moving average, or SMA, takes a specified number of past selected prices, adds them, and divides by that number. When a new completed observation enters the window, the oldest one leaves. This rolling process makes the number change over time. The line is a summary of the selected interval and data source, not an executable market price or a stop-loss guarantee. It matters whether the chart uses daily closes, five-minute closes, bid quotes or adjusted prices. This chapter teaches the arithmetic of a five-point SMA and why its reaction is delayed, without promoting crossover strategies or a trade.

Concept in depth — equal weighting creates both clarity and lag

An N-period simple moving average places equal weight, 1/N, on each of the latest N selected completed observations. For a five-period average, each contributes 20%. A new price of ₹106 does not replace the whole SMA with ₹106: only one-fifth of that new observation enters while the oldest contribution drops out. As the lookback N grows, the line generally becomes smoother and responds less quickly to a single new move, but this relation depends on the actual data series. A two-period average reacts more to the most recent pair than a five-period one. The word 'period' means chart bars, not necessarily calendar days; five five-minute bars cover a different window from five completed daily bars. Market holidays, missing observations and adjustments can affect how an app selects bars. The SMA is backward-looking by construction. Lag is not a software fault; it follows from using older prices in the average.

Concept 1

A simple moving average (SMA) adds the last N selected prices and divides by N; each observation receives equal weight. A five-bar SMA is not necessarily five calendar days if bars are intraday. Source A · Source B. Desi example: A kirana shop averages the last five daily rice-bag closing quotes to see a smoother history. Hypothetical worked example: Fictional closes ₹100, ₹102, ₹101, ₹103, ₹104: sum 510; five-bar SMA =510÷5=₹102. No real instrument is implied. Invented classroom numbers; no live quote, conversion, fees or trade instruction.

Concept 2

The average is 'moving' because the oldest observation leaves and a new one enters at the next bar. Recalculate with the same window length, not all prices ever recorded. Source A · Source B. Desi example: A mandi clerk removes Monday's price when adding the next day's rate to a five-entry notebook. Hypothetical worked example: Imaginary new close ₹106 replaces oldest ₹100. New set 102,101,103,104,106 totals 516; new five-bar SMA =516÷5=₹103.20, versus prior ₹102. Invented classroom numbers; no live quote, conversion, fees or trade instruction.

Concept 3

SMA smooths short-term jumps but lags them. A shorter window reacts faster; a price crossing an average is not a reliable instruction or guarantee. Source A · Source B. Desi example: A truck company compares a two-trip freight average with a five-trip average after a sudden costly trip. Hypothetical worked example: Imaginary latest prices ₹104 and ₹106 give two-bar SMA (104+106)÷2=₹105. Five-bar SMA is ₹103.20; difference ₹1.80. Latest close ₹106 is ₹2.80 above five-bar SMA; tomorrow is unknown. Invented classroom numbers; no live quote, conversion, fees or trade instruction.

How it works — rolling window as a transparent ledger

First fix a source, currency, bar length and a lookback N. Check that the bars are complete; an unfinished current interval's close may later change. Write the latest N closing observations in chronological order and add them. Divide the sum by N, keeping the output in the same quoted price unit. On the next observation, cross out the oldest price, include the newest and divide the new total by N. Verify the change by computing (newest − dropped oldest)/N: this shortcut is algebraically equivalent when N is unchanged. Be careful not to average the already-computed SMAs again and claim it is the exact next-period SMA. If fewer than N eligible observations exist, a conventional fully formed SMA cannot yet be computed unless the chart has a special start-up rule. Lastly compare the latest close with the computed line as two descriptive values, not an instruction to buy or sell.

Desi example — the rice seller's five-tag notebook

Desi udaharan: Desi example: A fictional kirana shop writes the final rice-bag asking price for five days: ₹100, ₹102, ₹101, ₹103 and ₹104. The owner averages them to avoid focusing excessively on one unusual day. After the sixth day's tag of ₹106 appears, the owner discards the oldest ₹100 entry and averages ₹102, ₹101, ₹103, ₹104 and ₹106. The resulting number summarizes the latest five selected tags, but it is not the exact price at which the next shopper can buy rice. Nor does it reveal how many bags sold or the shop's net profit, since procurement and overhead are absent. If two branches choose different closing times they may create different equally valid averages. The analogy illustrates rolling selection and unit discipline; financial markets have execution and contractual rules beyond a retail price diary.

ILLUSTRATIVE ONLY — two exact SMAs and a rolling shortcut

Use fictional five-bar closes ₹100, ₹102, ₹101, ₹103 and ₹104. The total is ₹510 and SMA5 = 510/5 = ₹102. Add an invented new closing price ₹106 and remove the oldest ₹100. The new list ₹102, ₹101, ₹103, ₹104, ₹106 totals ₹516 and SMA5 = 516/5 = ₹103.20. The shortcut change is (106−100)/5 = ₹1.20, matching 103.20−102.00. For comparison, the latest two completed closes ₹104 and ₹106 give SMA2 = (104+106)/2 = ₹105, closer to the latest ₹106 than the five-bar line. A current closing price of ₹106 exceeds SMA5 ₹103.20 by ₹2.80, but this is a historical comparison only. In another independent dollar example, $20 and $22 average to $21, with no assumed rupee-dollar conversion. All market-style prices here are made up.

Key points

- SMA is the equal-weight mean of a fixed window. - Rolling forward drops one old observation and adds one new. - Smoothing creates lag, not a prediction.

Common mistakes

- Keeping the dropped value in the next SMA. - Calling an SMA crossing a guaranteed profit signal.

Common Mistakes — averaging the wrong window

A frequent error is adding the new close to the previous five closes and dividing by six, then labelling the result SMA5. Another is keeping the old ₹100 in the five-bar window after the newest ₹106 appears; this silently changes the period count. Some learners calculate an average of five calendar dates while ignoring weekends or market holidays, though a chart's five bars may exclude those days. An SMA's smooth appearance does not mean volatility or risk has disappeared; abrupt current prices can sit far from it. A crossover between a price and an SMA, or between two SMAs, does not prove that the next price will continue in a particular direction. Look-ahead bias occurs if a supposedly historical SMA includes a later price unavailable at the time. Published account returns require actual fills, sizes, commissions and other costs, none of which this arithmetic contains.

Real-world-style scenario — a suspiciously perfect backtest

An imaginary online tutorial claims an SMA caught every turning point perfectly. The lesson screenshot shows the final completed chart but never identifies which observations were visible at the time each claimed decision was made. The author may unintentionally be judging earlier periods using future closes. A responsible learner would reconstruct the SMA chronologically: after bar five the first five-point mean is available, and only after bar six closes is the updated mean known. If a signal is alleged during an unfinished bar, later revisions must be accounted for. Even an accurately recomputed series cannot establish executable buy/sell rates, liquidity or realistic costs. The scenario is fictional and not an allegation of dishonest conduct by any named analyst. It demonstrates the difference between hindsight explanation and evidence available at the moment.

Summary and recap — window size, data type, no guarantee

SMA is the equally weighted average of a fixed number N of selected observations. A five-bar mean uses five completed bars, then drops the oldest and adds the next one when the window moves forward. Fictional prices ₹100, ₹102, ₹101, ₹103 and ₹104 produce SMA5 ₹102; replacing ₹100 with ₹106 gives ₹103.20. The shortcut change equals (new − old)/N and provides a useful cross-check. A shorter SMA may react faster but neither the short nor long line predicts a guaranteed next move. Always specify bar duration, quote source, completion status, adjustments and units. The line is derived from earlier observations; it does not prove a trade occurred or that the strategy is appropriate for anyone.