OverlapIQ

LEVEL 2: READING PRICES · LESSON 9 OF 9 · 6 MIN

Trends, moving averages and a fair test

Moving averages are the most popular chart tool. Instead of taking claims on trust, we test one on real data.

THE 5-YEAR-OLD VERSION

The market zig-zags like a puppy on a walk, but the walk still goes somewhere. A trend is the direction of the walk. A moving average is a smooth line showing where the puppy has been on average over the last stretch, so you can see the walk without the zig-zags.

What a moving average is

A 50-day moving average is the average of the last 50 daily closes. Each day it drops the oldest and adds the newest, so it "moves". The 200-day is the same over 200 days. The longer the average, the smoother and slower it is.

8,00010,00012,00014,000Jul '19Jan '20Jul '20Jan '21
Nifty 5050-day average200-day average
Nifty 50 (blue) with its 50-day (orange) and 200-day (green) averages, mid-2019 to March 2021. Notice the averages turn only after price has already moved. Source: daily closing levels of the Nifty 50 price index (NSE Indices), as published by Yahoo Finance, 17 Sep 2007 to 13 Apr 2026. The price index leaves out dividends (roughly 1% to 1.5% a year), so total returns were somewhat higher. Past performance does not predict the future.

Moving averages lag: they describe the past and react late. Traders often read "price above the 200-day average" as an uptrend and "below" as a downtrend.

A fair test: the 200-day rule

Rule: hold the Nifty 50 only while it closes above its 200-day average (trading the next day). When it is below, sit in cash. Tested from July 2008 to April 2026:

124620102014201820222026
Always investedOnly when above the 200-day average
Growth of ₹1 from July 2008 on a log scale: always invested (blue) versus the 200-day rule (orange). No costs, no tax, and cash earns nothing in this test. Source: daily closing levels of the Nifty 50 price index (NSE Indices), as published by Yahoo Finance, 17 Sep 2007 to 13 Apr 2026. The price index leaves out dividends (roughly 1% to 1.5% a year), so total returns were somewhat higher. Past performance does not predict the future.
Always invested200-day rule
Total price return+512%+226%
Worst fall from a high-45%-23%
Time invested100%73%
Number of switches0144

What this test shows, and what it does not

The rule roughly halved the worst fall (-45% to -23%) but gave up more than half the total gain, and flipped 144 times (about 72 round trips), each of which would cost money and, outside a retirement account, trigger tax. It is one market, one rule, with the benefit of hindsight. Rules that look great on past data often fail on new data.

That is the real lesson of technical analysis: tools like this are trade-offs, not magic. They change the shape of your ride (smaller falls, smaller gains, more activity), which may suit one person and not another.

Remember

Check yourself

Three quick questions. Nobody sees your answers.

← PreviousCandlesticks: a whole day in one pictureNow practise →The replay lab: trade ₹10 lakh of virtual money through a real market

OverlapIQ Learn is for education only. It does not recommend buying, selling or holding any security, and nothing here is investment, tax or legal advice. The replay lab uses virtual money. Charts and figures use daily closing levels of the Nifty 50 price index (NSE Indices) as published by Yahoo Finance; the price index excludes dividends, and past performance does not predict the future.