Technical Analysis: An Honest Introduction and Its Limits

Direct Equity — Arc 2, Article Four

The Claim

What technical analysis actually claims

The three articles before this one studied the business — its accounts, its returns, its competitors — to work out what a share is worth. Technical analysis studies something entirely different: the record of past prices and trading volumes. It rests on three claims. That prices move in trends which last long enough to be worth riding. That patterns repeat, because the human behaviour producing them repeats. And that everything worth knowing is already reflected in the price, so the price’s own behaviour is the only study you need.

The two approaches genuinely differ in kind. Fundamentals ask what is this business worth? Charts ask what is the crowd doing, and when is it changing its mind? Neither question is silly. The trouble starts when the second one is sold as having a reliable answer.

The Toolkit

The core tools, explained without the mystery

Trend. The plain observation that prices often keep moving in one direction for a stretch. Higher highs and higher lows make an uptrend, the opposite makes a downtrend, and neither makes a range. Riding trends is the oldest idea in charting and — worth saying clearly — the best supported by evidence. The tendency of recent winners to keep winning for a while, which researchers call momentum, is one of the few chart-based patterns that shows up consistently across markets and across decades, India included. Even that carries a warning: momentum strategies suffer occasional violent reversals that wipe out long stretches of gains in a few weeks.

Support and resistance. Price zones where falls have repeatedly paused (support) or rises have repeatedly stalled (resistance). There is nothing mystical about why they exist: memory and unfinished business. A lot of people bought at ₹500, and many of them have been waiting to get their money back, so a cluster of sell orders sits there. Support and resistance are real as zones where orders gather. They are unreliable as promises. Every support level holds until it doesn’t, and the moments it doesn’t are precisely the expensive ones.

Moving averages. The average closing price over the last 50 days, or 200, drawn as a line — a way of seeing the direction through the daily noise. Investors sometimes use “above or below the 200-day average” as a rough switch for how much risk to carry. The cost is built into the arithmetic and cannot be designed away: lag. An average of the last fifty days can only tell you about the last fifty days, so every signal it produces arrives after the move that produced it.

This is why crossover systems — the famous “golden cross” and its relatives — test respectably in decades that trend and bleed slowly to death in decades that don’t. In a sideways market they whipsaw you — repeatedly signalling a buy just before a fall and a sell just before a rise, paying brokerage, taxes and slippage (the gap between the price on screen and the price you actually get) each time, for no gain.

Volume. How many shares changed hands. A price move on heavy volume reflects broad participation; the same move on a trickle of volume reflects a thin order book and very few people. Volume’s most defensible use for a retail investor is a negative one: distrust dramatic price moves in small, thinly traded stocks. That is not a coincidence — it is exactly the signature of the promotion schemes covered later in this pillar.

RSI and other momentum indicators. Formulas that convert the speed of recent price moves into a reading, usually labelled “overbought” above 70 and “oversold” below 30. These are mean reversion tools (mean reversion is the assumption that a stretched price snaps back toward its average) — and they work until a genuine trend shows up, at which point “overbought” stays overbought for six months while the person following the indicator is busy selling a stock that keeps rising. Treat the RSI indicator as background information about how stretched a move is, never as an instruction.

Candlestick patterns and chart formations. Head-and-shoulders, flags, doji, hammer, and the rest of the vocabulary. This is the folklore layer, and individually these have the weakest evidence of anything in the toolkit — studies that try to define the patterns strictly enough to test them find results close to chance once costs are included. Learn to read a candle, which is genuinely useful literacy: each one just shows the open, high, low and close for a day. Hold the pattern predictions to the standard set out next.

The Evidence

What the research actually shows

This is the section the marketing leaves out, so let us be precise about it.

The idea	How well it holds up
Momentum — recent winners keep winning for a while	Real and persistent across markets and decades, with occasional sharp crashes that give back years of gains
Simple trend-following rules	Regime-dependent (they work in one kind of market and fail in another) — often profitable in trending decades, loss-making in sideways ones once costs are counted
Chart patterns and indicators	Largely indistinguishable from chance when tested on data they were not designed on

Why the winning examples are so convincing. Give anyone enough indicators and enough settings to adjust, and some combination will fit any past chart beautifully. That is not discovery. It is picking the rule from the history you are then testing it on — like being handed yesterday’s newspaper and correctly forecasting yesterday’s weather. Statisticians call it overfitting; the plain version is that hindsight has perfect vision. This single point explains why every chart course can show you spectacular past examples and why the live results so rarely match.

The sharper evidence is what happens to real traders. SEBI’s own studies of retail traders in derivatives — a group that is overwhelmingly short-term and overwhelmingly chart-driven — found the great majority losing money, persistently, year after year. Our article on futures and options presents those numbers in full. To be fair, that result bundles leverage and costs together with charting, so it is not a clean test of chart reading alone. But it demolishes the promise the trading-course industry actually makes: that chart skill converts into retail profits at scale. It does not.

There is a simpler argument too. If chart patterns reliably predicted prices, computer-driven funds — who can test every rule against every market’s full history in an afternoon — would have found and exhausted the easy ones long ago. What survives at that level is a narrow set of ideas, mostly momentum, executed with a cost discipline no individual with a broking app can match.

So why does it remain so popular? Four reasons, none of them conspiracy. Humans crave patterns — we see faces in clouds and shoulders in noise. Memory is selective: the calls that worked get retold, the ones that missed evaporate. Charts genuinely are useful for the narrow purposes below. And an entire ecosystem — brokers, platforms, course-sellers, influencers — earns more when you trade more, whatever your results.

The Honest Uses

What charts are genuinely good for

Strip away the prophecy and something useful remains. Naming it honestly matters as much as naming the limits.

1. Getting in and out sensibly. For a position you have already decided on through business analysis, a chart informs the mechanics — staging your entry over a few weeks rather than putting everything in at the top of a vertical run, or placing a standing GTT order at a level where orders visibly gather rather than at a round number you picked because it ends in zeros. The chart refines the when and how of a decision made on what and why.
2. Deciding in advance when you were wrong. Writing down “if this breaks decisively below X, my timing was wrong” turns vague worry into a rule you can follow, which is the whole subject of position sizing and risk management. Used this way, a level on a chart is a convention you have agreed with yourself. Its power comes from your discipline, not from the line.
3. Reading the temperature of the crowd. How many stocks are rising, how far prices have run from their averages, whether volume is frantic or dull — these give a quick sense of whether the market is calm or feverish. That is useful mainly for noticing when you are being swept along, not for calling tops.
4. Knowing what to walk away from. The one predictive use with a decent hit rate is negative. A small stock going vertical on thin volume is a “stay away” flag, and charts genuinely help you resist catching a falling knife in a business whose accounts have already broken.

What charts are not for, on the evidence: predicting prices, replacing business analysis, or turning a salaried person into a full-time trader on the strength of a fourteen-day indicator. And one boundary worth stating plainly in our own voice — any service selling chart-based “calls” is making investment recommendations, which requires SEBI registration. No registration means it belongs in the same category as the tips and finfluencer schemes covered later, however impressive the screenshots look.

Key Takeaways

---

Frequently Asked Questions

Your questions answered

Should a long-term investor learn technical analysis at all?

The reading layer, yes — about an afternoon’s worth. Understanding candles, spotting whether a stock is trending or drifting, and knowing why certain levels attract orders will protect you from both awe and panic when someone shows you a chart. The predicting layer earns your time only after the business-analysis articles are solid, and even then only in the narrow supporting role described above.

My friend consistently makes money from chart trading. Doesn’t that disprove the statistics?

One person’s results do not overturn the base rate — what happens to the whole population of people doing the same thing. In any large group of traders, some will be ahead over any given stretch through luck alone, and those are the stories that get told at weddings. If your friend has multi-year records, after all costs and taxes, that beat simply holding an index — genuinely rare — the interesting question is whether that is skill or a rising market in disguise. Only a proper bear market answers it.

Is fundamental analysis provably better, then?

It has the firmer anchor, because businesses have real earnings and cash flows, so a value estimate is about something. But the next article is honest that most professional fundamental managers also fail to beat the index. The pillar’s actual ranking: costs and behaviour matter more than which method you choose, process beats prediction, and the index remains the benchmark every method has to justify itself against.

What timeframe charts should a beginner look at?

Weekly and daily. Anything shorter is noise dressed up as information, and intraday charts pull people toward the trading frequency whose results the derivatives article documents. An investor’s questions about a chart — what has this done over three years, where do the obvious levels sit, what is volume saying about this move — are all answerable at daily resolution or slower.

Are paid chart courses and indicator subscriptions worth it?

The ideas in this article are the durable core, and they are free — in the exchanges’ own investor education material, in classic books available for a few hundred rupees, and in the documentation of the platform you already use. A ₹25,000 weekend course mostly adds branded vocabulary and survivor stories to the same material. The clearest way to think about it: the instructor’s edge is the fee, collected with certainty, from you.

Keep Learning