Technical Analysis · Field Notebook · Page 2 of set
Chapters 3 and 4 — The Chart Types and Getting Started with Candlesticks. Why line and bar charts fall short, how a candlestick is built from OHLC, how timeframes change the picture, and the assumptions candlestick patterns lean on.
Notebook Summary · liftable, derivable from the notes below
| Bullish (blue) | Bearish (red) | |
|---|---|---|
| Rule | C > O | C < O |
| Body top | Close | Open |
| Body bottom | Open | Close |
| Upper shadow | high→close | high→open |
| Lower shadow | low→open | low→close |
We settled that OHLC is the right summary of a period. Now the problem is drawing it. A single trading day is four numbers; a year is roughly a thousand. A chart that shows one number at a time cannot hold that, so most familiar charts are out before we start.
Ten days of trading is forty data points (four per day). Six months is a wall of them. The regular charts you know — column, pie, area — each plot one value per point in time, so none can carry OHLC. Only three chart types are built for the job: the line, the bar, and the Japanese candlestick. This module lives on candlesticks, but you should see why the other two lose first.
The line chart plots one value — the close — and joins the dots. Sixty days of closes becomes one line. Its strength is its weakness: a single glance shows the trend, and nothing more. Open, high, and low are all discarded.
The bar chart is more honest — it carries all four prices in one vertical bar with two ticks:
Left tick below right tick means close is above open — a positive (bullish) day, drawn blue. Left tick above right tick means close is below open — a negative (bearish) day, drawn red. The central line's length is the day's range (high minus low): a long line is a wide day, a short line a quiet one.
So the bar carries everything — yet traders still avoid it. The reason is purely visual: the ticks are thin and easy to misread, and across dozens of bars a forming pattern hides in the clutter. All the data is there; the eye cannot use it fast enough.
Candlesticks come from 18th-century Japan, credited to a rice merchant named Homma Munehisa — likely the oldest form of price analysis anywhere. The West stayed unaware until the 1980s, when a trader named Steve Nison came across them and introduced them widely, in his book Japanese Candlestick Charting Techniques. Many patterns keep their original Japanese names, which is why the vocabulary ahead sounds the way it does.
A candlestick draws the same four numbers as a bar, but the open-to-close distance becomes a filled rectangular body instead of two ticks — and a solid block is far easier to read at a glance than a pair of stubs. It has three parts: the real body (open to close), the upper shadow (wick to the high), and the lower shadow (wick to the low). Colour still follows the convention: blue for bullish, red for bearish.
The single thing to internalise: which end of the body is the open flips with direction. On a bullish candle the close sits on top; on a bearish candle the open sits on top.
Body length is a signal in itself. A long body means strong, one-sided buying or selling; a short body means little happened and the two sides roughly cancelled.
A timeframe is simply the duration each candle represents. Change it and the same stock tells a different story, because the number of candles changes sharply.
| Frame | Open | Close | Candles / period |
|---|---|---|---|
| Monthly | 1st day's open | last day's close | 12 / year |
| Weekly | Monday open | Friday close | 52 / year |
| Daily (EOD) | day's open | day's close | ~252 / year |
| 30-min | 1st minute | 30th minute | ~12 / day |
| 15-min | 1st minute | 15th minute | 25 / day |
| 5-min | 1st minute | 5th minute | 75 / day |
As the frame shrinks, candles multiply — and not all of them are information. The rest is noise. A long-term investor reading a 5-minute chart sees mostly noise; an intraday scalper reading a monthly chart sees almost nothing actionable. The frame has to match the trade.
In one breath: line shows only trend, bar shows all four but reads slowly, the candlestick shows all four instantly — and the timeframe you pick decides how much of what you see is signal.
Candlestick patterns rest on the fourth assumption from Chapter 1, so it is worth sharpening it. Imagine a stock where three factors line up: (1) it has fallen four sessions straight, (2) today's fall is on lower volume, and (3) today's range is unusually small. The next day, the fall stops and the stock closes up.
Months later, the exact same three factors appear over five sessions. What do you expect on the sixth? The refined assumption answers it: when a specific set of factors that produced an outcome before repeats, expect the same outcome — provided the factors truly match. History does not repeat blindly; it repeats when the setup does.
Patterns form by grouping candles in a sequence — but sometimes a single candle is signal enough. So the whole vocabulary splits in two, and each pattern carries a built-in risk level: it tells you where to enter and where your stop-loss sits.
Before any pattern, three working rules. They will not be fully clear yet — keep them in the back of your mind; the module leans on them constantly.
In one breath: candlestick patterns work because matched factors reproduce outcomes — so buy strength, allow bounded flexibility, and only trust a pattern that has a prior trend to reverse.
They carry identical data. The candlestick just encodes it as a filled body your eye reads in a fraction of a second, where the bar's thin ticks take longer and blur across many bars. In a craft built on speed and pattern recognition, that readability difference is the entire advantage.
No — it gives you a busier one. More candles mean more short-lived moves, most of which are noise for a trade you intend to hold days. Accuracy comes from matching the frame to your holding period, not from zooming in.
Colours are cosmetic and configurable. Read the structure instead: is the close above the open (up candle) or below it (down candle)? Many Indian tickers default to green-up/red-down rather than the blue/red used here. The logic never changes.
Because these are reversal signals — they mark a possible turn. A bullish reversal needs a decline to reverse; without one, there is nothing for it to signal. A "bullish" shape inside an ongoing uptrend is just continuation noise, not a setup.
The mechanism is human, not mystical: traders react to the same conditions with the same greed and fear, so similar setups produce similar moves. The addendum matters — the outcome recurs only when the factors genuinely match, not on a vague resemblance.
It is relative, not absolute — compare a candle's body to the recent candles around it. A body clearly larger than its neighbours signals conviction; one clearly smaller signals hesitation. This is exactly where "quantify and verify" applies: pick a rule (say, body > the last N bodies) rather than eyeballing it.
Two only. Both run in your head or a scratch file — no software, no invented tools.
The module's own plotting drill. For each day, decide the colour, then say where the body's top and bottom sit and how far each shadow reaches.
Day 1 O 430 H 444 L 425 C 438
Day 2 O 445 H 455 L 438 C 450
Day 3 O 445 H 455 L 430 C 437Step 1 — colour first. Compare C to O. Day 1: 438 > 430 → bullish. Day 2: 450 > 445 → bullish. Day 3: 437 < 445 → bearish.
Step 2 — body ends. The body always runs open↔close; direction decides which is on top. Bullish → close on top, open on bottom. Bearish → open on top, close on bottom.
Step 3 — shadows. Upper shadow runs from the body's top up to H; lower shadow from the body's bottom down to L. On a bullish candle that means H→close (upper) and open→L (lower); on a bearish candle H→open (upper) and close→L (lower).
Day 1 BULLISH (438>430)
body top = C 438 body bottom = O 430
upper shadow= 438 -> 444 (len 6)
lower shadow= 430 -> 425 (len 5)
Day 2 BULLISH (450>445)
body top = C 450 body bottom = O 445
upper shadow= 450 -> 455 (len 5)
lower shadow= 445 -> 438 (len 7)
Day 3 BEARISH (437<445)
body top = O 445 body bottom = C 437
upper shadow= 445 -> 455 (len 10) <- long upper wick
lower shadow= 437 -> 430 (len 7)A new trader summarises Chapters 3–4. Four statements are wrong. Find each and correct it.
A. "On a bearish candle the close sits at the top of the body."
B. "A line chart drops open, high and low, so it's the richest chart."
C. "I'll switch to the 5-minute chart to trade my week-long swings more accurately."
D. "This is a textbook bullish pattern in a strong uptrend, so I'll buy."A — wrong. Bearish means C < O, so the open is on top and the close is at the bottom. The close-on-top case is the bullish candle. Fix: "On a bearish candle the open sits at the top."
B — wrong. Dropping O, H and L makes the line the poorest chart, not the richest — it keeps only the close. Fix: the line shows trend alone; the candlestick is the rich one.
C — wrong. A 5-minute chart is noise for a week-long swing; it does not add accuracy. Fix: match the frame to the hold — daily or weekly for multi-day swings.
D — wrong. A bullish (reversal) pattern needs a prior downtrend. Inside an uptrend there is nothing to reverse. Fix: require a prior downtrend before trusting a bullish pattern.
A wrong -> bearish: OPEN on top, close on bottom
B wrong -> line = poorest (close only); candle = richest
C wrong -> match frame to hold; 5-min != week-long swing
D wrong -> bullish pattern needs a PRIOR DOWNTREND