Support, Resistance, Trends & Chart Patterns

1. What you'll learn and why it matters

Chapter 2 taught you to read one candle and small groups of candles. This chapter steps back and looks at the shape of a chart over weeks and months. These are the tools almost every swing trader and position trader uses to answer four practical questions:

1. Which way is the market going? (trend, Dow Theory, multi-timeframe analysis)
2. Where are buyers and sellers likely to show up? (support, resistance, role reversal, Fibonacci, 52-week levels)
3. Is a move real or a trap? (breakouts, volume confirmation, gaps)
4. Where do I get in, where do I get out, and is it worth it? (chart patterns, measured-move targets, the reward-to-risk ratio and Varsity's checklist)

Every example below uses real NSE prices from 2025–2026, computed with Python from the course data set (data/ohlc, daily OHLCV to 30 Sep 2026). You will see the tools work, and you will also see them fail. The failures teach more. No pattern in this chapter is a guarantee. They are ways of organising probabilities and placing stops sensibly.

A note on the prices you will see. Old prices are adjusted for later splits and bonus issues, the way most charting platforms show them. For example, KOTAKBANK split its shares 1:5 in Jan 2026, so the "440" resistance in Section 4 actually traded at about ₹2,200 in 2025. Likewise HDFCBANK (1:1 bonus, Aug 2025), NESTLEIND (Aug 2025), BAJFINANCE (Jun 2025) and HDFCAMC (Nov 2025) are shown on their post-adjustment scale. Patterns and percentages are unaffected, but old news articles and contract notes will quote the unadjusted rupee figures.

2. Trends: higher highs, higher lows

The plain-language idea

Prices do not move in straight lines. They zig-zag. A trend describes the direction of the zig-zag:

• Uptrend: each rally peaks above the previous peak (higher high, HH) and each dip stops above the previous dip (higher low, HL).
• Downtrend: lower highs (LH) and lower lows (LL).
• Sideways (range): highs and lows keep repeating at roughly the same levels.

The logic: why trends persist

Varsity lists "prices move in trends" as one of the core assumptions of technical analysis. The reasons are behavioural. Big investors (mutual funds, insurers, foreign institutional investors or FIIs) cannot build or sell a position in a day. They accumulate over weeks, and that keeps buying pressure on a stock. News also spreads gradually: analysts revise estimates one after another, and late investors chase what has already worked. A higher low means buyers stepped in earlier than last time, so they are getting more eager. A lower high means sellers are getting more eager. Structure is a direct record of who is gaining control.

How to identify swings objectively

The hard part is deciding which wiggles count. Two common methods:

1. Window method: a swing high is a bar whose high is the highest of the k bars on either side (e.g. k = 5 for short swings, 10–20 for bigger ones).
2. Percentage (zig-zag) method: a swing is only confirmed after price reverses by at least x% from the extreme (we use 6–8% for stocks and 3–4% for the Nifty in this chapter).

With a 1% tolerance, two highs or lows within 1% of each other count as "equal" (EH or EL). Equal levels are the raw material of double tops, double bottoms and support or resistance.

3. Trendlines and channels

What they are

A rising trendline joins two or more higher lows. A falling trendline joins two or more lower highs. A line parallel to the trendline through the opposite extreme forms a channel. Varsity points out that horizontal support and resistance cannot be drawn when a stock trades at all-time highs or keeps making new highs. In that situation a sloping trendline plays the support role.

The logic

In a steady uptrend, buyers are willing to pay a little more each week. The trendline is a picture of that rising "willingness to pay". When price falls back to the line, it is reaching the price at which buyers have recently stepped in. A close below the line says buyers are no longer paying up at the usual pace. That does not prove a reversal, but the trend has lost momentum.

The calculation

4. Support and resistance

Definitions

Support is a price zone below the current price where buying has repeatedly stopped declines. Resistance is a zone above the current price where selling has repeatedly stopped rallies. Varsity defines them relative to the current market price: the same horizontal line is resistance if price is below it and support if price is above it.

Why S&R works: three forces

1. Memory and regret. Investors who bought near a top and watched the price fall often sell when the price comes back to their purchase price ("let me just get my money back"). That supply caps rallies at old highs. People who missed buying at a low often place orders there if the price returns, which creates demand at old lows.
2. Supply and demand of large players. A fund that wants to accumulate 1 crore shares below ₹400 will absorb every dip to ₹400 for weeks. The chart sees a floor at 400. That is Dow's "accumulation" (Section 8).
3. Round numbers and self-fulfilment. Levels such as Nifty 22,000, 25,000 and 26,000, or a stock at ₹1,000, attract large option positions (open interest, Chapter 10), stop orders and media attention. Because many traders watch the same level, orders cluster there and the level "works" partly because people expect it to.

Varsity's summary: support and resistance show where a reversal is possible, not certain. History tends to repeat, but it is not obliged to.

Varsity's 4-step construction method

1. Load enough history. Varsity suggests 3–6 months for very short-term trades and 12–18 months for swing trades. Older levels are less relevant.
2. Find price-action zones. Mark at least three places where price either hesitated (stalled after a move up or down) or reversed sharply (a V-shaped turn).
3. Align them. The zones must sit at about the same price, and they should be well spaced in time. Varsity's example puts the second zone at least a couple of weeks after the first, and says the further apart the zones, the stronger the level. This chapter uses a stricter rule of about 4 weeks (20 sessions). Three touches over six months mean much more than three touches in one week.
4. Draw the line and widen it into a zone. Fit a horizontal line through the zones. If it is above the current price it is resistance; if below, it is support. Then allow a small margin either side. S&R is a band, not a single price.

Using S&R in a trade (Varsity's approach)

• Long trade: enter near support after a bullish signal. Put the stop just below the support zone (or the pattern low, if that is lower). Put the target just below the next resistance.
• Short trade: the mirror image. Enter near resistance, stop just above it, target just above the next support. Reminder from Chapter 2: in India you cannot carry a short in the cash market overnight, so a swing short needs stock futures or options (Chapters 9–10). Until then, read every "short" setup in this chapter as a signal to exit or avoid a long.
• Varsity's final checklist adds a sanity test: the pattern's stop and the nearest S&R level should be close to each other, within about 4%. If the nearest support is far below your logical stop, the setup has no structural anchor, and Varsity suggests dropping it.

In your framework: framework.indicators.pivots(df) returns swing highs and lows, and framework.patterns.support_resistance(df) clusters them into zones using the same touch-count and spacing rules as above.

5. Role reversal: old ceilings become floors

When resistance is decisively broken, it often turns into support, and broken support often turns into resistance. Varsity calls this role reversal.

The logic: imagine a stock repeatedly fails at ₹2,480. Sellers there were happy to exit. When it finally breaks above ₹2,480 and runs, people who sold at 2,480 regret it and want back in "at the price I sold". People who missed the breakout want to buy the first dip. Both groups put buy orders around 2,480, so the old ceiling becomes a floor. On the downside, people who bought at a support that later broke want to sell at break-even when price returns to it, so old support becomes resistance.

6. Breakouts and false breakouts

A breakout is a close beyond a well-established resistance (upside) or support (downside). Varsity notes that ranges form when there is no strong trigger, such as quiet results or waiting for an announcement. They end when something changes the balance of buyers and sellers.

What separates a real breakout from a trap

Varsity's two key tests are volume (well above average, which shows institutional participation) and momentum (price keeps moving away from the level over the next 3–4 sessions instead of drifting back). Additional filters that many traders use:

• Close, not intraday: the breakout bar must close beyond the level.
• Margin: the close is beyond the level by at least about 1% (or a fraction of ATR, Chapter 4), not 0.1%.
• Context: breakouts in the direction of the higher-timeframe trend succeed more often.
• Room: there should be space to the next resistance, at least 1.5× your risk (Varsity's minimum RRR).

In your framework: framework.strategies.breakout_52w(df) and framework.strategies.donchian_trend(df) are rule-based breakout systems; Chapter 12 backtests them honestly with costs.

7. Volume: does the crowd agree?

What volume is

Volume is the number of shares that changed hands in a period. Every trade has a buyer and a seller, so 100 shares traded counts as volume 100, not 200. A price move on heavy volume means many participants, and probably large institutions ("smart money"), agreed with the move. A move on thin volume may just reflect a lack of sellers or buyers that day.

Varsity's volume-price table

Honest check: is the table a trading signal by itself?

We tested the table on every trading day for 35 large-cap stocks in the course data (Jan 2016 – Sep 2026, 81,524 stock-days), measuring the dividend-adjusted return over the next 10 sessions:

What this tells us. On ordinary days the four rows are almost identical, so the table is not a stand-alone daily signal. The useful part is in the big-move rows: heavy-volume down days were followed by much weaker returns (0.32%) than light-volume down days (1.94%). That matches Varsity's "price down + volume up = bearish". Volume matters most at decision points: breakouts, breakdowns and pattern completions. Use it as confirmation there, not as a signal on every candle. (The sample is today's large caps, which carries survivorship bias, so treat the absolute returns with caution.)

8. Dow Theory

Charles Dow (co-founder of the Wall Street Journal) wrote about market behaviour around 1900. His ideas, organised by later writers, are the foundation of most of this chapter. Varsity summarises them as the following tenets, rephrased here in eight points.

The tenets

1. The market discounts everything. Known information, and the crowd's expectations about what is still unknown, is reflected in price quickly.
2. There are three kinds of trend running at once:
   • Primary trend: the main direction, lasting from about a year to several years.
   • Secondary trend: a correction against the primary trend, lasting weeks to months (a pullback in a bull market, a rally in a bear market).
   • Minor trends: day-to-day noise, lasting days to a few weeks.
3. The primary trend moves in phases. Accumulation, markup (public participation) and distribution, followed by markdown. These are described below.
4. Indices must confirm each other. A trend is more trustworthy when related indices move together. Dow used the industrial and transport averages; in India you might compare the Nifty 50 with Nifty Bank or the broader indices.
5. Volume must confirm the trend. In an uptrend, volume should expand on rallies and shrink on dips.
6. A sideways range can replace a secondary trend. Instead of a pullback, the market can simply go flat for a while.
7. The closing price matters most. Intraday spikes can be noise; the close is the price the market agreed on at the end of the day.
8. A trend stays in force until it gives a clear signal that it has reversed. In practice that signal is the swing rule from Section 2: an uptrend is intact until price closes below the last higher low. Until then, assume continuation rather than trying to call the top.

The market cycle: four phases

1. Accumulation: after a long decline, informed investors quietly buy. Price goes sideways at low levels, and support forms.
2. Markup: the floor holds, the news improves, trend followers join and the price rises quickly.
3. Distribution: at high prices and peak optimism, informed investors sell to latecomers. Price stalls, and repeated failures create resistance.
4. Markdown: supply overwhelms demand and the price falls, often fast. Eventually it reaches levels where accumulation starts again.

Using Dow Theory as a trader

• Identify the primary trend on the weekly chart and trade mainly in its direction.
• Treat secondary trends as your entry opportunities. A pullback in a primary uptrend is where swing traders buy.
• Varsity's caution: if you take a bullish candlestick signal during a secondary downtrend inside a primary uptrend, it can work, but be quick. Use tighter targets and exit if the move stalls.
• Dow's patterns (double and triple tops and bottoms, ranges and flags) are covered next, with the other classic patterns.

9. Classic chart patterns and measured moves

Chart patterns are recurring shapes produced by the tug-of-war between buyers and sellers. They fall into two families:

• Reversal patterns, which mark the end of a trend: double and triple tops and bottoms, head & shoulders and its inverse, and the cup (rounding) base.
• Continuation patterns, which are pauses within a trend: triangles, rectangles (ranges), flags and pennants. Triangles and rectangles can break either way, so treat the direction as unknown until the breakout.

This section has a real example of every pattern in the gallery below, and each one ends with the same trade card: entry, stop, risk, target, reward-to-risk and what actually happened. First the rules that produce those numbers.

The measured-move rule

The logic: the height of a pattern measures how far the two sides were prepared to push each other before one side gave up. Once one side gives up, the move often travels at least that far again. Treat this as a rule of thumb that is often roughly right, not a law.

Entry, stop and target: one recipe for every pattern

Every pattern has a trigger line: the neckline, boundary or flag edge whose break completes it. The table gives the trigger, the stop, the height and the target for each pattern in this chapter. "Beyond" means above for an upward break and below for a downward one.

Why tops and bottoms need the halfway stop. The "safe" stop for a double bottom sits under the lows, a full \(H\) below the neckline. The target is only \(H\) above the neckline, and you enter a little above it. So the risk is always bigger than the reward, and the RRR is always below 1. Every double and triple top or bottom in this section shows it. A close back halfway into the pattern already tells you the breakout has failed, so that is where this chapter puts the working stop.

Plan the trade before you enter, in this order:

1. Draw the trigger line and measure \(H\) as in the table.
2. Wait for a close beyond the line. A close less than 1% beyond it is weak evidence. The safer filters from Section 6 are a close at least 1% beyond the line, or two closes in a row beyond it. Volume above its 10-day average (Section 7) adds weight. The trade cards below use that day's close as the entry. If you do your analysis after the close (the EOD workflow in Section 14), you would enter at the next day's open, at a slightly different price.
3. Set the stop from the table and work out the risk per share: |entry − stop|.
4. Set the target = trigger ± \(H\), and compare it with the next support or resistance level. Use whichever comes first as your first target (the checklist in Section 14 does the same).
5. Compute the RRR = reward ÷ risk. If it is below your minimum (Varsity's is 1.5), skip the trade. Never move the stop closer than the table allows just to pass this test. A stop belongs where the pattern is proven wrong.
6. Size the position from the risk: shares = (capital × risk %) ÷ risk per share. With ₹5,00,000 and 1% risk, a ₹50 risk per share allows 100 shares. Chapter 11 covers this in depth, and framework.risk.position_size(capital, risk_pct, entry, stop) does the arithmetic.
7. Manage the trade. The stop only ever moves in your favour. Many traders book part of the position (say half) at the first support or resistance level, or at +1R (a gain equal to the risk), and trail a stop on the rest. Expect throwbacks: price often comes back to retest the broken line, and the stop in the table is placed to survive them. When the target is hit, exit or trail tightly. Reaching the target tells you nothing about what happens next.
8. Remember gaps. A stop is an instruction, not a guarantee. If the stock opens beyond your stop, you are filled at the open, and the loss can be much bigger than planned (the HDFCBANK rectangle below). Size positions so that a gap through the stop hurts but does not cripple you.

Two more rules from experience. Bigger patterns matter more: a weekly pattern built over 6 months outweighs a daily one built over 6 days. And bearish patterns in the cash market are mostly a signal to exit or avoid a long. A retail swing short needs stock or index futures (Chapters 9–10). The short "trade cards" below assume futures.

9.1 Double and triple tops and bottoms

A double top means price tests the same high twice and fails. Varsity requires the tops to be at least about two weeks apart; small differences in height are fine. A triple top means three failures, which Varsity regards as stronger. The pattern is confirmed only when price closes below the neckline, the low between the tops. Bottoms are the mirror image.

9.2 Head & shoulders (and the inverse)

A head & shoulders top has three peaks: a left shoulder, a higher head and a lower right shoulder. The neckline joins the two troughs between them and can slope. Varsity's Innerworth module reads it as a map of crowd psychology. The left shoulder is the last confident rally. At the head, enthusiasm peaks and then fails. The weak right shoulder shows buyers are tiring. The neckline break is where hope turns to fear. Volume is ideally heaviest at the left shoulder and head and lighter on the right shoulder. Treat the pattern as unconfirmed until the neckline breaks. An "almost head & shoulders" proves nothing. The inverse head & shoulders is the same shape upside down, at the end of a downtrend.

Because the right shoulder is lower than the head, a stop above the right shoulder is closer than the head, and it usually gives a workable RRR. That is the stop used in the trade cards below.

9.3 Triangles, rectangles, flags and pennants

• Ascending triangle: flat resistance with rising lows. Buyers keep paying more while sellers defend one price, which is usually (not always) resolved upward.
• Descending triangle: flat support with falling highs. The mirror image, with a bearish bias.
• Symmetrical triangle: lower highs and higher lows converging. Neither side dominates, so wait for the break.
• Rectangle (range): flat top and flat bottom (KOTAKBANK in Section 4, HDFCBANK in Section 7). Target = range width.
• Flag / pennant: a sharp, almost vertical move (the pole) followed by a short, tight, low-volume pause. In a flag the pause is a small parallelogram, often sloping slightly against the pole. In a pennant it is a small symmetrical triangle. Varsity describes the pole as a sustained, almost vertical rally and the pause as lasting about 5–15 sessions. Retail traders book profits during the pause while institutions hold, and the trend often resumes abruptly. Target = pole length added to (or, for a bear flag, subtracted from) the breakout.

9.4 Cup & handle (rounding base)

A cup is a rounded, U-shaped base. Sellers are gradually exhausted, then buyers gradually return. A handle is a small, shallow pullback near the right rim that shakes out the last weak holders before a breakout. It should stay in the upper half of the cup and usually lasts one to four weeks. Cups that are too deep (say more than about 35%) or V-shaped are less reliable. Target = cup depth added to the rim. The handle also matters for risk: its low gives a close, logical stop.

In your framework: framework.patterns.detect_candles(df) handles the candle patterns from Chapter 2; framework.risk.r_multiple and framework.risk.position_size(capital, risk_pct, entry, stop) turn any pattern's entry and stop into a position size. Every trade card in this section is rebuilt by course/scripts/ch03_pattern_trades.py.

10. Gaps

A gap is a price range where no trading happened between two consecutive sessions. Varsity's candlestick chapter illustrates the cause: a stock closes at ₹100, strong results come out overnight, and buyers are willing to pay ₹104 from the first trade, so nothing trades between 100 and 104. In India gaps form in the 9:00–9:15 pre-open auction.

The four types (plus the island)

11. Fibonacci retracements

The numbers

The Fibonacci sequence (known in ancient Indian mathematics, popularised in Europe by Leonardo of Pisa) starts 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144 … and each number is the sum of the two before it. Ratios between its terms converge to a few constants:

Traders add 50% (not a Fibonacci ratio, but a natural "half-back" level) and often 78.6% (the square root of 0.618).

How to draw and compute

1. Pick the most recent clear swing: trough to peak for an uptrend, peak to trough for a downtrend.
2. Compute the retracement levels:

Why it might work

Varsity is candid: nobody has proved why these levels should matter. The most plausible reason is self-fulfilment. Millions of traders and algorithms watch the same levels, so orders cluster there. The levels also land where a healthy pullback tends to end anyway (a third to two-thirds of the prior move). Varsity's advice is to use Fibonacci as confirmation for a trade that already passes the other checklist items, not as a reason on its own.

12. 52-week highs and lows

The 52-week high and low are the highest and lowest prices of the past year (about 252 trading sessions). Exchanges, brokers and media publish lists of stocks making new 52-week highs and lows every day, which makes these levels very widely watched.

Why they matter

• At a new 52-week high there is no overhead supply. Everyone who bought in the past year is in profit, so there is no "get me out at break-even" crowd. That is the bullish argument behind momentum strategies.
• Anchoring works the other way: many investors feel a stock is "too expensive" near its high and hesitate. When good news finally pushes it through, the late buyers can fuel a move.
• New 52-week lows show that every buyer of the last year is underwater. That supply caps rallies, which is why "buying because it's cheap" often catches a falling knife (ITC in 2026 kept making new lows: 403 → 262.75).

Honest reading. In this small, survivorship-biased large-cap sample, buying the first new 52-week-high close had no meaningful edge over buying on any random day. It was slightly better over 20 sessions and slightly worse over 60. New lows lagged over 20 sessions but caught up later. Large academic studies on broader universes do find a modest momentum effect, but this data says a 52-week high is context, not a signal. It is most useful as one item on the checklist: trade breakouts to new highs that also have a base, volume, an acceptable RRR and a supportive market trend. Chapter 12 tests a complete 52-week breakout system with costs.

13. Multi-timeframe analysis: weekly trend, daily entry

Varsity's TradingView chapter recommends viewing several timeframes side by side. For a swing or positional trader the standard pairing is:

• Weekly chart = direction. What is the primary trend (Dow)? Are weekly highs and lows rising? Is price above a rising long average (e.g. 30 weeks ≈ 150 days; moving averages are covered in Chapter 4)?
• Daily chart = timing. Wait for a pullback (secondary trend) to stop, ideally at support, a Fibonacci level or a rising trendline. Then enter on a sign that buyers are back: a bullish candle pattern, or a close above the previous few days' highs.

The logic: the weekly chart sets the odds and the daily chart sets the risk. Buying daily dips in a weekly uptrend means you trade with the institutional flow, and the dip gives you a nearby, well-defined stop.

14. The technical-analysis checklist (Varsity, adapted)

Varsity ends its technical-analysis module with a checklist that combines everything: candlesticks (Chapter 2), S&R, volume, Dow Theory, indicators (Chapter 4) and RRR. Here it is adapted for a swing or positional trader who also uses the material in this chapter. Varsity's scouting routine is to shortlist from a liquid universe (it suggests Nifty 50 stocks for beginners), look at the last 3–4 candles for a recognisable pattern, then spend 15–20 minutes per shortlisted chart on the full list. It also says that deciding not to trade is itself a trading decision. Many days will give you nothing, and that is fine.

• Liquidity: liquid stock (tight bid-ask, several lakh shares a day; Nifty 50/100 names to start). Avoid operator-driven stocks.
• Higher-timeframe trend (Dow): weekly HH/HL for longs (LH/LL for shorts). Know whether you are trading with the primary trend or a secondary one.
• Setup: a recognisable candlestick pattern in the last 3–4 sessions or a confirmed chart pattern (neckline or boundary close).
• Prior trend fits the pattern: bullish reversal patterns need a preceding decline; bearish ones need a preceding rise.
• Volume: signal-day volume at or above the 10-day average; breakouts ideally well above it.
• S&R alignment: for a long, the pattern low/stop sits at or just below support (for a short, the high/stop at or just above resistance). Varsity's limit: S&R within ~4% of the pattern stop, otherwise drop it.
• Room to run: the next resistance (for a long) is far enough away. Check old highs, round numbers, gaps and 52-week levels.
• Confluence (optional bonus): a Fibonacci level, trendline or role-reversal zone agrees with your entry or stop.
• RRR: at least 1.5 for swing trades, measured to the first realistic target (S&R or measured move, whichever is nearer).
• Indicators (Chapter 4): if RSI or MACD agree, you may size normally or slightly larger; if they disagree, size down or skip. Indicators confirm; they do not originate trades.
• Event check: results, budget, policy or index-rebalance dates inside your holding period mean gap risk (Section 10).
• Position size: risk per trade ≤ 1% of capital, computed from entry − stop (Chapter 11).
• Execution discipline: place the stop as soon as you enter; do nothing until target or stop, except trail the stop or book a partial profit you planned in advance; record the trade in your journal.

15. Try it yourself

These snippets need a copy of the course repository and Python with pandas and numpy (setup steps are in its framework/README.md). Run them from the repository's root folder, for example with .venv/bin/python, because the file paths are relative to that folder. Each reproduces a calculation from this chapter on the course data. The full figure script is course/scripts/ch03_figures.py; it prints every number quoted above.

(a) Find swing highs and lows and test the trend

import pandas as pd
df = pd.read_csv("data/ohlc/TITAN.csv",
                 index_col=0, parse_dates=True).loc["2025-03-01":]
k = 10                                   # a swing high = highest high of the 10 bars either side
hi = df.High[(df.High == df.High.rolling(2*k+1, center=True).max())]
lo = df.Low[(df.Low == df.Low.rolling(2*k+1, center=True).min())]
print("Swing highs:\n", hi.round(1).tail(6))
print("Swing lows:\n", lo.round(1).tail(6))
print("Last two highs rising?", hi.iloc[-1] > hi.iloc[-2], "| last two lows rising?", lo.iloc[-1] > lo.iloc[-2])

With k = 10 this finds a minor swing low at 3,985.1 (13 May 2026) just above the 3,962.8 low (11 Jun), so "lows rising" prints False even though the bigger structure is up. Try k = 5 and k = 20 and see how the answer depends on swing size. That is why you fix the parameter before you look.

(b) Varsity's S&R test: enough well-spaced touches?

import pandas as pd
df = pd.read_csv("data/ohlc/KOTAKBANK.csv",
                 index_col=0, parse_dates=True).loc["2025-03-01":"2026-02-27"]   # ~12 months
level, tol = 446.0, 0.015                     # candidate resistance and a +/-1.5% zone
k = 6
swing_hi = df.High[df.High == df.High.rolling(2*k+1, center=True).max()]
touches = swing_hi[(swing_hi - level).abs() <= level * tol]
spaced = [touches.index[0]]
for t in touches.index[1:]:
    if (t - spaced[-1]).days >= 28:          # this chapter's spacing rule (~4 weeks)
        spaced.append(t)
print(touches.round(1))
print("Well-spaced touches:", len(spaced), "->", "valid S/R" if len(spaced) >= 3 else "not yet")

(c) Volume ratio and Varsity's volume-price table

import pandas as pd
df = pd.read_csv("data/ohlc/HDFCAMC.csv",
                 index_col=0, parse_dates=True)
df["vol10"] = df.Volume.rolling(10).mean().shift(1)     # average of the PREVIOUS 10 sessions
df["vol_ratio"] = df.Volume / df.vol10
df["price"] = df.Close.diff().apply(lambda x: "up" if x > 0 else "down")
df["volume"] = (df.vol_ratio > 1).map({True: "high", False: "low"})
table = {("up", "high"): "bullish", ("up", "low"): "caution (weak rally)",
         ("down", "high"): "bearish", ("down", "low"): "caution (weak fall)"}
df["read"] = [table[(p, v)] for p, v in zip(df.price, df.volume)]
print(df.loc["2025-10-27":"2025-10-31", ["Close", "Volume", "vol10", "vol_ratio", "read"]].round(2))

Expected: 29 Oct 2025 shows a ratio of 4.11 and "bearish".

(d) Fibonacci levels

def fib_levels(low, high, trend="up"):
    """Retracement levels of the move low->high (trend='up') or high->low (trend='down')."""
    move = high - low
    out = {}
    for r in (0.236, 0.382, 0.5, 0.618, 0.786):
        out[f"{r:.1%}"] = round(high - r * move, 1) if trend == "up" else round(low + r * move, 1)
    return out

print("Nifty fall 5 Jan -> 2 Apr 2026:", fib_levels(22182.55, 26373.2, trend="down"))
print("TITAN rally 11 Jun -> 27 Aug 2026:", fib_levels(3962.8, 5186.7, trend="up"))

(e) Measured-move target, RRR, and a 52-week-high scan

import os, pandas as pd
def measured_target(pattern_high, pattern_low, breakout_level, direction):
    """Minimum target = breakout level +/- pattern height."""
    h = pattern_high - pattern_low
    return breakout_level + h if direction == "up" else breakout_level - h

def rrr(entry, stop, target):
    return (target - entry) / (entry - stop)

print("HDFCAMC triple top target:", measured_target(2967.25, 2715.25, 2715.25, "down"))   # 2463.25
print("DIXON triple bottom target:", measured_target(11847, 9600, 11847, "up"))           # 14094
print("TITAN Jun-2026 setup RRR:", round(rrr(4184.0, 3950.9, 4605.0), 2))                  # 1.81

D = "data/ohlc/"
rows = []
for f in sorted(os.listdir(D)):
    df = pd.read_csv(D + f, index_col=0, parse_dates=True)
    hi52 = df.High.iloc[-252:].max(); lo52 = df.Low.iloc[-252:].min()
    rows.append((f[:-4], df.Close.iloc[-1], round((df.Close.iloc[-1] / hi52 - 1) * 100, 1),
                 round((df.Close.iloc[-1] / lo52 - 1) * 100, 1)))
t = pd.DataFrame(rows, columns=["symbol", "close", "% from 52w high", "% above 52w low"])
print(t.sort_values("% from 52w high", ascending=False).head(5).to_string(index=False))

On the 30 Sep 2026 data, even the instrument closest to its high (KOTAKBANK) was 8% below it. No instrument in the universe was near a 52-week high at the cut-off, which is consistent with the weak index structure in Section 4.

In your framework: these snippets become framework.indicators.pivots, framework.patterns.support_resistance, framework.risk.position_size and framework.strategies.breakout_52w. python -m framework.scan will run the daily checklist across the universe (Chapter 17).

16. Key takeaways

17. Quiz and exercises

18. Sources

Sources: Zerodha Varsity, Module 2 (Technical Analysis): Introducing Technical Analysis · Multiple Candlestick Patterns (Part 3), gap-up and gap-down openings · Support and Resistance · Volumes · Dow Theory (Part 1) · Dow Theory (Part 2), ranges, breakouts, flags, RRR and checklist · Fibonacci Retracements · The Final Lap / Helping you get started · Interesting features on TradingView (multi-timeframe layouts). Zerodha Varsity, Module 12 (Innerworth): Head-and-Shoulders Pattern is a Psychological Map · Classic Chart Patterns: Know How to Use Them. Market structure (verified Oct 2026): Zerodha Z-Connect, Closing Auction Session (effective 3 Aug 2026) · Kotak Neo, NSE pre-open session changes from 7 Sep 2026 · NSE market timings. News context: Dhan, Closing bell 3 Feb 2026 (India–US trade deal) · Kotak Securities, ITC falls on cigarette excise duty (Jan 2026). Data: NSE daily OHLCV via Yahoo Finance (course data set data/ohlc, to 30 Sep 2026); all figures and statistics computed by course/scripts/ch03_figures.py. Explanations are written independently; Varsity text is not reproduced.