Skip to content
getcoinsight

Technical Analysis — Complete Guide

Exhaustive coverage of the discipline: from Dow theory to order flow, cycles, on-chain metrics and risk management.

1. Philosophy and Postulates

  • Dow's three tenets: (1) the price discounts everything; (2) prices move in trends; (3) history repeats itself (crowd psychology).
  • TA does not predict the future — it weighs probabilities based on repeatable behavior.
  • It all comes down to one skeleton: context → structure → levels → confluence → trigger → risk.

2. Chart Types

Types

  • Japanese candlesticks — OHLC, body + wicks; the standard in crypto.
  • OHLC bars — the same data drawn as ticks; line chart — close only, pure trend; Area — a line chart with a fill.
  • Heikin Ashi — smoothed candles from averaged OHLC; hides actual prices.
  • Renko — fixed-size bricks, time is ignored (a noise filter).
  • Point & Figure (X/O) — movement by price only; box size + reversal; targets via horizontal/vertical count.
  • Kagi — variable-thickness lines based on direction change; Range bars — a bar closes once a set range is covered.
  • The chart you choose changes what you see: for trend — Heikin Ashi/Renko, for precise levels — candles/P&F.

Signal methods

  • Heikin Ashi: candles with no lower wick = a strong uptrend (no upper wick = a downtrend); a color change + small bodies with wicks on both sides = weakening/reversal. Hold the position until the color changes.
  • Renko: a brick changing direction = a signal; a breakout confirmed by a brick is more reliable (noise filtered out). Brick size = sensitivity.
  • Point & Figure: signals — a breakout of an X column (buy) / O column (sell); targets are counted via vertical (from the reversal base) and horizontal (from the base width) count.

3. Market Structure

Swing points and trend

  • Swing high/low — a local extreme (fractal); a trend = a sequence of these points.
  • Uptrend: HH/HL; downtrend: LH/LL; range: horizontal boundaries.
  • A trend is alive as long as the sequence holds; a break in the sequence = a signal.

Structure break events

  • BOS (Break of Structure) — a break in the direction of the trend (continuation).
  • CHoCH (Change of Character) — the first break against the trend (an early reversal).
  • MSB / MSS (Market Structure Shift) — a confirmed change of direction.
  • Internal vs swing structure — minor structure inside the major one; don't confuse the scales.

Ranges

  • Boundaries = resistance/support; a deviation/sweep beyond a boundary + a return = a trap (often the best entries).
  • A range is traded from its boundaries to the opposite one; a breakout is traded only with volume confirmation.

Market cycles

  • Accumulation → markup → distribution → markdown.
  • Psychological cycle: disbelief → hope → optimism → euphoria → complacency → anxiety → denial → panic → capitulation → despair.

4. Theories and Methodologies

Dow Theory

  • 6 tenets: the market discounts everything; three types of trends (primary/secondary/minor); a trend has 3 phases (accumulation/public participation/distribution); confirmation by indices; volume confirms the trend; a trend stays in force until a clear reversal signal.

Elliott Waves

  • Impulse: 5 waves (1/3/5 motive, 2/4 corrective). Correction: 3 waves A-B-C.
  • Inviolable rules: wave 2 does not retrace beyond the start of wave 1; wave 3 is not the shortest (often the longest); wave 4 does not overlap the price territory of wave 1.
  • Fibonacci ratios: wave 2 ≈ 0.5–0.618 of 1; wave 3 ≈ 1.618 of 1; wave 4 ≈ 0.382 of 3; wave 5 ≈ 1 or 0.618 of 1.
  • Corrective patterns: zigzag (5-3-5), flat (3-3-5), triangle, combination; diagonals (leading/ending).
  • Weakness: subjectivity of the count — different analysts see different waves.

Wyckoff Method

  • Accumulation (phases A–E): PS → SC (selling climax) → AR (automatic rally) → ST (secondary test) → Spring (false break down) → Test → SOS (sign of strength) → LPS (last point of support) → BU.
  • Distribution: PSY → BC (buying climax) → AR → ST → UTAD (false break up) → SOW → LPSY.
  • 3 laws: supply/demand; cause/effect (accumulation → the size of the move); effort/result (volume vs result).
  • Composite Man — a hypothetical operator who accumulates/distributes.

Smart Money Concepts / ICT

  • Order Block (OB) — the last opposite candle before an impulse (a zone of institutional orders).
  • FVG / imbalance — a gap between the wicks of three candles; price tends to fill it.
  • Liquidity (BSL/SSL) — stops above highs / below lows; a liquidity grab / stop hunt before a reversal.
  • Breaker / Mitigation block — a spent OB acting as an opposite level.
  • Premium / Discount / Equilibrium — sell in premium (the top), buy in discount (the bottom) of the range.
  • ICT deeper: kill zones — London/New York open windows with a higher probability of a move; Judas swing — a false impulse right at the session open that traps early traders; OTE (0.62–0.79 Fib) — the 'optimal trade entry' zone within a retracement; SMT divergence — a divergence between BTC and ETH (or another pair) at the same level, an early sign of weakness; displacement — a sharp impulsive move that leaves an FVG behind it; liquidity void / BPR — a zone with no trading activity that price tends to return to.

5. Levels and Tools

Support / Resistance

  • How to draw them: by bodies vs by wicks (bodies = 'agreement', wicks = extremes); these are zones, not lines.
  • Level strength = number of touches × volume at them × freshness × timeframe.
  • Supply/demand zones — where an impulse move started (the base before the candle).
  • Flip — broken resistance becomes support (and vice versa).

Pivot Points

  • Standard (P, R1–R3, S1–S3), Fibonacci, Camarilla (narrow, intraday), Woodie (weights the close), DeMark (based on open/close).
  • Daily pivots — reference points for intraday.

Fibonacci

  • Retracement: 0.236 / 0.382 / 0.5 / 0.618 / 0.786; the 'golden pocket' 0.618–0.65 — the most important entry zone in a trend correction.
  • Extension/projection: 1.272 / 1.414 / 1.618 / 2.618 / 4.236 — targets.
  • Fan / Arcs / Time Zones / Channel; draw from a significant swing to a swing in the direction of the trend.

Gann and median lines

  • Gann Angles / Fan (1×1…), Square of 9, Gann Box.
  • Andrews' Pitchfork — a median line + parallels (price gravitates to the median); Schiff / Modified Schiff.
  • Fibonacci confluence / clusters — an overlap of several Fibs = a strong zone.

Trendlines and channels

  • Construction: at least 2 points, the 3rd touch confirms; in an uptrend — along the lows, in a downtrend — along the highs; by bodies (conservative) vs by wicks (aggressive).
  • Validity: the more touches and the shallower the angle, the more reliable; a line that is too steep breaks quickly.
  • Channel: a parallel to the trendline along the opposite extremes — gives entry zones (from a boundary) and targets (the opposite boundary).
  • Fan principle: after the first line breaks, a second, shallower one is drawn; three consecutive fan breaks = a trend reversal.
  • Break vs throwback: after a break price often retests the line from the inside — a more reliable entry point than the break itself.

6. Chart Patterns

How to read any pattern

  • Volume: should decline during formation and spike on the breakout (otherwise a fakeout).
  • Measured move (target): a projection of the pattern's height from the breakout point (e.g. H&S = head height from the neckline).
  • Breakout vs throwback: price often returns to retest the broken level — a second, more reliable entry point.
  • Failure: a breakout without volume / a quick return into the pattern = a trap; reversal patterns on thin volume are unreliable.
  • Statistics (Bulkowski): no pattern is 100%, the most reliable are ~70–80% under clear conditions — weigh a pattern, don't 'believe' in it.
  • Compare the pattern across several coins at once in Compare — it shows whether this is a single asset's setup or a move across the whole sector.

Reversal

  • Head and shoulders + inverse (the neckline as the trigger).
  • Double / triple top / bottom, rounded (saucer), Cup with handle.
  • V-spike, Diamond, Island reversal, Bump and Run.
  • Quasimodo (QM), Adam & Eve, Wolfe Waves (the EPA target line), Pipe/Horn, Dead Cat Bounce.

Continuation

  • Flag, Pennant; triangles: ascending (bullish) / descending (bearish) / symmetrical (neutral).
  • Rectangle, Channels, Measured Move, Scallop, Running flag, Dragon.
  • Three Rising/Falling Valleys (Peaks).

Bilateral / harmonic

  • Wedge (rising/falling wedge), Broadening formation / megaphone.
  • Harmonics — precise Fibo patterns across 5 points X-A-B-C-D, where each type is defined by its own set of Fibonacci ratios: AB=CD, Gartley, Bat/Alternate Bat, Butterfly, Crab/Deep Crab, Cypher, Shark, Three Drives, 5-0, NenStar, White/Black Swan, Leonardo, anti-patterns.
  • Entry — at point D (PRZ, potential reversal zone).

7. Candlestick Patterns

Reading principles

  • Location: a candlestick signal only works at S/R, Fib, OB, range boundaries; in the middle of nowhere = noise.
  • Confirmation: the next candle must confirm (a close beyond the pattern).
  • Size/volume: a large body + volume = a stronger signal; long wicks = a rejection of a level.
  • Reliability: three-candle patterns (morning/evening star) are more reliable than single dojis.

Single

  • Doji (long-legged, gravestone, dragonfly).
  • Hammer / hanging man, Inverted hammer / shooting star.
  • Spinning top, Marubozu, Belt Hold, High Wave, Rickshaw Man, Takuri.

Double

  • Engulfing (the strongest two-candle reversal at a level), Harami + harami cross.
  • Piercing line / Dark cloud cover, Tweezers, Kicking.
  • On-neck / In-neck / Thrusting — weak bullish continuation patterns in a downtrend (the second candle barely pulls back from the first candle's low); Matching low/high — two closes at the same level, a possible equilibrium zone; Homing Pigeon / Separating — rare trend-continuation patterns; Last engulfing / Hawk — reversal variants of engulfing near the end of a move.

Triple

  • Morning / evening star (+ doji-star), Abandoned baby.
  • Three white soldiers / three black crows, Three inside/outside, Tri-Star.
  • Stick Sandwich, Deliberation, Advance Block, Two Crows, Tasuki Gap, Gap three methods, Three stars in the south.

Five candles

  • Rising/falling three methods, Mat Hold, Breakaway, Ladder Bottom/Top.

8. Gaps

  • Common gap — in a range, fills quickly.
  • Breakaway gap — at the start of a trend/breakout; often does not fill.
  • Runaway / Measuring gap — in the middle of a trend; target = a doubling of the move.
  • Exhaustion gap — at the end of a trend; exhaustion.
  • Gap fill — the tendency to return and close the gap.
  • CME Bitcoin gap — CME futures don't trade on weekends → a Fri-Sun gap that the market statistically fills (a common intraday target).

9. Indicators

  • Principle: at most 1–2 from each category; indicators confirm structure, they don't replace it. Almost all of them lag.

Trend

  • MA (SMA/EMA): direction + dynamic support/resistance; key ones 20/50/200; Golden/Death Cross (50×200). EMA is faster. Price above/below the MA = a trend filter.
  • MACD (12/26/9): signal cross, zero-line cross (trend change), histogram (momentum), divergences.
  • ADX/DMI: >25 a strong trend, <20 a range (strength only, not direction).
  • Ichimoku: TK-cross, price above/below the Kumo (the cloud = an S/R zone), Chikou confirmation; the future cloud = a forecast.
  • Supertrend, PSAR — trend trailing; Alligator+Gator (Bill Williams).

Oscillators / momentum

  • RSI (14): >70/<30, but more valuable are divergences and behavior within a trend (in a bull 40–50 holds, in a bear 50–60 is resistance), failure swings.
  • Stochastic — more sensitive, good in a range; StochRSI even more sensitive.
  • CCI, Williams %R, ROC — alternatives to RSI/Stochastic with different sensitivity; WaveTrend — a crypto-community favorite for its fast reaction to reversals; QQE — a smoothed RSI with its own signal line; TD Sequential (TD9) — a candle counter that signals likely trend exhaustion on the 9th candle.
  • Divergences: regular (reversal) vs hidden (trend continuation) — a key distinction.

Volatility

  • Bollinger Bands (20, 2σ): squeeze (a contraction = an impulse ahead), walking the bands (a strong trend), %B/Bandwidth.
  • ATR — stop size (1.5–2×ATR) and sizing; Keltner; TTM Squeeze (BB inside Keltner); Chandelier Exit (trailing).

Volume-based

  • Volume — confirmation; OBV (accumulation/distribution + divergences); VWAP + Anchored VWAP (an institutional reference, anchored to an event).
  • Volume Profile (POC/VAH/VAL); MFI ('volume-weighted RSI'); CMF, A/D, Force Index, Klinger, BW MFI.

Overlays / add-ons

  • ZigZag, Fractals, Pivot Points, Auto S/R, Auto Fib.

10. Volume and Order Flow

Volume analysis / VSA

  • Volume = a verifier: a breakout on volume is real, on thin volume — a fakeout.
  • VSA: a wide candle on huge volume with no progress = absorption/reversal; a narrow one on high volume = absorption; climax volume = exhaustion.
  • You don't have to hunt for anomalous volume manually across charts — the 'Relative volume' column in the screener sorts coins by exactly that.

Auction Market Theory / Market Profile

  • Initial Balance (IB) — the range of the first hour; Value Area (~70% of volume).
  • Day types: trend / normal / double distribution.
  • Single prints (a fast move, get filled), poor high/low (no tail, retested), naked POC (a magnet).

Order flow / microstructure

  • DOM / order book — densities, 'walls'; bid/ask spread.
  • Footprint — bid×ask volume at each price; Delta — aggressors; CVD (cumulative delta) + CVD divergences (price up, CVD not = weakness).
  • Absorption (liquidity absorbing aggression), imbalance, iceberg, spoofing, liquidation cascades (crypto-specific).

11. Cyclical and Time Analysis

Time cycles

  • Hurst cycles (cycles of varying length nested within one another) and the dominant cycle (MESA/Ehlers) — an attempt to find repeating periodicity in price; unreliable on low timeframes, more stable on higher ones.
  • Seasonality: 'Uptober' (historically strong October for BTC), thin weekend liquidity (sharper moves on the same volume), the Bartels test — a statistical check for whether a seasonal pattern is more than chance.
  • BTC's 4-year halving cycle — a time reference, not a guarantee; for a deeper look at the phases and how to read the cycle, see the FA Guide, 'Positioning in the market cycle'.
  • Gann time cycles — cyclicality based on calendar intervals; popular among followers but weakly verified statistically.

Trading sessions

  • Asian / London / New York sessions — liquidity and volatility differ; the London–New York overlap is historically the most active stretch of the day even for 24/7 crypto.
  • Kill zones (ICT) — narrow windows around session opens where institutional participants more often trigger a move; useful for narrowing when to expect a trigger, not for picking the trade's direction.
  • In practice: cycles and seasonality are probability context, not a standalone entry signal. Use them to raise confidence in a setup found through structure and levels, not as a substitute for it.

12. Statistical / Quant Layer

Statistical tools

  • Z-score — how far price is from the mean, in standard deviations; the basis for mean-reversion signals (on short timeframes crypto reverts to the mean far more often than it holds momentum).
  • Correlation / beta / relative strength (RS) — how much an asset moves together with BTC or the market; linear regression channels show price's deviation from trend.
  • Historical (realized) volatility vs implied (from options) — a gap between the two signals underpriced or overpriced market fear.

Strategy backtesting

  • Expectancy E = win% × avgWin − loss% × avgLoss — a strategy is worth pursuing only if E > 0 across a representative sample of trades, not 5–10 cases.
  • Sharpe, Sortino, profit factor, max drawdown — measure the quality of the equity curve, not just the final return; Monte Carlo simulations show the range of possible outcomes for the same strategy.
  • Pitfall: testing on the same data the strategy was fitted to (in-sample) inflates the result — split history into in-sample/out-of-sample (walk-forward), or the numbers will lie.
  • Our screener's Score column uses a different, pillar-based weighting (technical, momentum, patterns, structure, derivatives, on-chain, sentiment) computed automatically — not the Z-score/backtesting statistics on this page. Think of this section as tools for testing your OWN strategies, separate from that composite score.

13. Intermarket Analysis

Correlations and dominance

  • BTC ↔ alts / Nasdaq-S&P / DXY (inverse) / gold; risk-on / risk-off.
  • BTC.D, ETH/BTC, Total / Total2 / Total3, USDT.D (an inverse sentiment indicator).
  • Capital rotation: BTC → ETH → alts.
  • BTC.D and capital rotation — the same mechanism is broken down in more detail alongside the cycle phases in the FA Guide.

Market breadth

  • Altseason Index, % of coins above MA50/MA200, the number of coins at new highs vs lows.
  • They show how broad a rally is: a thin rally on BTC alone = weak; broad participation = a healthy bull.

14. Crypto Market Specifics

Basics

  • 24/7 (no gaps between sessions except CME; thin liquidity on weekends); long wicks take out stops.
  • Correlation with BTC: an alt can have a perfect setup and fall with Bitcoin.

Derivatives (reading)

  • Funding Rate: positive = longs pay (overheating); an extreme = long-squeeze risk; persistently negative while price rises = fuel for growth.
  • Open Interest: price↑ + OI↑ = a strong move on new money; price↑ + OI↓ = shorts closing (weaker).
  • Long/Short ratio, Liquidation Heatmap (liquidation magnets), Basis (contango/backwardation).

Options

  • IV / DVOL, skew (fear via puts), Max Pain (a magnet into expiry), put/call, gamma zones.

On-chain (as TA context)

  • MVRV / MVRV-Z, NUPL, SOPR, Realized Price.
  • Exchange Net Flows (outflow = bullish), Puell, HODL Waves, Hash Rate.
  • This is a compact snapshot for trade context — for the full breakdown with cohorts (LTH/STH), Thermocap, and forensics, see the FA Guide, 'On-chain metrics'.

Sentiment

15. Multi-Timeframe + Trading Styles

Multi-timeframe (top-down)

  • Higher TF (1W/1D) — direction and key levels; middle (4H/1H) — structure and zones; lower (15m/5m) — the trigger and the precise entry.
  • Three-screen rule: don't trade against the higher TF; enter on the lower TF trigger. TF alignment = the strongest setups.

Trading styles (the same framework, different TFs)

  • Scalping — seconds to minutes; TF 1m/tick + order flow; many trades, small R, fees and spread are critical.
  • Intraday — hours, closed by end of day; TF 5–15m, context 1–4H; sessions/VWAP.
  • Swing — days to weeks; TF 4H/1D, context 1W; classic patterns/Fibs.
  • Position — weeks to months; TF 1D/1W; macro/cycle + FA matter more.
  • Style determines stop size, the number of trades, and the role of each tool.

16. Confluence

  • One signal = nothing. Probability rises with each independent overlap at a point.
  • Overlap: level + Fib 0.618 + OB/FVG + RSI/CVD divergence + a round number + a volume spike + POC + the higher TF.
  • You're not looking for a 'holy grail' but for an accumulation of probabilities.

17. Workflow

  • 1. Context (the higher TF + intermarket/BTC).
  • 2. Structure (HH/HL? CHoCH?).
  • 3. Level map (S/R, supply/demand, Fib, POC, OB/FVG, gaps).
  • 4. Search for confluence.
  • 5. Wait for a trigger (a candle / breakout / order flow on the lower TF).
  • 6. Confirmation vs anticipation (a conscious choice).
  • 7. Trade plan before entry (entry, stop, target, R:R).
  • 8. Found a setup with this checklist? Set an alert at the trigger price — don't hold the level in your head.

18. Risk Management

  • Everything rests on this. The best analysis = 50–60% winning trades; profit comes from the math of risk.

Position sizing

  • Formula: size = (risk-$ per trade) / (distance to the stop). Risk-$ = % of account × account (usually 1–2%).
  • Size adapts to the stop, not the other way around. A wider stop → a smaller size.
  • Once you've calculated size, stop, and target, lock them in right away with an alert instead of keeping them in your head during the trade.

Stop, target, math

  • Stop behind structure (behind a level/swing/zone), not at a random distance; a buffer for wicks/volatility (ATR).
  • R:R at least 1:2 (at 40% wins it's still positive). Think in R-multiples, not in $.
  • Expectancy E = (win% × avgWin) − (loss% × avgLoss); must be >0. Losing streaks are inevitable — sizing must survive them (risk of ruin).

Position management

  • Move to breakeven after a 1R move; trailing (ATR/Chandelier/by structure).
  • Scaling in/out (partial entries/exits); take profit in parts at targets.
  • No averaging into losses against structure.

19. Pitfalls and Psychology

  • Indicator lag — structure leads, indicators confirm.
  • Chart overload — 10 indicators = conflicting signals = paralysis.
  • Curve-fitting — seeing patterns where there are none; 'drawing them in'.
  • FOMO and revenge trading — the biggest destroyer of accounts.
  • Illusion of control — any trade can be a loser; what matters is following the system and the sizing.
  • Discipline > forecast: a plan before entry, execution without emotion, a trade journal.
  • Formula: Context → structure → levels → confluence → trigger → risk. TA tells you when; for what — fundamental analysis.

Formula: Context → structure → levels → confluence → trigger → risk. Patterns, candlesticks, indicators, order flow, cycles and statistics are tools within this skeleton, not a replacement for it. TA tells you when; for what, use fundamental analysis.