Technical Analysis: VSA, Wyckoff & Volume

The FlowSense Approach to TA

Most retail TA fails because it treats indicators as signals rather than descriptions. RSI overbought isn't a sell signal — it's a description that says "this has been going up for a while." Whether you should sell depends on context the indicator can't see.

The TA philosophy embedded in this platform is closer to proprietary spread/volume analysis and Wyckoff method: read what the institutional money is doing by watching where it left footprints in price and volume. The A/M/D scoring, divergence detection, and phase classification you see across FlowSense are all expressions of this lineage. To see these concepts applied in real time, the Market Scanner ranks S&P 500 tickers by A/D phase, and the Methodology page documents the exact computations.

📊 proprietary spread/volume analysis

Volume Spread Analysis was developed by Tom Williams in the 1990s, building on the original work of Richard Wyckoff in the early 1900s. The core insight is simple but uncomfortable: price alone tells you almost nothing. A 1% up day on tiny volume means something completely different from a 1% up day on 5× average volume — but most retail charts treat them as identical.

VSA reads each bar as a transaction between two groups: the "smart money" (institutions, market makers, professional desks) and the "crowd" (retail traders, momentum chasers, news reactors). The smart money has information advantages, capital advantages, and patience advantages. The crowd has none of those. VSA tries to identify which group is the dominant actor on a given bar — and that tells you what's about to happen next.

The four foundational VSA setups:

No-supply bar. A down bar with a narrow range and below-average volume. The interpretation: the smart money isn't selling. If they were, you'd see wide range and high volume on the down bars. The narrow range tells you the selling pressure is exhausted. After a no-supply bar, the next direction is usually up. The FlowSense Top Accumulation screener finds tickers where multiple no-supply bars are clustering — a textbook accumulation signature.

No-demand bar. The opposite: an up bar with narrow range and below-average volume. The smart money isn't buying. The rally is being carried by retail enthusiasm or short covering, not institutional accumulation. After a no-demand bar, the rally typically stalls or reverses. These are the bars that catch FOMO buyers right before the drawdown.

Stopping volume. A wide-range down bar on extreme volume that closes in the upper half of its range. Reading: the selling was met by aggressive buying. The smart money used the panic to load up. After stopping volume, you typically see a sharp recovery within a few bars. Stopping volume is one of the highest-conviction VSA reversal signals.

Climactic action / buying climax. A wide-range up bar on extreme volume that closes in the lower half of its range. Reading: the rally was met by aggressive selling. The smart money used the euphoria to distribute. After a buying climax, you typically see a sharp decline. Climaxes are one of the highest-conviction VSA distribution signals.

What separates VSA from typical "volume confirmation" advice is the focus on the spread (the high-low range of the bar) relative to volume and relative to recent bars. Volume without spread context is meaningless. Wide spread on low volume is suspicious. Narrow spread on high volume is even more suspicious. Every FlowSense A/M/D score weighs both factors together — that's why the platform's accumulation calls hold up where simple "volume spike" alerts produce constant false positives.

For a practical introduction to VSA, see Volume Price Analysis by Anna Coulling — listed in Recommended Books.

🎯 The Wyckoff Method

Richard Wyckoff was a Wall Street operator in the early 1900s who studied the most successful traders of his era — including J.P. Morgan and Jesse Livermore — and reverse-engineered a framework that's still in use a century later. The Wyckoff method describes the four-phase cycle that every traded instrument goes through: Accumulation → Markup → Distribution → Markdown, then back to Accumulation.

Accumulation Phase. After a downtrend, large operators absorb the supply being dumped by capitulating retail holders. Price moves sideways in a range. Volume on down bars decreases over time (no-supply); volume on up bars increases (institutional buying). The range itself shows a characteristic structure: a "selling climax" creates the floor, an "automatic rally" defines the ceiling, then secondary tests of the floor on lower volume confirm accumulation is complete. The "Spring" is a final fake breakdown below the floor that's immediately reversed — designed to flush out the last weak hands before the markup begins.

Markup Phase. Once accumulation is complete, the large operators have a positioned book and now allow price to rise. Markup phases show higher-highs and higher-lows, expanding range on up bars, contracting range on down bars (pullbacks aren't selling, they're profit-taking). Most of the trend's gains are made in markup. The Markup phase is where momentum and trend-following strategies work best.

Distribution Phase. After the markup has run far enough, the operators start unloading. Distribution looks structurally similar to accumulation — sideways range, defined floor and ceiling — but the volume signature inverts. Up bars now have decreasing volume (no-demand); down bars have increasing volume (institutional selling). A "buying climax" creates the ceiling; an "upthrust" is the distribution equivalent of the spring — a final fake breakout above the ceiling that's immediately reversed to trap late longs.

Markdown Phase. Distribution complete, the operators allow price to fall. Lower-highs, lower-lows, expanding range on down bars, contracting range on up bars (bounces aren't real recovery, just short covering). The markdown phase is where shorts win and where stop-losses on long positions matter most.

The Composite A/M/D Phase classifier on FlowSense maps directly onto these Wyckoff phases. When the platform labels a ticker "Accumulation," it's identifying the structural signature Wyckoff described: range-bound price, narrowing down bars, expanding up bars, declining volume on selloffs, rising volume on rallies. The Composite Signal Engine aggregates these signals across multiple timeframes to filter out noise.

The key insight that makes Wyckoff durable: institutional traders cannot enter or exit large positions instantly. Their positions take days or weeks to build — and that building process leaves footprints. Wyckoff is the study of those footprints. To dig deeper, Trades About to Happen by David Weis and The Three Skills of Top Trading by Hank Pruden are the modern authoritative texts on Wyckoff method.

↕ Divergence Patterns

A divergence occurs when price moves one way and an indicator moves the opposite way. The premise: if price makes a new high but momentum or volume doesn't confirm, the move is losing strength internally even if it looks strong externally.

Bullish (positive) divergence: price makes a lower low, but the indicator makes a higher low. The downside momentum is fading even though the price is still falling. Bullish divergence often precedes meaningful reversals — though "often" is not "always." On the Divergence page, FlowSense filters divergence signals through volume and trend context to dramatically reduce false positives.

Bearish (negative) divergence: price makes a higher high, but the indicator makes a lower high. The upside momentum is fading. Buying climaxes in Wyckoff terms are often accompanied by bearish divergence — the rally to the new high happens on shrinking momentum and volume, the buying climax bar exhausts the move, and the distribution phase begins.

Hidden divergence is the less-discussed but more actionable variant. Hidden bullish: price makes a higher low while the indicator makes a lower low. This is a continuation signal in an uptrend — momentum reset, trend resuming. Hidden bearish: price makes a lower high while the indicator makes a higher high — continuation signal in a downtrend. Hidden divergences fire more often than regular divergences and have better win rates in trending markets.

Why most divergence trading fails: retail traders treat any divergence as a reversal signal. In reality, divergence in a strong trend can persist for many bars without reversal — the smart money simply absorbs all the supply or demand. Divergence becomes actionable only when paired with structural context: a key support/resistance level, a Wyckoff phase change, a volume climax, or convergent signals from multiple indicators. FlowSense's divergence filtering requires this multi-factor confirmation.

The indicators that produce the most reliable divergences (in our backtesting): On-Balance Volume (OBV), Chaikin Money Flow (CMF), Accumulation/Distribution Line, and Composite RSI. RSI alone is the most-quoted divergence source and one of the least reliable — too many false signals in trending markets.

🕯 Candlestick Patterns — The Ones That Actually Work

The candlestick pattern literature is enormous and most of it is garbage. Academic backtests have repeatedly shown that the vast majority of named candlestick patterns have no statistically significant edge after costs. A few do — and only a few — and even those work only in specific contexts.

Patterns with documented edge:

Hammer (and inverted hammer) at a meaningful support level after a downtrend. Long lower wick (lower wick at least 2× body length), small body at the top of the range, ideally on above-average volume. Reading: aggressive selling met aggressive buying — a stopping-volume bar in VSA terms. Win rate improves significantly when the hammer occurs at a previously-tested support level or after multiple no-supply bars.

Engulfing patterns at the end of a trend. A bullish engulfing has a body that completely engulfs the previous bar's body, ideally on above-average volume. Bearish engulfing inverts. The pattern works because it represents a definitive change in who's in control of the bar's range — the previous bar's high and low are both taken out.

Doji at extremes. A doji is any candle with open and close near each other (small or no body). At the extreme of a move (after a long advance or decline), a doji represents indecision after sustained one-way flow. The follow-through bar matters more than the doji itself — a doji that closes near its midpoint, followed by a strong bar in the opposite direction of the prior trend, is a reasonable reversal signal.

Patterns that are mostly noise: three-line strikes, abandoned baby, three black crows, three white soldiers, harami, harami cross, gravestone doji, dragonfly doji (in isolation). Backtests consistently show these have no edge — or what edge exists is too small to overcome slippage and commissions.

The deeper principle: candlestick patterns describe what already happened, not what will happen next. A hammer tells you sellers were absorbed at this level. Whether the absorption continues — whether the bottom holds — depends on context the candle can't see. The candle is one input. Volume is another. Trend structure is another. Pair them and patterns become useful; trade them in isolation and you'll churn your account on noise.

For a rigorous statistical treatment, see Evidence-Based Technical Analysis by David Aronson (referenced in Recommended Books) — it puts dozens of candlestick patterns through proper statistical testing and demolishes most of them.

📐 Chart Patterns — Volume Is Everything

Triangles, flags, pennants, head-and-shoulders, double tops/bottoms — these patterns are real and they can be tradeable, but the textbook treatment of them produces consistently mediocre results. Why? Because the textbooks describe the shape and skip the volume context.

A breakout from a symmetrical triangle is a real signal — IF volume expands meaningfully on the breakout bar AND the breakout closes meaningfully beyond the trendline. Without volume confirmation, breakouts fail at roughly the rate of a coin flip (Aronson's data). With volume confirmation, the win rate climbs into the 55-60% range — still not extraordinary but with significant edge over costs when combined with proper position sizing.

Head-and-shoulders is the most-discussed reversal pattern and one of the most overcalled. The actual pattern requires: a defined left shoulder peak with volume X, a higher head peak with volume LESS than X (a momentum divergence), a right shoulder peak with even less volume, a clear neckline, and a decisive close below the neckline on volume that exceeds the head's volume. Most "head and shoulders" patterns retail traders point out are missing two or three of these confirmations.

The FlowSense approach to chart patterns: don't trade the pattern, trade the structural context the pattern describes. A failed head-and-shoulders right shoulder that resolves upward is often a more powerful signal than the original pattern would have been — because everyone short the pattern is now wrong and has to cover.

🧰 Indicator Library — What Each One Actually Measures

Indicators are mathematical transformations of price and/or volume. Each transformation answers a specific question. Knowing what question an indicator answers is more important than knowing whether it's "overbought" or "oversold."

RSI (Relative Strength Index): Ratio of recent up-day magnitude to total recent magnitude. Range 0-100. Question answered: "How asymmetric has the recent move been?" RSI above 70 means recent up moves dominate; below 30 means down moves dominate. RSI is NOT a reversal indicator — in strong trends, RSI can stay above 70 for many bars while price continues higher. Useful for: divergence, comparing relative strength across stocks. Not useful for: bare overbought/oversold calls.

MACD (Moving Average Convergence Divergence): Difference between two EMAs (typically 12 and 26), with a 9-period EMA of that difference as the "signal line." Question answered: "Is short-term momentum accelerating or decelerating relative to medium-term momentum?" MACD crossovers above zero are continuation signals in uptrends; below zero in downtrends. Useful for: trend confirmation. Not useful for: range-bound markets.

Bollinger Bands: Moving average ± 2 standard deviations of price. Question answered: "How far is current price from its recent statistical center?" Bands compress in low volatility, expand in high volatility. The compression itself is the signal — a tight Bollinger squeeze often precedes a directional breakout. Useful for: volatility regime detection, mean-reversion in ranges. Not useful for: trend strength measurement.

ATR (Average True Range): Average of true range (high - low, adjusted for gaps) over a lookback period. Question answered: "How much does this stock move on a typical day?" ATR is the foundation of position sizing — your stop-loss distance should be set as a multiple of ATR, not as a fixed percentage. Useful for: position sizing, trailing stops, volatility comparison across stocks. The Trader's Toolbox uses ATR for its position size calculator.

OBV (On-Balance Volume): Cumulative sum of volume signed by direction (up day adds, down day subtracts). Question answered: "Is volume accumulating on up days or down days?" Rising OBV in a sideways market suggests accumulation. Falling OBV at new price highs is one of the most reliable bearish divergence signals known. Useful for: divergence analysis, confirmation of breakouts. Not useful for: short-term timing.

Chaikin Money Flow (CMF): Volume weighted by where the close falls within the day's range. Question answered: "Are buyers or sellers winning the intraday battle for the close?" CMF above zero indicates accumulation; below zero distribution. The Chaikin family of indicators is at the core of FlowSense's A/D scoring — see the Methodology for the exact computation.

⚡ Options Flow & Dealer Gamma — The Modern Market Structure Reality

Options flow has become an enormous portion of US equity market volume — often exceeding the underlying cash equity volume on heavily-optioned names. This isn't a side effect to ignore; it shapes price action directly.

Gamma exposure (GEX): When you buy an option, a market maker takes the other side. To stay delta-neutral, that market maker must hedge by trading the underlying stock. As price moves, their hedge requirements change — and the rate of change is their gamma exposure. When dealers are long gamma (typically near major OPEX, with high call open interest), they buy on dips and sell on rallies, dampening volatility. When dealers are short gamma (typically when puts dominate, or after a sharp move), they sell on dips and buy on rallies, AMPLIFYING volatility.

The FlowSense GEX Scanner ranks tickers by current dealer gamma exposure. A heavily negative GEX reading on a stock that's already breaking down is one of the most dangerous setups in the market — every additional tick down forces dealers to sell more underlying. This is how single-stock crashes accelerate beyond any "fundamental" explanation.

Put/Call Ratio: The ratio of put open interest (or volume) to call open interest. A high P/C ratio means more downside hedging is in place. Counterintuitively, a very high P/C ratio is often bullish — when everyone is hedged, the next surprise is more likely to be upside than downside. The crowd is rarely right at extremes.

0DTE (Zero days to expiration) flow: Same-day expiry options have become a massive force in the market since 2022. 0DTE volume can equal or exceed underlying cash volume on a typical SPY day. These options have explosive gamma in the final hour of trading, which is why end-of-day volatility on the indices has structurally increased. The Market Fragility index includes a 0DTE component for this reason.

Modern equity trading without understanding options market structure is like driving without checking the weather. The flows that move price are increasingly originating in the options market and rippling INTO the cash market, not the other way around.

🏗 Market Structure & Auction Theory

The auction theory framing — popularized by Peter Steidlmayer's Market Profile work and J. Peter Steidlmayer's "Mind Over Markets" — treats every trading day as an auction. Price moves up or down until it finds enough opposing volume to halt, then rotates back to a "fair value" range.

Key concepts: Value Area (the price range where 70% of the day's volume traded — represents "accepted" price), Point of Control (POC) (the single price with the highest volume — the most "agreed-upon" price), Initial Balance (the first hour's range — establishes the day's auction structure), and Excess (price moves beyond the value area that are rejected with little volume — represents disagreement).

The Hybrid Analytical engine that powers FlowSense's deeper analysis combines Wyckoff-style volume reading with Market Profile structural analysis. A "no-supply" bar in VSA carries even more weight when it occurs at a POC from a prior session — the institutions are buying at the price the market previously agreed was fair. Read the full computation in the Methodology page.

The most important lesson from auction theory: price extension without volume is rejection, not direction. When price moves above the value area on shrinking volume, that's the market saying "we don't agree this is fair." Expect rotation back to value. When price moves above the value area on expanding volume, the value area itself is migrating up — expect continuation. The volume tells you which is happening; price alone cannot.

Putting It Together

Technical analysis done well is an integrated discipline, not a checklist of indicators. The skilled practitioner reads volume (who's actually transacting), structure (where in the Wyckoff cycle we are), spread (whether range supports the move), and context (macro regime, options positioning, recent surprises) — and integrates them into a probability assessment, not a binary call.

FlowSense's tools are designed around this integration. The Composite Signal Engine combines multiple A/M/D inputs to filter noise. The Real vs Fake Movers page identifies which of today's gainers are backed by institutional volume versus retail momentum chasing. The Market Fragility index aggregates the conditions that historically preceded crashes. Each of these is doing the integration work most retail traders skip.