Learn Orderflow Trading
Orderflow analysis looks at what traders are actually doing — not just where price went, but who is buying, who is selling, and with what conviction. These are the tools institutional desks use. Here's how to read them.
Getting Started — Free Tools
These tools are available to everyone, no account needed. Start here to understand the basics of orderflow.
CVD — Cumulative Volume Delta
Net buying vs selling pressure over time.
What it measures
CVD tracks the running total of buy volume minus sell volume. When someone aggressively buys (market order hitting the ask), it adds to CVD. When someone aggressively sells (hitting the bid), it subtracts.
Why traders use it
Price can move on low volume, but CVD shows whether the move has real conviction behind it. A price rally with rising CVD is backed by actual buying. A rally with flat or falling CVD is weak and likely to reverse.
How to read it
Green and rising = buyers in control. Red and falling = sellers in control. The most powerful signal is divergence: when price goes one way but CVD goes the other.
Key signals to watch
Example scenario
BTC drops from $100K to $98K but CVD keeps climbing — buyers are absorbing the dip. Price likely recovers.
VPIN — Flow Toxicity
Probability that informed (smart) money is dominating the flow.
What it measures
VPIN splits trades into fixed-volume buckets (e.g., every $500K traded) and measures how one-sided each bucket is. A bucket where 90% is buy-side is highly imbalanced = informed flow. VPIN averages this over 50 buckets.
Why traders use it
When VPIN is high, "smart money" is moving — they know something the market doesn't. This is when market makers pull liquidity and volatility spikes. VPIN famously predicted the 2010 Flash Crash hours before it happened.
How to read it
Below 45% = safe, normal two-sided trading. 45-70% = elevated, watch closely. Above 70% = danger zone, someone is aggressively positioning. Consider reducing exposure or tightening stops.
Key signals to watch
Example scenario
VPIN spikes to 78% on BTC. Within 20 minutes, price drops 4% as the informed sellers finish exiting and stop hunts trigger.
Whale Detection
Real-time alerts on unusually large trades.
What it measures
We track every trade and flag those above a dynamic threshold (top 1% by size). Each whale trade shows direction (buy/sell), USD value, price, and time. The threshold adapts per pair — a $100K trade is whale-sized for a small alt but normal for BTC.
Why traders use it
Large trades often come from informed participants — institutions, funds, or large individual traders with an information edge. Tracking their direction gives you a high-signal indicator of where the "big money" thinks price is going.
How to read it
Green = large buy. Red = large sell. Watch for clusters — 3+ whale buys in quick succession is a much stronger signal than a single one. Also check if whales are trading with or against the current trend.
Key signals to watch
Example scenario
Three whale buys of $500K+ on SOL in 2 minutes, all during a -3% dip. Smart money is loading. Price recovers 5% by end of session.
Session Analytics
Trading activity broken down by global time zone.
What it measures
Splits the trading day into sessions — Asia (00-08 UTC), Europe (07-16 UTC), and US (13-21 UTC) — and tracks volume, CVD delta, whale count, and price change for each. Crypto trades 24/7 but behavior patterns differ by session.
Why traders use it
Different sessions have different characters. Asian sessions often see accumulation and range-bound action. European opens bring volatility. US sessions tend to set the direction. Understanding which session you're in helps you calibrate expectations.
How to read it
Compare today's session stats with historical averages. If Asian session volume is 3x average, something unusual is happening. If delta is strongly negative in Asia but price hasn't dropped, expect the EU session to react.
Key signals to watch
Example scenario
Asian session shows $50M volume (3x average) with strong positive delta. By the time US opens, price is up 4% — Asia front-ran the move.
Open Interest Analytics
Tracks new positions being opened and closed across the market.
What it measures
Open Interest (OI) is the total number of outstanding derivative contracts. When OI rises, new positions are being opened (fresh money entering). When OI falls, positions are being closed (money leaving). Delta OI shows the rate of change.
Why traders use it
OI context transforms price analysis. Rising price with rising OI = new longs entering (trend continuation). Rising price with falling OI = shorts covering (exhaustion, potential top). OI divergence from price is one of the most reliable signals in derivatives trading.
How to read it
Watch the delta (rate of change) more than the absolute level. A sudden OI spike without price movement = hidden accumulation. An OI drop with stable price = quiet position exit. The divergence indicator highlights when OI and price are moving in opposite directions.
Key signals to watch
Example scenario
ETH OI jumps $200M in an hour but price is flat. Someone is building a huge position. When the move comes, it's explosive.
Funding Rate Cross-Exchange
Compares Hyperliquid funding with Binance and Bybit.
What it measures
Perpetual futures use funding rates to keep the price aligned with spot. Positive funding = longs pay shorts (market is bullish/overleveraged long). Negative = shorts pay longs. We show HL, Binance, and Bybit funding side by side.
Why traders use it
Extreme funding is a contrarian signal — when everyone is long (high positive funding), a squeeze to the downside is likely. Cross-exchange comparison reveals arbitrage: if HL funding is 0.1% but Bybit is 0.01%, you can potentially capture the difference.
How to read it
Compare the three exchanges. Large spreads = arbitrage opportunity. Extreme rates on all three = market-wide overleverage. Watch for funding flips (positive to negative) as momentum shift signals.
Key signals to watch
Example scenario
BTC funding on HL is +0.08%, Binance is +0.01%. HL traders are much more leveraged long. HL-specific long squeeze hits, price drops 2% on HL while Binance barely moves.
Pro Orderflow Analytics
Advanced tools that reveal the microstructure of the market. Available with Pro subscription or 7-day free trial.
OBI — Order Book Imbalance
Ratio of bid vs ask liquidity in the order book.
What it measures
OBI measures how much buying interest (bids) vs selling interest (asks) exists in the top 10 levels of the order book. It ranges from -100% (all asks, no bids) to +100% (all bids, no asks).
Why traders use it
The order book shows intent — where traders are willing to buy and sell. A heavy bid side suggests support; a heavy ask side suggests resistance. Unlike CVD which tracks what already happened, OBI shows what might happen next.
How to read it
Positive = more bids than asks (buying wall). Negative = more asks (selling wall). Watch for sudden shifts — when OBI flips from positive to negative, the order book just changed character.
Key signals to watch
Example scenario
ETH has OBI at +35% — heavy bids stacked. Price pulls back to those bids and bounces. The wall held.
OFI — Order Flow Imbalance
How aggressively orders are being placed between snapshots.
What it measures
OFI measures the change in order book volume between consecutive snapshots. It captures not just where orders sit (OBI), but how fast they're being added or pulled. Think of it as the "velocity" of the order book.
Why traders use it
OFI is more sensitive than OBI because it catches movement before it settles. A sudden surge in OFI means aggressive order placement — someone is in a hurry to get positioned.
How to read it
Positive = bids being added faster than asks (aggressive buying intent). Negative = asks building fast (aggressive selling). Spikes are more important than steady values.
Key signals to watch
Example scenario
SOL OFI spikes to +80% in seconds — someone just placed a massive bid ladder. Price launches 3% in the next minute.
Smart Money Delta
CVD broken down by who is trading: whales, HLP, or retail.
What it measures
Every trade on Hyperliquid reveals the wallet addresses of buyer and seller. We classify wallets into three groups: Whales (top 5% by volume), HLP (the protocol's own market maker), and Retail (everyone else). Then we show each group's contribution to the CVD separately.
Why traders use it
Knowing total CVD is useful. Knowing that the whale CVD is +$2M while retail CVD is -$500K is game-changing. It means whales are buying what retail is selling — a classic setup for a move in the whales' direction.
How to read it
Watch for divergences between cohorts. When whales and retail disagree, whales usually win. HLP is often a counterparty — if HLP is heavily one-sided, it might be absorbing flow that will later reverse.
Key signals to watch
Example scenario
ARB whale delta is +$800K while retail is -$200K. Whales are buying the dip retail is selling. Price rallies 8% over next 4 hours.
Absorption Detection
Spots hidden buyers or sellers defending a price level.
What it measures
Absorption happens when large limit orders silently consume aggressive market orders without the price moving. It's like a wall that absorbs all the selling (or buying) thrown at it. We detect this by looking for high volume delta with minimal price change.
Why traders use it
Absorption is one of the strongest reversal signals in orderflow trading. If someone is willing to absorb millions in selling at a specific price, they have conviction that price belongs higher. When the selling exhausts, the absorber often pushes price their way.
How to read it
Buy absorption (green): heavy selling + price holds = hidden buyer. Sell absorption (red): heavy buying + price capped = hidden seller. Higher strength = more significant. Multiple absorption events at the same level = very strong.
Key signals to watch
Example scenario
BTC at $97,000 — 3 buy absorption events in 10 minutes, all strength 80+. Someone is defending $97K hard. Price bounces to $99K.
Footprint Chart
Volume heatmap showing buy vs sell pressure at each price level.
What it measures
The footprint chart breaks down volume into a grid: each row is a price level, each column shows bid volume (green, left) and ask volume (red, right). The Δ column shows the net (bids minus asks). Color intensity = volume concentration.
Why traders use it
Regular candles hide where the action happened within the bar. The footprint shows exactly which price levels attracted the most volume and whether it was buying or selling. You can see absorption, exhaustion, and initiative at specific levels.
How to read it
Dark green cells = heavy buying at that level. Dark red = heavy selling. The Δ column is key: big positive delta = buying dominated, big negative = selling. Look for imbalances (one side much stronger) at support/resistance levels.
Key signals to watch
Example scenario
ETH footprint shows $3,450 level has $5M bid volume vs $500K ask. The buyers own this level — it becomes strong support.
Markout Analysis
Measures whether recent traders are "smart" or "noise."
What it measures
After every trade, we track where the price goes at 1 second, 5 seconds, 30 seconds, and 5 minutes. If buy trades consistently lead to price increases, the buyers are "informed" — they're trading on real signal. If buys lead to price drops, they're noise traders getting it wrong.
Why traders use it
Markout reveals flow quality. When takers are consistently right (toxic flow), it means smart money is active — be cautious trading against them. When takers are wrong (safe flow), it's mostly retail noise — mean reversion strategies work better.
How to read it
Positive buy markout = price goes UP after buys (toxic — buyers are right). Negative buy markout = price goes DOWN after buys (safe — buyers are wrong). The 30-second horizon is usually most informative.
Key signals to watch
Example scenario
BTC buy markout at 30s is +0.03% — every buy is followed by a small price increase. Buyers are informed today, don't fade them.
Kyle's Lambda — Market Impact
How much price moves per $1M of volume traded.
What it measures
Kyle's Lambda estimates the price impact of trading. It runs a regression of price changes against signed volume. A high lambda means the market is thin — even moderate trades move the price. A low lambda means deep liquidity.
Why traders use it
Lambda tells you how "expensive" it is to trade. Before entering a position, you want to know: will my order move the market? Lambda also reveals market structure — when lambda spikes, liquidity has dried up and volatility is coming.
How to read it
Low lambda (< 0.001) = thick book, safe to trade size. High lambda (> 0.005) = thin book, be careful with size. Sudden lambda increase = liquidity pulled, potential for sharp moves.
Key signals to watch
Example scenario
DOGE lambda triples in 5 minutes — market makers pulled orders. A $100K market order that normally moves price 0.1% now moves it 0.5%.
Regime Detection
Classifies market as trending, ranging, or volatile.
What it measures
Analyzes recent price action to determine the current market "regime" — whether it's trending (strong directional move), ranging (oscillating between levels), or volatile (large swings without clear direction). Uses ATR, directional strength, and range analysis.
Why traders use it
Different regimes need different strategies. Trend-following works in trending markets but gets chopped up in ranges. Mean-reversion works in ranges but gets destroyed in trends. Knowing the regime helps you pick the right approach.
How to read it
Trending (blue arrow) = follow the direction, ride momentum. Ranging (orange arrow) = trade the boundaries, buy support, sell resistance. Volatile (red lightning) = reduce size, widen stops, or wait for clarity.
Key signals to watch
Example scenario
BTC regime flips from ranging to trending with 85% confidence. The breakout from the range is real — price runs 6% in the trend direction.
Cross-Asset Correlation
Shows how orderflow in one asset predicts moves in another.
What it measures
Computes rolling correlation between the CVD (volume delta) of different pairs. Also estimates lead-lag — whether one asset's orderflow predicts another's price movement. For example, BTC CVD often leads ETH price by a few seconds.
Why traders use it
When historically correlated assets suddenly decorrelate, it's a strong signal. If BTC and ETH CVDs are normally 0.8 correlated but drop to 0.2, something unusual is happening in one of them — likely an informed trade or structural shift.
How to read it
High correlation (> 0.7) = assets moving together, normal. Low correlation (< 0.3) = divergence, investigate. Lead-lag positive = asset A leads B. Watch for correlation breakdowns as early warnings.
Key signals to watch
Example scenario
BTC CVD is rising but ETH CVD is flat (correlation dropped from 0.85 to 0.3). BTC is being bought specifically while ETH is ignored — BTC-specific catalyst.
Ready to see the orderflow?
Open the screener — it's free, no signup required. Pick any pair and start analyzing.
Open Screener — Free