OPEN-SOURCE SCRIPT
Updated Levy Area Flow Sequencer Flow Price Lead Lag
Lévy-Area Flow Sequencer — Flow/Price Lead-Lag
What it is
Correlation says flow and price move together; it cannot say which moves first. But the sequencing is the interesting part: when aggressive flow precedes price, moves are being built by participation before they print; when price precedes flow, price is running ahead and flow is chasing — squeeze / stop-run character. Traced together, the two series form a path in the plane, and the signed (Lévy) area that path encloses measures its rotation — a scale-free, lag-free read of lead–lag, including non-linear lead–lag that fixed-lag cross-correlation misses. This is the most experimental tool of this suite, and it is framed that way.
The mathematics (signature lead–lag metric)
The metric is the antisymmetric part of the second-level path signature of the pair (flow, price): the window sum of (X·dy − Y·dx), with both increment series normalized to unit scale so the area is dimensionless. Per the literature's interpretation, the metric is positive and grows when moves in the first series are followed by same-direction moves in the second. The first series here is cumulative order-flow delta (from lower-timeframe signed volume, with bar-shape fallback) and the second is price, so AREA > 0 → FLOW LEADS and AREA < 0 → PRICE LEADS.
The honesty steps
Significance gate — a raw signed area is noisy, so the reading is ranked against its own recent history, and a lead is declared only when rotation is unusually strong for this symbol/timeframe. Otherwise the state is BALANCED: no claim.
Sequencing ≠ causation — the literature is explicit that a signed area alone cannot establish causal direction. This tool reports a temporal-ordering tendency of past bars; treat it as tape character.
Known limitation, stated — persistent inverse co-movement between flow and price can contaminate the sign. On liquid futures they co-move and the read behaves; on instruments where they reliably anti-correlate, don't trust it.
The stability & multi-timeframe layer
States are dwell-filtered (standard anti-chattering): FLOW LEADS / PRICE LEADS / BALANCED is announced only after surviving a set number of confirmed bars, so the read doesn't flip-flop. STABILITY shows how settled it is; PENDING shows a forming state with a countdown. Cost: a few bars of lag — stated and adjustable.
The lead lane — a thin strip at the pane bottom — gives the glance-read: green = flow leads (moves better backed), amber = price leads (flow chasing, be sceptical), gray = balanced. Trust/caution colors, never direction.
The HTF STACK row shows the raw lead state on three higher timeframes derived as multiples of the chart (defaults 3×, 5×, 15×). Honesty note: lower-timeframe data cannot be requested inside a higher-timeframe request, so the HTF slots use the bar-shape delta proxy — a stated approximation. ✓ = all timeframes agree on the same significant lead; ⚠ = a higher timeframe shows the opposite lead.
How to use it
Add to a liquid intraday chart. Read the dashboard: FLOW LEADS → breakouts/drives carry more weight (participation came first); PRICE LEADS → be sceptical of extensions (flow is chasing); BALANCED → the tool makes no claim.
Tags print when the lead flips while significant; alerts fire on flips.
Use it as context alongside order-flow and structure tools — never as a standalone signal.
What makes it original
Path-signature methods are frontier quantitative machinery (rough-path theory) that has reached systematic trading but, to the author's knowledge, not chart platforms. Applying the signature lead-lag metric to the flow-vs-price pair — the pair an order-flow trader actually cares about — with an honest significance gate and stated limitations, is the contribution.
Concept credits
Signed area of stochastic paths — P. Lévy. Rough-path / signature theory — T. Lyons; Levin, Lyons & Ni (2016). Signature lead-lag metric and interpretation — I. Chevyrev & A. Kormilitzin (2016). Market applications — Bennett, Cucuringu & Reinert (2022); Cartea, Cucuringu & Jin (2023). Implementation and charting design are the author's own.
Important disclaimer
Research and education only. Not financial advice, not a signal service, not a guarantee of future results. The area measures a sequencing tendency in past data; it is not causal proof and not a prediction. Validate independently and manage your own risk.
What it is
Correlation says flow and price move together; it cannot say which moves first. But the sequencing is the interesting part: when aggressive flow precedes price, moves are being built by participation before they print; when price precedes flow, price is running ahead and flow is chasing — squeeze / stop-run character. Traced together, the two series form a path in the plane, and the signed (Lévy) area that path encloses measures its rotation — a scale-free, lag-free read of lead–lag, including non-linear lead–lag that fixed-lag cross-correlation misses. This is the most experimental tool of this suite, and it is framed that way.
The mathematics (signature lead–lag metric)
The metric is the antisymmetric part of the second-level path signature of the pair (flow, price): the window sum of (X·dy − Y·dx), with both increment series normalized to unit scale so the area is dimensionless. Per the literature's interpretation, the metric is positive and grows when moves in the first series are followed by same-direction moves in the second. The first series here is cumulative order-flow delta (from lower-timeframe signed volume, with bar-shape fallback) and the second is price, so AREA > 0 → FLOW LEADS and AREA < 0 → PRICE LEADS.
The honesty steps
Significance gate — a raw signed area is noisy, so the reading is ranked against its own recent history, and a lead is declared only when rotation is unusually strong for this symbol/timeframe. Otherwise the state is BALANCED: no claim.
Sequencing ≠ causation — the literature is explicit that a signed area alone cannot establish causal direction. This tool reports a temporal-ordering tendency of past bars; treat it as tape character.
Known limitation, stated — persistent inverse co-movement between flow and price can contaminate the sign. On liquid futures they co-move and the read behaves; on instruments where they reliably anti-correlate, don't trust it.
The stability & multi-timeframe layer
States are dwell-filtered (standard anti-chattering): FLOW LEADS / PRICE LEADS / BALANCED is announced only after surviving a set number of confirmed bars, so the read doesn't flip-flop. STABILITY shows how settled it is; PENDING shows a forming state with a countdown. Cost: a few bars of lag — stated and adjustable.
The lead lane — a thin strip at the pane bottom — gives the glance-read: green = flow leads (moves better backed), amber = price leads (flow chasing, be sceptical), gray = balanced. Trust/caution colors, never direction.
The HTF STACK row shows the raw lead state on three higher timeframes derived as multiples of the chart (defaults 3×, 5×, 15×). Honesty note: lower-timeframe data cannot be requested inside a higher-timeframe request, so the HTF slots use the bar-shape delta proxy — a stated approximation. ✓ = all timeframes agree on the same significant lead; ⚠ = a higher timeframe shows the opposite lead.
How to use it
Add to a liquid intraday chart. Read the dashboard: FLOW LEADS → breakouts/drives carry more weight (participation came first); PRICE LEADS → be sceptical of extensions (flow is chasing); BALANCED → the tool makes no claim.
Tags print when the lead flips while significant; alerts fire on flips.
Use it as context alongside order-flow and structure tools — never as a standalone signal.
What makes it original
Path-signature methods are frontier quantitative machinery (rough-path theory) that has reached systematic trading but, to the author's knowledge, not chart platforms. Applying the signature lead-lag metric to the flow-vs-price pair — the pair an order-flow trader actually cares about — with an honest significance gate and stated limitations, is the contribution.
Concept credits
Signed area of stochastic paths — P. Lévy. Rough-path / signature theory — T. Lyons; Levin, Lyons & Ni (2016). Signature lead-lag metric and interpretation — I. Chevyrev & A. Kormilitzin (2016). Market applications — Bennett, Cucuringu & Reinert (2022); Cartea, Cucuringu & Jin (2023). Implementation and charting design are the author's own.
Important disclaimer
Research and education only. Not financial advice, not a signal service, not a guarantee of future results. The area measures a sequencing tendency in past data; it is not causal proof and not a prediction. Validate independently and manage your own risk.
Release Notes
v1.1 — Finishing pass (no engine change)- Cleared a Pine warning (CW10004): two smoothing averages called ta.ema from inside a
"skip while warming up" check. They now run every bar via a seed-on-first-valid EMA that
reproduces the exact same output after warm-up — only the warning is gone.
- Added a Data Window (EXP_) export bus so the signed area, rotation rank and lead state can feed
other scripts via input.source().
- Added the MPL-2.0 licence header.
- No change to the Lévy-area engine, the lead state, the multi-timeframe reads or the dashboard —
everyday behaviour is identical.
Descriptive lead-lag research tool, not investment advice.
Open-source script
In true TradingView spirit, the creator of this script has made it open-source, so that traders can review and verify its functionality. Kudos to the author! While you can use it for free, remember that republishing the code is subject to our House Rules.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.
Open-source script
In true TradingView spirit, the creator of this script has made it open-source, so that traders can review and verify its functionality. Kudos to the author! While you can use it for free, remember that republishing the code is subject to our House Rules.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.