Skip to content
FXOpen Ledger
Directory

Abook Bbook Stp Bridge Routing

Five-Second Markout Profiling: Segmenting Sharp vs Benign FX Flow

Wholesale foreign exchange market makers care very little whether a trader makes money over three weeks; dealing desks segment sharp versus benign flow by measuring statistical Z-scores of mid-price movement 100ms to 5 seconds after entry.

By Alistair Kerr, CFAReview not recorded
Five-Second Markout Profiling: Segmenting Sharp vs Benign FX Flow — FXOpen Ledger 2026 quantitative & supervisory data card
FXOpen Ledger (A-Book STP, B-Book Internalization, C-Book Hedging & LP Toxicity): Volatility-normalized +5s t-statistics separating stale-quote snipers, news EAs, and swing winners.Source: FXOpen Ledger Institutional Research DeskLicense: CC-BY-4.0

A macro hedge fund that buys EUR/USD at 1.0800 and sells six weeks later at 1.1050 (+250 pips profit!) is frequently welcomed as VIP Tier-1 benign flow by every wholesale FX market maker on Wall Street and Canary Wharf! Conversely, a high-frequency scalper whose monthly net account PnL is barely break-even after commissions can be banned from three liquidity providers within 48 hours. Why? Because automated market makers hedge and recycle inventory on horizons of 5 to 30 seconds, not six weeks! What determines whether an account destroys a liquidity provider's book is its Short-Horizon Post-Trade Markout Profile—specifically at +100ms, +1s, and +5s.

Connected to our A-Book, B-Book & C-Book FX Dealer Architecture guide and indexed inside the A-Book, B-Book & Bridge Routing hub, multi-horizon markout Z-scoring is the universal language spoken between broker risk desks and wholesale LPs.

Distinguishing Latency Snipers, Momentum Chasers & Benign Swing Traders#

By evaluating mid-price displacement simultaneously across four micro-horizons (+50ms, +500ms, +5s, and +60s), dealing-desk data scientists separate incoming accounts into four unmistakable behavioral archetypes:

  1. Sub-100ms Stale-Quote Sniper (+50ms > +0.50 bps, flat thereafter): Exploits feed lag between CME futures/NY4 ECNs and a slower retail trade server—analyzed in cross-feed latency arbitrage detection. Toxic to both B-Books AND conditional A-Book LPs!
  2. 1s–5s Order-Flow / News Momentum Sharp (+50ms ~ 0, +5s > +0.85 bps): Trades genuine short-term breakout momentum or fast news parsing. Profitable over 5–60 seconds; must route 100% to deep firm/ZAHT A-Book pools via Bayesian C-Book profiling.
  3. Long-Horizon Macro / Fundamental Winner (+5s ~ 0.00 bps, +4h > +15 bps): Harmless to wholesale 30-second LP warehouses (great A-Book flow!), YET dangerous to an unhedged multi-week retail B-Book unless portfolio delta caps externalize the slow trend!
  4. Contrarian Retail Mean-Reverter (+5s = -0.18 bps, negative spread-adjusted alpha): Fades short-term spikes right into continuing momentum, giving the counterparty immediate positive markout on top of half-spread.
Multi-horizon FX markout classification matrix separating snipers, news sharps, macro funds, and benign retail
Multi-horizon FX markout classification matrix separating snipers, news sharps, macro funds, and benign retailVerified Exhibit

Vol-Normalized Markout t-Statistic Across Heterogeneous Pairs#

Raw pip or basis-point markouts cannot be averaged naively across EUR/USD (daily vol ~ 55 bps) and GBP/JPY or XAU/USD (daily vol ~ 115-140 bps), nor across quiet Tokyo afternoons versus volatile US payroll minutes. Quantitative dealing rooms normalize each trade $k$'s 5-second mid-price drift by concurrent rolling 1-second local volatility $hat{\sigma}{1s}(tk)$ to compute a unitless Volatility-Standardized Markout t-Statistic ($T_{5s}$):

Formula or model from the articleNot independently verified
Normalized_5s_Shock (z_k) = ( d_k * (M(t_k + 5s) - M(t_k)) / M(t_k) ) / ( sigma_1s(t_k) * sqrt(5) )
  • NotionalWeightedtStat (T5s) = ( Sum{k=1..N} wk * zk ) / sqrt( Sum{k=1..N} w_k^2 )
  • Where:
  • d_k = +1 for Client Buy, -1 for Client Sell on ticket k
  • M(tk) = Independent composite ECN mid-price at execution timestamp tk
  • sigma1s(tk)= EWMA 1-second mid-price volatility of that currency pair at t_k
  • wk = Ticket USD notional Qk (weighting larger lot sizes proportionally)
  • Worked Segmentation Audit (Account #40912, N = 36 Equal-Sized $500k Trades across EUR/USD & XAU/USD):
  • Mean normalized 5-second drift z_bar = +0.48 standard deviations per trade:
  • T_5s = (36 * 0.48) / sqrt(36 * 1^2) = 17.28 / 6.0 = +2.88 (p = 0.0020 < 0.01!).
  • Spread-Adjusted LP 5s Margin: Mean half-spread earned = +0.18 bps; Mean 5s drift = +0.52 bps -> Net = -0.34 bps LOSS!
  • Automated Action: Account #40912 exceeds T_5s > 2.33 cutoff -> Promote immediately to 100% Firm A-Book STP!

Routing Matrix: Matching Trader Horizon to Optimal Counterparty Venue#

Look how cleanly multi-horizon markout segmentation resolves the tension between retail brokers—such as City Index, 4XC, and ATG World—and institutional Prime-of-Prime desks (AMB Prime / ALB Prime) under FCA and ASIC execution governance:

Quantitative Distribution · +50ms Drift
+50ms Drift+5s $T$-Stat
01Cross-Feed Latency Sniper+0.65 bps02Intraday Breakout EA+0.04 bps03Fundamental Swing Winner-0.02 bps04Retail Range Contrarian-0.03 bps
Trader Archetype+50ms Drift+5s $T$-Stat>1h Closed PnLOptimal Broker RoutingUpstream LP Reception
Cross-Feed Latency Sniper+0.65 bps+6.40Highly PositiveFix Server Quote Lag + Firm CLOBRejected / Banned on Conditional Feeds
Intraday Breakout EA+0.04 bps+2.95Moderately Positive100% Firm / Speed-Bump ECN STPAccepted on Firm Anonymous CLOBs
Fundamental Swing Winner-0.02 bps-0.15Strongly Positive100% Relationship ZAHT A-Book!Loved by Tier-1 Bank Warehouses!
Retail Range Contrarian-0.03 bps-1.85Negative (-Spread)B-Book Warehouse (Inside VaR Cap)Kept In-House (High Spread Retention)

Multi-Horizon Markout Decay Profiles by Client Flow Archetype#

Decomposing post-trade mid-price drift from +50 milliseconds out to +60 seconds separates sub-second feed snipers from slower fundamental momentum accounts:

Comparison of +50ms, +5s, and +1h markout alpha across four FX trader archetypes
Comparison of +50ms, +5s, and +1h markout alpha across four FX trader archetypesVerified Exhibit

The "Golden Synergy" Trade: Routing Profitable Swing Traders to Tier-1 Bank LPs#

Notice Row 3 in the matrix above—one of the most profitable institutional insights in OTC FX brokerage management! A disciplined fundamental swing trader who holds positions for four days and earns +$120,000 a year is dangerous to a broker's B-Book, yet exhibits T_5s = -0.15 (zero 5-second toxicity!). When the broker routes that swing trader to a Tier-1 relationship ZAHT A-Book stream, the broker locks in risk-free STP markup/swap revenue, AND the upstream Tier-1 bank loves the flow because its automated 30-second SDP warehouse internalizes the fill at +0.18 bps positive 5-second markout!

Frequently asked questions

Why does a Tier-1 FX bank welcome a hedge fund that makes millions on 2-week trends?

A Tier-1 e-FX spot engine typically holds warehoused client risk for 10 to 45 seconds before matching opposing flow; as long as the hedge fund entry does not move the market within 60 seconds, the bank earns steady spread while the fund profits from 2-week macro drift.

Why normalize 5-second markouts by 1-second EWMA volatility instead of measuring raw pips?

A +1.5 pip move in 5 seconds during a calm EUR/USD Asian session is a 5-sigma informed shock, whereas +1.5 pips in GBP/JPY during a BoE press conference is random background noise; volatility normalization weights both accurately.

How does +50ms markout separate broker server misconfiguration from skilled trading?

No human or macro model predicts 50-millisecond ticks without cross-venue arbitrage; if an account posts +0.60 bps markouts at +50ms with zero additional drift from +50ms to +5s, the broker trade server quote feed is lagging primary ECNs.

Can a trader manipulate their +5s markout score to stay on tight LP streams?

Attempting to mask sharp trades by firing decoy random micro-lots fails against notional-weighted t-statistics (w_k = Q_k), which weight $5M sweeps 500x heavier than $10k decoy tickets.

Publication record

Published
14 November 2026
Updated
14 November 2026
Author
Alistair Kerr, CFA
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
3 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Alistair Kerr, CFA. (2026, November 14). Five-Second Markout Profiling: Segmenting Sharp vs Benign FX Flow. FXOpen Ledger. https://fxopenledger.com/orderbook/five-second-markout-profiling-to-segment-sharp-vs-benign-fx-flow
Explore Abook Bbook Stp Bridge Routing Hub (20) →

FXOpen Ledger Data, Benchmarks & Contributor Records

5-Second FX Markout Profiling: Sharp vs Benign Flow | FXOpen Ledger