Sentinel

Trade Surveillance & Market Conduct

Agent StatusAll Active

Market Conduct Team

Surveillance & Compliance

Spoofing & Layering Detection

Rapid order placement/cancellation patterns consistent with order book manipulation across lit and dark venues

Orders Analyzed (24h)

92,840

Patterns Flagged

26

Under Investigation

2

Highest Cancel Rate

97%

AccountInstrumentPatternOrdersCancel RateScoreStatus
Prop Desk — Futures Book 2FTSE 100 Future (Z)Spoofing34097%88 Under Investigation
Client A/C 55621 (algo)DAX Future (H)Layering21294%81 Under Investigation
Market Making Unit 4EURO STOXX 50 (M)Layering16889%62 Analyst Review
Client A/C 30410 (algo)GBP/USD SpotSpoofing9691%58 Analyst Review
Prop Desk — Rates Book 110Y Gilt Future (Z)Momentum Ignition4472%29 Closed — No Action

About This Agent

Data on This Page

Order book patterns by account and instrument (spoofing, layering, momentum ignition) with order count, cancellation rate, alert score and investigation status. Stat tiles cover orders analyzed in the last 24 hours (92,840), patterns flagged (26), open investigations, and the highest observed cancel rate (97%).

What This Agent Does

Replays full order book event streams (new, amend, cancel, execute) per account and instrument, detects statistically abnormal placement-to-cancellation patterns on one side of the book intended to move price before trading the opposite side, and scores each pattern by order count, cancel rate, time-to-cancel, and correlation with a subsequent opposite-side execution.

Worked Examples

  • Prop Desk Futures Book 2 placed and cancelled 340 orders on the FTSE 100 future within 90 seconds, a 97% cancel rate immediately followed by an opposite-side execution — scored 88, under investigation.
  • Client A/C 55621's algo showed a repeating layering pattern on the DAX future across three sessions this week, each episode preceding a favourable fill on the account's resting order.
  • Prop Desk Rates Book 1's gilt future activity was closed with no action — the 72% cancel rate was consistent with normal market-making quote refresh behaviour, not manipulation.