Market Anomaly: Tick-Size / Queue Gaming
In short: Tick-Size / Queue-Gaming describes deliberately exploiting price increments (tick sizes) and queue rules (price–time priority) in spot orderbooks to step-ahead/penny, defend queue positions (e.g. amend without loss of priority) or distort market quality through unnatural re-quoting. For professionals: shortened best-quote half-life, high OTR, step-ahead rate (1-tick minimal improvements), cancel→fill ratios at the touch and coherence with tick-change events.
Documented Scenarios (CEX-based)
- 15.03.2019 – Coinbase Pro “Market Structure Update” (Spot): New tick sizes, order limits & protective measures; tick calibration affects queue density and step-ahead incentives. (Coinbase)
- 2024–2025 – Binance & Binance.US: Tick-size adjustments (Spot): Multiple adjustments “to increase liquidity” - shifting step-ahead costs and potentially driving short-term OTR/quote churn. (Binance)
- 27.09.2024 – Kraken “Atomic Amends” (Spot): In-place amends with queue priority maintained where possible - facilitates queue retention and step-ahead loops. (docs.kraken.com)
Clarification: Perpetual/futures tick pilots are not considered here - focus is on CEX spot orderbooks and rules.
Functional Principle
- Tick economy: Small ticks favor cheap step-ahead; large ticks compress price levels and increase pennying costs. Tick calibration shifts incentive zones. (SEC)
- Queue rule & amend: Under price–time priority it's worth defending a timestamp. In-place amends (where allowed) preserve queue position; cancel+resubmit loses it. (Kraken)
- Gaming sequence: Micro-quote → 1-tick step-ahead (minimal improvement) → cancel before fill → repeat step-ahead → touch flicker without genuine execution intent.
- Tick events: Around tick-size changes: quote bunching & OTR spikes in the UTC migration window; subsequent re-calibration of queue topology. (Binance)
Distinct Detection Features (live observable)
- Step-ahead rate ↑: Clustering of 1-tick improvements at minimal sizes (rounding clusters) immediately before expected execution proximity.
- Best-quote half-life ↓ with spread jump ≤ P50: touch flickers while spread remains relatively tight.
- OTR spikes & quote churn: Many updates per fill at top-of-book; cancel→fill quote ratio at the touch increases.
- Event coherence (UTC): Patterns start/end sharply around venue tick-change announcements (API notices/announcements).
- Lead/lag asymmetry: One venue leads step-ahead bursts, others follow with latency (1-s bins).
Why CEXs Are Vulnerable
- Fine-grained tick control: Regular tick adjustments at exact UTC times → predictable re-calibration windows.
- Price–time priority + amend path: Queue retention where possible reduces opportunity cost for persistent re-quoting.
- Transparency limits: L2 without queue IDs - signature detection must rely on timing, t½, OTR; hidden/iceberg orders obscure residual size.
- 24/7 & fragmentation: Different tick tables & latencies → asynchronous step-ahead trails, fertile ground for queue-gaming.
Comparison: Regulated Markets
EU tick regime (RTS 11): Tick bands by liquidity/price; systematic control of pennying costs. (ESMA)
OTR guardrails (RTS 9): Order-to-trade ratio ceilings to limit disorderly trading. (EU Commission)
US tick pilot: Evidence on trade-offs (spread/liquidity vs undercutting/complexity) - instructive for tick calibration. (SEC)
Bottom line: Ticks are calibrated and OTR is tracked - queue-gaming becomes visible and constrained earlier in regulated markets.
Why Early Detection Is Critical - and What Changes in the EU
- MiCA (EU): Since 30.12.2024 central CASP duties; 2025 ESMA guidelines on detection/prevention of market abuse (behavioural patterns). Transitional periods until 01.07.2026 possible. (ESMA)
- Surveillance path: Intent-agnostic indicators (t½, OTR, step-ahead rate, event coherence) gain importance due to lack of a CAT-equivalent.
- Practice benefit: Early detection prevents phantom-depth traps, stabilises execution quality and produces auditable dossiers for MiCA/MAR reviews.
Concrete Thresholds / Alert Rules (E3 / E2 / E1)
Baseline: 30-day history, hourly bins (UTC), tolerance ±15 min, per symbol × venue. Thresholds relative to percentile baseline.
Metrics (selection): Step-ahead rate (share of 1-tick improvements at the touch), best-quote t½ (ms), OTR (orders/trades) Top-5 depth, cancel→fill at the touch, spread jumps (bps), VWAP drift vs mid (bps), round-size ratio, Benford-L1, lead/lag (1-s bins, ≥3 venues), event coherence to tick-change (UTC).
E3 – high (persistent step-ahead + queue retention)
- Step-ahead rate ≥ P99 and best-quote t½ ≤ P1 over ≥ 3 min,
- OTR ≥ P99.5 and cancel→fill ≥ P99 at the touch,
- Spread jump ≤ P50 (tight spreads despite churn) with VWAP drift ≥ P99,
- Event coherence: start ≤ 15 min around venue tick-update UTC (if published).
E2 – medium (partial signals)
- Step-ahead rate ≥ P98 or t½ ≤ P5,
- OTR ≥ P99 or cancel→fill ≥ P98,
- Lead/lag: a venue leads step-ahead bursts > 700 ms across ≥ 5 1-s bins.
E1 – low (early warning)
- BBO flicker (changes/s ≥ P95) with spread jump ≤ P60,
- Rounding clusters (round-size ratio ≥ 1.5×) or Benford-L1 ≥ P90 in the touch flow.
De-Escalation: Downgrade when t½ ≥ P20 returns and OTR < P80 and step-ahead rate < P80 for ≥ 10 min.
Practical Notes (Minimizing False Alarms)
- Event filter (UTC): Only raise E-levels if tick changes or fee/VIP incentives are not the trigger. Tick updates are communicated precisely. (Binance)
- Check venue rules: Amend semantics (queue retention “where possible”) vary by venue/field - crucial for queue retention logic. (Kraken)
- Cross-venue checks: E3 only when ≥ 3 venues show coherent patterns; single-venue flicker may be a feed artifact.
- Legitimate alternatives: Re-calibration after tick updates, MM inventory control, latency hedges; absence of post-event reversion and a consistent notional path argue against anomaly.
- Data quality: Discard windows with L2 gaps/errors > 1%; measure in bps/ms, UTC-stamp and record notional.
Why It Matters (Trader Benefit & Compliance)
- Execution: Recognize step-ahead traps - do not park limits in high-churn touch queues; route latency-/tick-aware to venues with stable t½.
- Cost / Risk: Slippage & opportunity loss fall when event coherence (tick changes) is respected and OTR/cancel spikes are avoided.
- Best-execution & supervision: Intent-agnostic E-level logs (step-ahead, t½, OTR, cancel→fill, event coherence) are auditable - aligned with MiCA guidelines and RTS 9/11 expectations.
Relevant Sources
- Coinbase – Coinbase Pro Market Structure Update (15.03.2019).
- Binance – Updates on Tick Size for Spot Trading Pairs (15.05.2024; 16.09.2024; 06.10.2025 / 14.10.2025).
- Kraken – Atomic Amends / Amend Order / Spot Atomic Amends (Queue Priority) (27.09.2024; ongoing).
- ESMA – RTS 11 Tick-Size Regime (12.12.2018; update 03.10.2025).
- ESMA – RTS 9 Order-to-Trade Ratio (18.05.2016) & MiFID II Final Report on Algorithmic Trading (28.09.2021).
- SEC / DERA – Tick Sizes and Market Quality: Revisiting the Tick Size Pilot (28.11.2022); NYSE – EMA Report on SEC Tick Pilot (07.04.2017).
- ESMA – MiCA: Guidelines on Prevention & Detection of Market Abuse (29.04.2025).