Article 30

Market Anomaly: Spread-Pinning (1-Tick Fixation of the Spread)

In short: Spread-pinning denotes phases in which the visible bid/ask distance remains exactly 1 tick for extended periods - sometimes minutes - regardless of actual risk or orderflow. For professionals: the pattern relies on tick-size calibration, price-time priority and post-only/amend paths. It shows as a high 1-tick-occupancy share (S1O), shortened best-quote half-life and OTR spikes with minimal VWAP drift. Pinning can be legitimate, competitive market-making; it becomes anomalous when fixation plus sequence signatures point to perverse incentives (e.g. rebate windows, tick changes) rather than real liquidity.

Market CEX Spot, CEX Derivatives

Evidence status & date Spot-based evidence · Updated: October 18, 2025, 12:00 UTC

Documented Scenarios (CEX-based)

18.11.2021 – Bitstamp analysis “Not all liquidity is created equal” (Spot): Explicitly describes cases where the bid/ask spread on leading CEXs (e.g. Coinbase / Kraken) gravitated to 1 tick; each print briefly opens the spread which then reverts to 1 tick. (The Bitstamp Blog)
23.05.2024 / 24.09.2024 – Binance tick-size adjustments (Spot): Wave-like tick adjustments “to increase liquidity”; changes shift minimum spreads and can strengthen or resolve pinning phases. (Binance)
11.06.2025 – Study on Bitcoin tick sizes (Kraken vs. Coinbase/Gemini): Empirical evidence shows differing tick regimes shift trading migration and costs; price quality varies by tick regime. (SSRN)

Clarification: Derivative phenomena such as locking/crossing rules in equities are not evidence for spot; they serve only as a rules contrast.

Functional Principle

Tick anchoring: If the tick is large relative to perceived risk, quotes bind at the minimum: best bid at mid − ½ tick, best ask at mid + ½ tick. Each execution briefly widens the spread - re-quotes immediately pull it back to 1 tick.
Queue & incentives: Price–time priority rewards front queue positions. Post-Only and in-place amends (where allowed) reduce the cost of permanent re-quoting ⇒ high OTR, low best-quote t½ - without proportional trade throughput.
Pinning vs price discovery: Genuine price discovery “breathes” with imbalance/impact. Spread-pinning keeps spreads artificially tight and pulls away once fill proximity threatens (quote-fade). Tick changes or promotions can temporarily force or resolve pinning.

Distinct Detection Features (live observable)

1-Tick-Occupancy (S1O) ↑: Share of time with spread = 1 tick significantly above baseline, often > 90% in quiet minutes (30-day comparison, same time of day).
Best-quote t½ ↓ + OTR ↑: Extremely short-lived touch quotes, many updates per fill; spread jumps remain moderate.
Quote-fade chain: Near fill → touch withdraws → immediate re-quote at the same price; cancel-to-fill ratio at the touch is above average.
Pinning persistence across events: 1-tick fixation for minutes without correlated directional trades (VWAP drift vs mid is small).
Cross-venue contrast: Core venue “breathes” while target venue pins → lead/lag (0.5–1.0 s) and a noticeable mirror-error divergence.

Why CEXs are vulnerable

Regular tick maintenance: Large CEXs calibrate ticks continuously, with precise UTC windows - predictable re-calibration windows for quoting algos.
Price–time + amend paths: Atomic amends / queue preservation (where available) promote queue retention at minimal price step - pennying loops become cheaper.
Rebate / VIP incentives: Maker/taker & VIP tiers reward tight spreads / resting quotes - pinning can be economically rational.
Transparency limit (L2): No queue IDs, hidden/iceberg orders possible ⇒ diagnosis via signatures (S1O, t½, OTR, fade-quote) only.

Comparison: regulated exchanges

Tick regimes (EU): MiFID II / RTS 11 calibrates tick bands; mis-calibrations are revisited. Goal: preserve spread quality without excessive pennying incentives. (ESMA)

OTR guardrails (EU): RTS 9 requires order-to-trade ratio controls - quote-churn becomes measurable & addressable. (European Commission)

US framework (Reg NMS): Rule 610 prohibits lock/cross quotes in equities; tick/access rules (including 2024 amendments) promote fair, accessible quotes. Crypto spot lacks market-wide duties of that breadth. (SEC)

Why early detection matters - and what changes in the EU

MiCA timeline: In force 29.06.2023; applicable 30.06.2024 (ART/EMT) & 30.12.2024 (CASP duties). Transitional regimes until 01.07.2026 possible - heterogeneous supervision in 2025/26. (ESMA)
Practical relevance: 1-tick fixation hides true execution depth, produces apparent depth / mis-routing. Early detection supports best-execution & surveillance (intent-agnostic).

Concrete thresholds / alert rules (E3 / E2 / E1)

Baseline: 30-day history, same time of day (UTC), hourly bins, ±15 min tolerance, per symbol × venue.

Core metrics: S1O (share-of-1-tick-occupancy %); best-quote t½ (ms); OTR (orders/trades) at top-5 depth; cancel-to-fill (%) at the touch; VWAP drift vs mid (bps); spread jump (bps); lead/lag (1-s bins, ≥3 venues); round-size ratio; Benford distance (L1).

E3 – high (“persistent 1-tick fixation + churn + re-fade”)

S1O ≥ P99 over ≥ 5 min and best-quote t½ ≤ P1,
OTR ≥ P99.5 and cancel-to-fill ≥ P99 at the touch,
VWAP drift vs mid ≤ P40 with spread jump ≤ P50 (tight spreads despite heavy activity),
Lead/lag coherence: pinning only on 1–2 venues; core venue “breathes” → delta pattern over ≥ 5 1-s bins.

E2 – medium (partial signals)

S1O ≥ P98 or t½ ≤ P5,
OTR ≥ P99 or cancel-to-fill ≥ P98,
RoundSize / Benford deviation ≥ P95 (rounding clusters at minimal improvements).

E1 – low (early warning)

sudden S1O uptick ≥ P95 ± 15 min around tick/fee events,
BBO flicker (changes/s ≥ P95) with spread jump ≤ P60.

De-escalation: Downgrade when S1O < P80, t½ ≥ P20 and OTR < P80 for ≥ 10 min.

Practical notes (minimizing false alarms)

Event filter (UTC): Tick-size changes, zero-fee / promotion windows, listings/relistings, maintenance windows - do not escalate otherwise.
Cross-venue check: E3 only when pinning coherence is not market-wide. Single-venue pinning + fade chain ⇒ higher anomaly probability.
Legitimate alternatives: highly competitive market-making, fine tick calibration or broad liquidity can create S1O highs without abuse. If fade/OTR excess is absent and VWAP drift is small, de-escalate.
Data quality: Discard time windows with L2 gaps / error rate > 1%; measure in bps/ms, UTC-stamp everything; document notional.

Why it matters (Trader benefit & compliance)

Execution & cost: Detect spread-pinning early ⇒ don’t route blindly into apparent depth; align limit placement to t½/OTR and reduce slippage.
Strategy: Tick-aware scheduling (before/after tick events), queue-aware sizing (avoid starving in endless 1-tick loops), venue selection by “breathing” vs pinned spreads.
Surveillance & best-execution: S1O / t½ / OTR / fade as intent-agnostic evidence - fits MiCA/MAR expectations, especially without a CAT equivalent in crypto spot.

Relevant sources

Bitstamp – Not all liquidity is created equal (18.11.2021).
Binance – Updates on Tick Size for Spot Trading Pairs (16.05.2024 / 24.09.2024 / 14.10.2025).
Coinbase – Coinbase Pro Market Structure Update (15.03.2019).
SSRN – Do Tick Sizes Need to be Regulated? Evidence from the Bitcoin Market (11.06.2025).
ESMA – Final Report on Amendments to RTS 11 (Tick Size Regime) (12.12.2018).
ESMA – RTS 9 Order-to-Trade-Ratio (18.05.2016) / Final Report on Algorithmic Trading (28.09.2021).
SEC – Regulation NMS: Access Rule / Minimum Pricing Increments (Rule 610; 2005/2024 Updates).
ESMA – Statement on MiCA Transitional Measures (17.12.2024).