Venue Fragmentation & SOR Slippage Playbook (Practical)

2026-02-22 · finance

Venue Fragmentation & SOR Slippage Playbook (Practical)

Date: 2026-02-22 17:04 KST
Category: research
Scope: Production execution for fragmented equity/crypto venues

Why this matters

In fragmented markets, "best price" is often a trap when you ignore:

A naive SOR that only chases top-of-book can increase realized slippage despite better displayed prices.


1) Cost model: choose by expected net fill quality

For each venue v, score the next child order by:

ExpectedCost_v = SpreadCapture_v + Impact_v + ToxicityPenalty_v + LatencyPenalty_v + RejectPenalty_v + FeeNet_v

Where:

Route to venues minimizing ExpectedCost_v, not just best displayed quote.


2) Minimal feature set per venue (live)

Maintain rolling features (EWMA + quantiles):

If you can only keep three features, keep:

  1. markout, 2. fill ratio, 3. latency p95.

3) Regime-aware routing states

Use a lightweight state machine:

State A: Normal

State B: Stress

Trigger: volatility or spread expansion above threshold, or venue latency p95 breach.

State C: Shock

Trigger: reject spike, data lag, crossed/stale quote anomalies.

Recovery with hysteresis (don’t flap): require stable metrics for N windows before demotion.


4) Venue health score (for hard gating)

Define:

Health_v = w1*LatencyScore + w2*RejectScore + w3*ToxicityScore + w4*DataFreshnessScore

Hard rules:

This prevents one broken venue from polluting portfolio execution.


5) Calibration loop (weekly)

  1. Recompute model coefficients per liquidity bucket.
  2. Compare predicted vs realized per-venue shortfall.
  3. Track calibration drift (MAE/p95 error).
  4. Auto-reduce confidence weight on drifting venues.
  5. Review top 10 worst routing decisions manually.

Don’t chase average error only—watch p95/p99 misses.


6) Practical anti-footgun checklist


7) KPIs that actually matter

Primary:

Secondary:

A good SOR is one that lowers tail slippage without exploding opportunity loss.


8) Start simple (MVP path)

Week 1:

Week 2:

Week 3:

Week 4:

Compounding edge comes from calibration discipline, not a giant first model.