
Lecture Scope
This econometric review unpacks regulator-grade market-structure statistics — routing pathways, speed benchmarks, and off-exchange liquidity — into teaching material for simulation-literate readers. It is the most technically demanding lecture in the series.
The lecture is deliberately technical: market-structure numbers are the raw material honest fill-assumptions are built from. A simulated 'instant fill' that knows the real-world distribution it approximates is a responsible simplification; one that doesn't is a fantasy.
Students are expected to have completed the market-versus-limit-orders material first; this lecture supplies the empirical floor those order-type discussions stand on.
Key Themes
- Order-routing statistics reveal how few retail orders ever see a lit exchange — the context every simulated 'instant fill' should cite.
- Execution-speed benchmarks are distribution tails, not averages; quoting the median conceals the regime that matters.
- Dark liquidity trends shift the visible book's meaning — displayed depth is a fraction of true interest.
- High-frequency regimes make queue position a first-order variable simulators almost always omit.
- Venue payment structures shape routing more than price improvement claims suggest — economics of routing is a modeling input.
- Every market-structure statistic carries a measurement window and venue scope; both belong in the journal.
Fill-Assumption Grounding
| Simulator Default | Real-World Structure | Journal Annotation |
|---|---|---|
| Instant fill at touched price | Queue position determines fill probability | Tag every 'touched = filled' as optimistic |
| Uniform spread | Spreads widen in stress, tails dominate | Note spread-regime per entry |
| Visible book = true depth | Displayed depth is a fraction of interest | Cite off-venue share per statistic |
| Single venue | Fragmented routing across dozens of venues | Mark entries "single-venue assumption" |
| Zero latency | Speed measured in microseconds with fat tails | Acknowledge omitted latency variable |
Working Through the Material
- Collect a routing statistic, a speed benchmark, and an off-exchange share figure — three measurement species, three caveats each.
- For each, record window and venue scope before the number itself.
- Map each statistic onto a simulator default it challenges — build the personal table shown above.
- Annotate an existing journal's fills against the table: which of your 'executed' orders would have faced queue, spread, or latency risk?
- Re-derive quarterly: market-structure stats drift, and stale citations are a silent journal defect.
Simulation Applications
The lecture's exercise retrofits a simulated order book with a queue model: students replay a past journal's fills under probabilistic queue rules and count how many 'fills' become 'missed' — the single most disillusioning and instructive exercise in the course.
A second exercise journals venue assumptions: for each simulated fill, students write the venue, queue, and spread regime a real fill would have required, then flag which of those the simulator actually provided (usually: none).
The exercises converge on a durable habit — every simulated fill carries a 'would this have filled?' field, and honest journals answer it.
Review Checklist
- Does every market-structure statistic carry window and venue scope?
- Are simulator fills annotated against queue, spread, and latency reality?
- Is the 'touched = filled' assumption flagged wherever used?
- Are citations version-stamped against drift?
Critical Notes
Market-structure research is the genre where educational simulation and real markets touch most closely — which is exactly why the boundary must be policed hardest here. The statistics are real; the fills they contextualize are not.
The lecture's closing argument: you do not need to simulate queue priority correctly — many engines can't. You need to journal that yours didn't, so your reviewers can discount your fills accordingly.
Frequently Asked Questions
The *genre* is real — routing, speed, and off-venue data are published by regulators — but the figures in the scenario are teaching examples. Any statistic you journal should carry window and venue scope.
Because real fills are probabilistic: your limit order competes with orders ahead of it. Most simulators display 'touched = filled,' skipping the single variable that decides whether a real fill would have happened.
A mandatory journal question per simulated fill: what queue, spread, and latency regime would a real order at that price have faced? Honest entries answer it; the simulator can't.
Because medians conceal tails. Quoting 'median execution speed' hides the stress regime where fills actually matter — the lecture journals distribution tails, not averages.
Review Note
Market-structure stats are the best available estimate of what your simulator isn't modeling — journal them as the gap.
Continue Exploring
High Roller ROLR Beta Testing
Laboratory evaluation of simulated volatility options architecture, behavioral pacing stress-testing, and sandbox risk variables.
NRF and Gartner 2026 Trends
Examines predictive artificial intelligence frameworks, automated supply metrics, and macroeconomic retail commerce shifts for paper simulations.
Charles Schwab Crypto Access Impact
Objective academic assessment evaluating the systemic institutional launch of expanded digital currency gateways across retail portals.