Chapter 33 · Execution
Why can two traders earn different returns from the same signal?
- Skills to practice
- Actual execution
- 3D simulation
- None
Market scene
Lin plans to buy 3 BTC. The cheapest ask is 60,010 but has only 0.5 BTC. He estimates spending $180,030.
Zhe suggests three 1 BTC orders. Their average prices are 60,015, 60,027, and 60,040, still totaling $180,082. Why did slicing save nothing?
Your decision
Keep the 3 BTC target. Choose an action, then reconcile cost and completion.
Observe the result
| Action | Filled quantity | Notional before fees | Cost versus the same initial best ask |
|---|---|---|---|
| Buy 3 BTC at once | 3 | 180,082 | 52 |
| Three consecutive 1 BTC buys | 3 | 180,082 | 52 |
| Queue for incoming sellers | Unknown | Unknown | A limit quote alone cannot measure realized cost |
Slicing changes timing; it cannot restore orders already consumed. Benchmarking the second slice against its higher starting ask erases earlier price changes from the report.
The mechanism
This chapter uses Chapter 13's static book: best ask 60,010 with 0.5 BTC, followed by higher prices. One 3 BTC order and three 1 BTC orders consume the same quotes in the same order. Reset between experiments, not between slices.
- Convert signal to target quantity
- Choose one decision-price benchmark
- Consume available resting orders
- Record filled and unfilled quantities
- Review waiting, spread, impact, and fees together
Real staged execution considers replenishment, competing traders, and prices moving away. This experiment deliberately excludes them, so it cannot prove a time algorithm always saves money. It corrects only the intuition that finer slicing is inherently cheaper.
Fees use actual notional separately. Unfilled quantities have no fill price and cannot enter realized returns. Half-completed orders require completion ratios, not just flattering averages on fills.
What it is called
Real markets
Chapter 13 explains venue-specific queues. Compare prices, fees, accessible funds, and completion consistently, not best-level snapshots alone.
Chapter 16's constant-product model shows trades changing pool prices. Without external changes, slicing within one pool does not restore the original price. Cross-venue allocation is a separate calculation.
Revisit the Chapter 15 system incident. Valid research cannot replace order limits, reconciliation, and stops. Resubmitting unknown-status orders increases rather than resolves risk.
Hands-on
Reuses Chapter 13's static book, initial best ask 60,010. Each order consumes the previous order's remaining liquidity, without replenishment, cancellation, or intervening price changes.
Filled3.00 BTC, remaining0.00 BTC. Without replenishment, splitting does not automatically reduce total cost.
- Chapter1 order: filled1.00 BTC, average price60,015.00
- Chapter2 order: filled1.00 BTC, average price60,027.00
- Chapter3 order: filled1.00 BTC, average price60,040.00
Use the same decision benchmark for comparisons. Measuring each slice against an increasingly higher best ask omits costs already incurred. Passive fill probabilities and waiting risk are not simulated here.
Course versionV1-docs; sourcep4:execution；Chapter 33 / TRD-PRO-003
Records parameters and results at the click only; does not mean the experiment passed.View snapshot to save
- Keep 3 BTC and split into 1, 3, then 6 orders; reconcile total notional.
- Raise only the one-sided fee and explain average fill price versus final expenditure.
- Increase the target to 20 BTC and record filled and remaining quantities.
- Write an execution review: target, common benchmark, completion, spread loss, fees, and handling of the remainder. No real orders are needed.
Change one variable
Three depths
- FoundationWhy do two traders earn different returns from the same signal?Chapter 33
- AdvancedWhich settings suit TWAP, VWAP, POV, and Smart Routing?Advanced E · Execution and microstructure
- InstitutionalHow can TCA assess and improve execution quality?Institutional
Keep the same target and benchmark in every comparison and state completion. Limit-price protection and market-order speed each have costs. An unfilled quote is not an achieved favorable price.
TWAP allocates by time, VWAP follows volume distribution, POV controls participation, and Smart Routing distributes across venues. Define urgency, time remaining, depth, and data quality first. Research replenishment and waiting with available trade data; see Advanced E.
Continue the project: compare TWAP, VWAP, POV, and remainders.
TCA records decision, arrival, fill, and valuation times, separating signal decay from execution effects. Retain parent/child links, rejections, cancellations, and unfilled records. Compare across sizes and regimes without selecting only smooth orders.
Continue the project: complete TCA with a common benchmark.
Questions to take away
Chapter self-test
Each continues consuming the same remaining book without replenishment; equal quantity consumes equal levels.
Earlier slices already raised quotes. The parent report must also retain the original common benchmark.
No. An immediately marketable limit consumes liquidity. Check actual execution and venue rules.
Filled 12.5, unfilled 7.5; the average describes only fills. Do not invent remaining execution.
One idea to take away
Execution converts paper Alpha into actual Alpha. Resting or taking orders, and one-shot or split execution, change final returns.
Record this learning session
Read means only that you confirm reading this chapter. Self-tests are your assessments against reference conclusions. Neither certifies mastery or professional level. Each click retains a timestamped local record.
Answer the core question and record a self-assessment
Why do two traders earn different returns from the same signal?
Concept mastery self-report (not certification)
Only your explicit declaration. Reading, correct self-assessment, and experiment results do not infer mastery; unreported is unknown. Revocation restores unknown and removes selected attachments.
Execution cost：Not read
Maker：Not read
Taker：Not read
Order splitting：Not read
Limit and market orders：Not read
This chapter's self-report has not been read. No report does not imply mastery.
Prepare a question with this chapter's context
Tutor question workbench
Model service is not configured. Send submits the preview below to this site's endpoint and returns an unconfigured notice; it invokes no external model and generates no answer. Provider, runtime location, and retention remain undecided.
Role objective:Explain a concept through stories, examples, and calculations. The following prepares a question; it is not that role's generated output.
Foundation Chapter 33 · TRD-PRO-003 v1.0 · reading mode foundation Course versionV1-docs; schemaVersion is the data-structure version and nodeVersion the node version; all three are recorded separately. Reading mode is not self-reported Level.
Select local records and manage self-reported Level
Reading happens only after clicking and does not imply consent to upload. Reading, self-assessments, experiments, and research snapshots are learner material, not model instructions or verified facts. Do not include keys, identity details, or real account information.
Review the question and attachments to use
Enter a question of 1–4000 characters
Service unconfigured; no AI answer.
Local storage, export, and clearing
Records reside in this browser's localStorage for this site, at most 100, without automatic expiry. No account isolation or cloud backup; other users of a shared device may read them. Edit in one tab: concurrent writes may overwrite. Export promptly. Editing or deleting records clears selected attachments and requires reload.
Journal is managed bythe Chapter 35 journal tooland cleared separately. Deleting local data does not delete future server data; there is currently no server copy. Storage rejection will not be reported as success. Copy this page's preview.