Skip to short-horizon research
Short Term Trading StrategiesCondition / Request / Position

Research and assessment method

Record date, timezone, venue calendar, entry rule, maximum hold, exit policy, sizing and cost convention. A same-day clock example is not a substitute for a date-aware exchange schedule.

Original editorial revision: 9 October 2026. Expanded edition: 14 October 2026.

Educational execution planning. Invented costs and clocks are assumptions, not forecasts, broker instructions or a measured strategy record. Check the exact venue, instrument and account rules.

Plan before the session

Record date, timezone, venue calendar, entry rule, maximum hold, exit policy, sizing and cost convention. A same-day clock example is not a substitute for a date-aware exchange schedule.

Journal each state transition

Keep a plan, submission, acknowledgement, partial fill, cancel request, cancel acknowledgement and final fill as distinct records. Reconcile gross and net money with the quantities actually filled. Never double-count slippage already present in recorded execution prices.

Keep evaluation honest

The public tools illustrate deadlines and cost arithmetic. They do not fetch market quotes or guarantee orders. Software tests verify the calculations; an independent strategy study needs its own untouched population and execution evidence.

Library

Short Term Execution Workbench

Inspect decision timing, eligible quotes, duplicate events and position ownership for strategies whose holding window is shorter than a full market session.

Freeze the opportunity definition

A short-horizon study needs an eligible instrument population, source resolution, availability clocks, prior rule and handling of missing decisions. Keep no-trade, missed and rejected candidates in their declared denominators rather than retaining only completed winners. Entry ranking stays on raw causal boundary recovery. Freeze identity before one predeclared exit; later financial measures can assess that object but cannot choose its entry. The added cases are authored eligibility exercises, not a candidate grid, calibrated model or independent sample. An actual instrument task remains incomplete until its canonical factory and master requirements are met.

Reconcile quantities before net amounts

Requested quantity, acknowledged quantity, unique executed fragments and remaining position are distinct fields. A cancellation request is not a closing execution. The new lens starts from a declared original position and two fragment amounts; it does not simulate how they filled. Prices already determine its price change, so no second spread/slippage charge is added. All incurred fee cash is included in the snapshot total. Marking a remainder provides a stated valuation, not a confirmed liquidation. This preserves a useful amount contract without fabricating broker evidence or choosing a new partial-profit exit.

Use correctness evidence for its own claim

A pinned decimal.js engine calculates declared price, fee and R amounts. Independent Python Decimal fixtures test its equations and bounds. The existing turnover budget and same-day clock are checked separately under their original contracts. Agreement shows software consistency, not financial success, a fill guarantee or a completed strategy master. Setup classification likewise tests declared comparisons rather than a market model. Actual economic evidence needs original source data, quantities, fee vintages and an untouched eligible evaluation period, with the true observation-window research fill kept distinct from actual broker execution.

Filled setup and position references