Knowable time · fixed rules · untouched holdout

ES Opening Range Strategies: A Beginner's Test Protocol

By Kyle ParrottUpdated August 27, 2026

A 15-minute opening range is not known at 9:31 a.m. New York time. Its high and low become final only after the declared interval ends, the last eligible message arrives and the system finishes its calculation. Any backtest entering earlier with the completed range has used future information.

Define the range Run the holdout

Start: Named session event
End: Range becomes knowable
Entry: Executable after latency
Guaranteed pattern: None

Write a machine-readable opening event

Define the Range and Its Availability Timestamp

The normal NYSE core session runs from 9:30 a.m. to 4:00 p.m. ET, while ES trades on a broader CME schedule. If a study anchors to the cash open, say so. Verify each date for holidays and shortened sessions instead of assuming every weekday is normal.

Fixed-time opening rangeORH = max eligible ask/trade high; ORL = min eligible bid/trade low over [T_open, T_open + k)Choose quote or trade inputs, interval inclusion, correction handling and k before running the study. The decision cannot occur before the range-close message is processed.

Field	Required choice	Common hidden ambiguity
Clock	UTC plus America/New_York timezone identifier	Hard-coded ET offset around daylight changes
Session event	Official cash core open or another named event	Platform “RTH” template with different boundaries
Input price	Trades, quotes or executable side	Last price used as if both sides could fill there
Duration	One preregistered k or a declared comparison set	Choosing 5, 15 or 30 minutes after seeing profit
Availability	Final source message + computation + order latency	Entering on the same bar that defines the boundary

Use the NYSE session guide for the normal cash core boundary and the current CME trading-hours and holiday calendar for ES availability. Both schedules can have exceptions.

Each variant asks a different question

Choose One Primary Opening-Range Test

Do not blend breakout, failed-breakout and range-reversion rules into one flexible narrative. Give each a complete entry, exit, invalidation and no-trade definition, then count every version attempted.

Breakout test

Accept beyond a boundary

Define the minimum executable distance beyond ORH or ORL, confirmation clock, order type, latency and maximum allowed spread. A price touch alone need not create an entry.

Failed-breakout test

Return inside after extension

Declare how far price must extend, how re-entry is confirmed and which later price is truly available. The later failure cannot be known at the first break.

Range-reversion test

Reject an extreme

Specify the reaction evidence, target, stop and deadline. Do not assume the midpoint or opposite boundary must be reached.

Opening range width is a feature, not a verdict.

A narrow or wide range can be normalized by price or a pre-open volatility measure, then tested. It does not automatically imply expansion, reversal or a valid stop distance.

Compare like openings

Condition on Gap, Event, Volatility and Liquidity

The cash open sits inside a longer information process. Separate routine opens from scheduled-release days, large overnight repricing, holiday sessions and contract-roll transitions. Register categories before inspecting outcomes.

Control	Available before entry	Why retain it	Invalid shortcut
Gap state	Declared prior reference and cash-open benchmark	Separates overnight repricing from opening movement	Mixing settlement and cash-close gaps
Scheduled event	Official publication calendar and timestamp	Separates routine and information-shock regimes	Assigning direction from the calendar
Volatility	Pre-entry returns/range state only	Scales opening-range width and risk	Using full-day range to classify the open
Liquidity	Spread, depth and quote integrity at decision	Determines whether a theoretical break is executable	Using volume as capacity
Contract/session	Dated contract, roll, holiday and schedule flags	Prevents unlike markets entering one sample	Using a continuous chart without audit fields

The official BLS, BEA and Federal Reserve calendars support scheduled-event labels. Preserve the announced timestamp and later revisions; do not invent a fixed ES reaction.

Turn a line crossing into an order simulation

Model Entry, Protection and Exit on Executable Prices

For every candidate, replay the order at the first eligible decision time. Use the correct side of the book, visible capacity, partial fills, acknowledgement delay and dated fees. A stop is an instruction, not a guaranteed price.

Before entry

Permission gate

Range complete and data valid
Event and roll state eligible
Spread/depth inside limits
Whole-contract risk within policy

At trigger

Execution branch

Order side and type fixed
Latency applied
Queue or sweep modeled
Reject if capacity disappears

After fill

Path accounting

Partial quantity retained
Stop gap-through allowed
Exit uses executable side
Fees and shortfall charged

Hypothetical risk inputmodeled dollars at risk = contracts × (stop distance × ES point value + stressed slippage) + feesThis is an arithmetic example, not a result. Pull the current point value from the canonical mechanics page and allow the whole-contract answer to be zero.

A strategy earns a claim only after a fair test

Freeze, Hold Out, and Retire

Split chronologically. Develop the range duration, trigger, filters and exits on training data. Lock them before one untouched forward evaluation. Report gross and net results, rejected signals, uncertainty and every attempted variant.

Leakage

The final range, day type or full-day volatility enters before it was knowable.

Result: invalid study

Execution failure

Price-path performance disappears after spread, depth, latency, slippage and fees.

Result: no tradable edge

Holdout failure

The locked rule misses its preregistered forward threshold.

Result: retire, do not retune

The opening event, duration and inclusion rules are immutable in the holdout.
Range availability occurs before every simulated order.
Gap, event, volatility, liquidity and roll controls use only prior information.
Outcomes include no-trade, rejected, partial-fill and censored states.
Performance includes all costs and an adverse slippage stress.
The research log preserves losing variants and parameter searches.

Research status as of August 27, 2026

No original ES opening-range breakout, failed-breakout or reversion result is reported. This page publishes a protocol, not a win rate, probability or trading edge.

Sources, methods and editorial disclosure — reviewed August 28, 2026

NYSE trading-session FAQ for the normal U.S. cash core-session boundary.
CME E-mini S&P 500 product page and CME trading-hours calendar for current ES and schedule context.
CME historical futures and options data catalog for trades, quotes, depth and order-level data categories.
BLS release schedule, BEA release schedule and FOMC calendar for scheduled-event timestamps.
U.S. DOT uniform-time guidance for daylight-saving and timezone controls.

Sources were reviewed August 28, 2026. All opening-range variants and thresholds are proposed test definitions; no original result is claimed.