Skip to main content

UTC anchor · local clocks · market quality

6M Trading Times: A DST-Safe Session Study

Under the 2026 time-zone rules, 14:00 UTC during U.S. daylight time is 08:00 in Mexico City and 09:00 in Chicago. In U.S. standard time, that same instant is 08:00 in both cities. A fixed “8 a.m.” bin can therefore compare different market moments across the year unless the study stores the timezone and resolved offset for every observation.

Anchor
UTC instant
Local labels
Versioned
Outcomes
Separate
Session ranking
Not reported

Define moments before sessions

One Instant Can Occupy Different Local Buckets

A session is an analyst-defined window, not an exchange finding. Store the event in UTC, then attach date-aware local labels for Chicago, Mexico City, and any other economic clock the hypothesis needs. Preserve the original timestamp and the timezone-database version used to resolve it.

Clock fieldRequired recordError prevented
UTC event timeUnambiguous instant with precision and source timestamp semanticsDifferent source clocks are sorted incorrectly
IANA timezoneNamed zone such as America/Mexico_City or America/Chicago, not “CST”Ambiguous abbreviation hides historical rule changes
Resolved offsetUTC offset calculated for that exact date and timeCurrent daylight rule is projected backward
Civil dateLocal calendar date for each economic regionRelease or holiday is joined to the wrong local day
CME trade dateExchange trade-date label stored separatelyOvernight records are assigned to the wrong futures session
Boundary ruleHalf-open intervals and policy for bars crossing an edgeOne trade or bar is double-counted

Mexico

Use the law in force for the observation date

Mexico’s 2022 Law of Time Zones established standard zones and seasonal treatment for specified northern-border areas. A Mexico City study must not inherit U.S. daylight changes merely because a vendor labels both clocks “Central.”

United States

Chicago changes offset during daylight time

NIST’s daylight-saving rules give the U.S. transition dates, including the 2026 interval. Code the rule from a maintained timezone database and retain the resolved offset.

Dated instruments, complete paths

Build the Sample Before Naming a “Best” Window

Use dated 6M contracts and a feed that contains the fields required by the outcome. A daily OHLC series cannot measure intraday spread, visible depth, queue behavior, or event-time slippage. Missing data must remain missing rather than being interpreted as quiet trading.

01

Inventory the feed

Record provider, dataset, schema, timestamp, trade, quote, depth and correction fields.

02

Audit each expiry

Check contract identity, gaps, duplicates, crossed books, stale quotes and impossible prices.

03

Assign windows

Map clean messages to UTC-anchored, half-open windows and preserve all local labels.

04

Summarize by day

Create one date-window-contract observation before estimating uncertainty.

Field acceptance is not negotiable

CME DataMine offers distinct trade, top-of-book, depth and Market by Order datasets. If a required spread, depth, or queue field is absent, the source fails for that outcome. It may remain usable for a narrower volume or range question.

No single activity score

Measure Participation, Movement, and Executability Separately

A window can have high completed volume, wide ranges, thin displayed depth, and poor market-order slippage at the same time. Calling it “active” or “best” collapses distinct questions. Publish distributions and tails for each outcome before considering any composite.

Displayed market

Spread and depth

Sample bid-ask width, time at each width, visible quantity at declared levels, locked/crossed states, staleness and missing-book intervals. Displayed orders can cancel and do not guarantee execution.

Completed participation

Trades and volume

Count contracts, trades, trade size and intertrade duration for the selected expiry. Completed volume does not reveal participant identity, motive, or unobserved liquidity elsewhere.

Price path

Range and realized movement

Freeze high-low range, absolute return, or realized-variance construction. Report direction, reversals and tails separately; a large range does not select a side.

Executable experience

Slippage and fill quality

Replay declared order sizes and latency. Measure arrival-to-fill slippage, partials, fill rate, wait, adverse selection, commissions and fees.

Primary comparison unit

Summarize the intraday path into one observation per eligible date, window, and contract. Quote messages within a window are dependent, and adjacent windows on the same date share conditions. Use day-level or longer blocks for uncertainty rather than counting every update as an independent trial.

Center
Median
Tails
Quantiles
Missingness
Reported
Direction
Separate

Calendar overlays

Separate Ordinary Windows From Scheduled Information and Holidays

Official calendars identify when information was scheduled for release. They do not establish the sign, size, or durability of a 6M response. Timestamp each event, define exclusion and event windows before measurement, and keep scheduled, rescheduled, surprise, and unscheduled events distinct.

Banco de México

Mexico policy
SourceBanxico
ControlDecision & minutes

Use the official 2026 publication calendar and actual release records.

Also identify CME maintenance and holidays from the contemporaneous CME trading-hours notices. Holiday schedules are subject to change, so preserve the version used instead of rebuilding history from the latest page.

Liquidity can split

Control the Contract Roll Before Comparing the Clock

A clock window near expiry may show weak volume or depth in the outgoing contract while participation has already moved to the next month. A front-month continuous series can silently switch at different times across vendors, creating a session effect that is actually a roll rule.

Observed weak window near migrationDo not label it a time-of-day effect until contract selection is reconciled.

Clock channel

Market participation may genuinely vary by time. Test it on the same eligible contract under a fixed window definition.

Migration channel

Volume and quotes can be split across expiries. Preserve both and apply an ex-ante selection rule.

Adjustment channel

Back-adjusted bars can change levels or returns at the splice. Execute only on dated prices and books.

Expiry channel

Termination and delivery proximity can alter eligibility. Use the canonical 6M contract guide for mechanics and roll-risk checks.

Try to move the ranking

Demand Robustness to Boundaries, States, and Tail Costs

Predeclare adjacent clock windows and economically motivated state splits. If small boundary changes reverse the ranking, confidence intervals overlap materially, or tail execution dominates the median, report no defensible ordering for that metric.

Clock guard

Shift boundaries by a declared small interval; repeat by UTC and named local-economic purpose.

Calendar guard

Repeat on ordinary days, policy days, macro-release days, holidays and transition weeks.

Contract guard

Use a second ex-ante roll rule and report outgoing, incoming and combined-liquidity views.

Feed guard

Vary missing-book, stale-quote and outlier policies; use a second source only where rights allow.

Cost guard

Increase latency, spread, slippage and size; retain missed and partial fills.

Time guard

Compare early/later years and walk forward; do not pool structural breaks into one universal hour.

Interpretation limit

A tighter historical median spread does not identify a profitable trade. A wider range does not supply direction. High displayed depth does not guarantee a fill. The only claim a completed study may make is the metric, sample, size, clock, contract, and conditions it actually measured.

Ending deliverable

Reproducible 6M Session-Study Protocol

Freeze this manifest before opening the final chronological holdout. Publication should include null or unstable results, not only the window that looks best.

  1. Question: declare one primary metric and the exact economic purpose of every window.
  2. Clock: store UTC, IANA timezone, resolved offset, local civil date, CME trade date and boundary policy.
  3. Universe: list dated 6M contracts, source files, fields, gaps, corrections and the ex-ante roll rule.
  4. Quality: freeze stale-book, crossed-book, sequence-gap, missing-quote and outlier treatment.
  5. Measures: specify spread, depth, trade, volume, movement and execution calculations independently.
  6. Overlays: timestamp official Mexico/U.S. events, DST regimes, holidays, maintenance and roll proximity.
  7. Inference: summarize at a day-aware unit, report effects and tails with block-aware uncertainty, and disclose all comparisons.
  8. Falsification: reject a universal ranking when adjacent boundaries, later dates, alternate rolls or cost stress overturn it.
Research status as of August 13, 2026

No original study is reported. No 6M market dataset was acquired or analyzed for this article. No window, hour, session, spread, depth, volume, range, slippage, or performance ranking is reported. This is a study protocol; “no defensible ranking” remains an acceptable completed outcome.

Sources, methods and editorial disclosure — reviewed August 13, 2026

Sources and methods were reviewed August 13, 2026. This page is unsponsored editorial analysis and presents a session-study protocol, not original 6M market findings.