Skip to forex research
Simple Forex StrategiesRules / Quote / Account

Test Forex Sessions and Rollover Costs

Define session clocks, news exclusions and holding costs before comparing a rule's historical results.

Original source review: 8 October 2026. Numerical scenarios are fictional; no broker account or paid signal feed was tested. Expanded edition: 14 October 2026.

Write a Clock That Survives the Calendar

A session label such as London open is incomplete without a timezone and a daylight-saving policy. Record both the named local zone and the stored UTC timestamp. Do not hardcode a permanent UTC offset and assume it represents a local opening hour throughout the year. When two regions change clocks on different dates, their overlap can move.

For an experiment, define the eligible decision times and the source of the session calendar before counting setups. A bar labelled 09:00 may indicate its opening or closing time depending on the feed. Save that convention. Comparing two datasets without aligning those meanings can turn a supposed execution improvement into a one-bar timing error.

Choose the Event Policy Before the Event

If the rule avoids scheduled announcements, save the calendar source, publication version, timezone and exclusion window. A calendar downloaded later may contain revised values that were not available at the original decision. Separate the scheduled event time from the actual release and from the arrival of a data feed.

An exclusion is a model decision, not an opportunity to remove historical losers after inspection. Keep every excluded setup in the journal with its reason. If the original event schedule cannot be reconstructed, label the limitation and avoid presenting the filtered sample as an exact causal simulation. You can study an unfiltered reference rule without pretending the missing filter was known.

Measure the Holding Clock Separately

A trade entered shortly before the broker's financing cutoff may incur a charge even though it was held for few hours. A longer intraday trade that closes before the cutoff may have different treatment. OANDA US's cited page identifies 5 p.m. ET for its own product; that is not a universal forex broker rule. Save your broker, legal entity, instrument, cutoff and rate convention.

For fictional arithmetic, suppose $20,000 of account-currency notional is charged an annualised 5% using a 365-day basis for one charged day. The simple cost is about $2.74. That is not a live rate, and actual charge-day multiples, conversions and holiday treatment may differ. A credit on one side of a pair does not mean the other side receives its exact opposite.

Compare Like-for-Like Outcomes

Keep entry logic, exit logic, risk reference and instrument fixed while studying a session filter. Publish the eligible count and unresolved count, not just the best session's average gain. Different sessions can have different spreads and data gaps; do not apply a single convenient spread if the hypothesis depends on those differences.

Use a later period not repeatedly inspected during filter selection. If the chosen interval fails there, record the failure instead of moving the session boundary and reusing the same period as untouched evidence. A session chart can describe historical observations, but it does not establish that the timing advantage persists or is executable in your account.

Build a Session Journal

Each row should retain the local zone, UTC decision time, calendar policy, observed bid and ask where available, entry eligibility, exit clock and financing exposure. Record skipped days and platform outages too. The objective is to explain why two otherwise similar setups received different treatment.

Finish with a short statement of uncertainty: which calendar was reconstructed, which rates were assumed and which fills were references rather than executions. This site has not performed an empirical session backtest or measured a broker's spread distribution. The lesson supplies a test design and a worked holding-cost calculation, not a claim that a particular hour is profitable.

Primary Sources

Forex Base-Unit Sizing Worksheet

Continue the Research

Cash signs do not identify a funding calendar

A declared financing amount of negative 1 means subtract one quote-currency unit in the new cash packet; positive 3 means add three. No annual rate, day count or universal rollover calendar is inferred from those amounts. A real study needs the actual applicable final financing record, account lane, holding interval and conversion event. A session label should retain its named zone and UTC mapping rather than a fixed offset assumed to hold across every date. The new timezone and financing-vintage guides isolate those requests without changing the original protocol.

The next record to acquire

Preserve the original product, source version, availability, eligible population and exact cost/quantity basis for the chosen question. An official mechanism reference does not authenticate a paid account, complete channel archive or current instrument entitlement. Failed variants and unresolved observations remain part of the study. The linked workbook contains authored examples, not extra independent evaluation data. A genuine instrument task must still complete its authorised factory and independently verified master; this educational addition neither promotes a model nor changes native exit ownership or the established research-fill convention.