1
Bar interval
Open, high, low, close, and volume depend on the aggregation window.
Lesson · Backtesting and Trading Systems
Timeframes, Sessions, and Bar Construction
Define how raw trades or quotes become bars, which sessions are eligible, and how timeframe choices change indicators, signals, and fills.
Core concepts
What to understand
Keep each definition operational so it can be implemented and tested.
2
Session rule
Regular, extended, overnight, and exchange-specific sessions produce different datasets.
3
Decision clock
Define the exact timestamp when indicators update and orders become eligible.
Interactive lab
Timeframe and Session Lab
Aggregate the same intraday sequence into different bars and sessions. Observe how signals and market interpretation change.
Timeframe and Session Lab
Deterministic synthetic educational data · no live market connection
BACKTEST MODEL
Process
Define the market clock
Bar boundaries and sessions must be fixed before testing.
Choose source frequency
Use data granular enough for the execution assumptions.
Define sessions
Include only times the system is allowed to observe and trade.
Aggregate consistently
Use one timezone, calendar, and boundary rule.
Knowledge check
Check your understanding
Select one answer for each question. Feedback appears immediately.
1. Why can a 60-minute test miss an intrabar stop?
OHLC shows the range but not the order in which high and low occurred.
2. Extended-hours inclusion can change:
Session data changes price history and eligible execution periods.
3. What should determine bar boundaries?
Consistent market calendars and timezones are required for reproducibility.
4. A regular session runs 390 minutes. Aggregated into 60-minute bars it produces:
390 ÷ 60 = 6.5, so six full bars and a seventh partial bar of 30 minutes — the lab prints the division and names the partial bar.
Practical rules
- Store timestamps in a clear canonical timezone.
- Use exchange calendars and session definitions.
- Use lower-frequency bars only when their information supports the fill model.
Common mistakes
- Assuming all daily bars close at the same time.
- Mixing extended-hours signals with regular-hours fills without rules.
- Using daily OHLC to infer an exact intraday sequence.
References
Official reading
The labs are simplified simulations. These sources cover historical performance, data, costs, or execution limitations.
CME Group · Historical Market Data
Describes historical market data used for market analysis and strategy validation.
CME Group · Trading Simulator
Provides a practice environment for testing trading decisions without live capital.