Skip to main content

​
Why data coverage matters

Backtesting results are only as good as the data they are based on. Understanding data coverage helps you:
  • interpret results correctly
  • avoid false confidence
  • understand strategy limitations
A strategy can be logically correct but statistically misleading if data coverage is poor.

​
What data Trinigence uses

Trinigence backtests use historical market data that includes:
  • Open
  • High
  • Low
  • Close
  • Volume (when available)
Data is sourced from:
  • supported crypto exchanges
  • standardized candle feeds
  • normalized historical datasets

​
Market coverage

Data coverage depends on the selected market. Examples:
  • Major pairs (BTC, ETH) → long, deep history
  • Newer or low-liquidity pairs → shorter history
If a market has limited data:
  • the backtest range is automatically constrained
  • Trinigence clearly shows the available period
You cannot backtest before data exists.

​
Timeframe coverage

Coverage also depends on timeframe. Typical behavior:
  • higher timeframes → deeper historical coverage
  • lower timeframes → shorter available history
Example:
  • 1D data may go back many years
  • 1m or 5m data may only be available for recent periods

​
Backtest date range

When you select a backtest range:
  • Trinigence uses the exact available candles
  • missing data is not fabricated
  • partial candles are excluded
If the requested range exceeds coverage:
  • the range is clipped
  • the effective range is shown explicitly

​
Gaps and missing data

In some cases, data may contain gaps due to:
  • exchange outages
  • maintenance periods
  • delisted pairs
  • low-liquidity periods
Trinigence:
  • detects gaps
  • handles them deterministically
  • surfaces warnings when gaps are material

​
Volume availability

Volume data:
  • is available for most crypto pairs
  • may be missing or unreliable on some markets
  • is ignored if not explicitly used in logic
If your strategy does not reference volume, volume quality does not affect results.

​
Data normalization

Before backtesting, data is:
  • normalized
  • aligned to timeframe boundaries
  • cleaned for consistency
This ensures:
  • indicators behave correctly
  • comparisons across periods are valid

​
How data coverage affects metrics

Limited data can:
  • inflate win rate
  • hide drawdowns
  • exaggerate performance
Best practice:
  • test across the longest meaningful period
  • validate behavior in multiple market regimes

Backtesting metrics overview

Learn how data length impacts metrics.

​
Common misunderstandings

Data availability varies widely between symbols.
Short periods rarely capture full market behavior.
Lower timeframes naturally have less historical depth.

​
Best practices

  • Always check the effective backtest range
  • Prefer longer data windows when possible
  • Test multiple timeframes
  • Be skeptical of extreme results on short data

How backtesting works

Review the full backtesting process.

Metrics overview

How results are calculated.

Common backtest pitfalls

Avoid misinterpretation.

Trade history & logs

Inspect individual trades.

Strategy structure

Understand what is being tested.
Good data reveals truth.
Poor data creates illusions.