Overview
Overview
Performance metrics on AITA provide a structured view of how an AI agent has behaved historically under specific market conditions.
These metrics are designed to describe behavior, not to predict outcomes. They help users understand consistency, volatility, and risk exposure over time.
What performance metrics show
Performance metrics summarize the historical results of an agent’s fixed strategy and parameters.
They show:
How the strategy responded to past market conditions
The cumulative impact of gains and losses
Patterns of volatility and stability over time
Metrics do not influence how an agent operates. They are outputs, not inputs.
Backtested and live data
Performance data on AITA may include:
Backtested results, which simulate strategy behavior using historical data
Live results, which reflect real execution in current market conditions
Backtests provide context and learning. Live performance reflects real-world constraints such as liquidity, slippage, fees, and capital availability.
Differences between backtested and live results are normal and expected.
No single metric tells the full story
No single metric can fully describe an agent’s behavior.
For example:
High returns may come with high volatility
Low drawdowns may come with lower growth
Short-term performance may differ from long-term trends
Metrics should always be interpreted together and in context.
Time horizon matters
Performance metrics become more meaningful over longer time periods.
Short time frames may reflect temporary market conditions, while longer histories provide insight into:
Strategy robustness
Risk management characteristics
Behavior across different market regimes
Longevity and consistency are often more informative than recent results.
Responsibility and interpretation
All performance metrics are historical and informational.
Users remain responsible for:
Interpreting metrics correctly
Understanding their limitations
Assessing suitability for their own use
AITA provides transparency and standardized metrics, not recommendations or guarantees.
Last updated
