BitThor surfaces three audit values on every completed simulation run — Signal Impact Score, SignalName, and SignalConfidence. This article defines each field, explains how it is calculated, and clarifies what it does and does not tell you about your strategy.
Signal Impact Score (SIS) is a proprietary diagnostic and educational visualization metric derived entirely from historical simulation data. It summarizes the weighted average confidence of the top recorded buy and sell signals on a normalized 0.000 to 1.000 scale. SIS is calculated by grouping recorded exits by signal family, weighting each group by its share of realized P&L, and averaging the associated confidence values. It is not a prediction of future performance, does not guarantee results, and is not a recommendation to trade.
SignalName identifies the specific technical indicator or rule that triggered a trade exit in the simulation. Each entry in SimulatedTradesLog carries the SignalName that closed the position, allowing BitThor to attribute realized P&L back to individual signal families. SignalName values are derived from the bot's configured indicator set and reflect historical backtesting behavior only — they do not indicate how any signal will perform in live market conditions.
SignalConfidence is a normalized value between 0.000 and 1.000 that describes the historical strength of the indicator reading that triggered a trade. A higher confidence means the indicator produced a strong reading at the time it fired in the simulation; it does not mean the trade was correct or that a similar reading will produce the same outcome in the future. SignalConfidence is used as a weighting factor when computing Signal Impact Score.
After a completed simulation, BitThor uses the read-only endpoint GET /api/v1/simulation/signal-breakdown/{sessionId} to populate the Simulation Summary Card's attribution layers. This endpoint returns the signal weights behind the run so you can inspect the raw component contributions that formed the final P&L and the educational Signal Impact Score summary.
Input: the sessionId from a completed simulation. Output: buy-side and sell-side weighting arrays, exact SignalName and SignalConfidence values, capped top-signal rows, total session P&L, signal diversity, and the historical weighting data used to explain the final result.
Disclaimer: All definitions and metrics described on this page are derived entirely from historical simulation data and user-defined strategy parameters. Past performance does not guarantee future results. Cryptocurrency trading involves substantial risk of loss and is not suitable for every user. Signal Impact Score, SignalName, and SignalConfidence are proprietary educational and diagnostic visualization metrics — they do not predict future market performance, constitute investment advice, or constitute a recommendation to trade. BitThor provides software tools only and does not provide investment advice. Users should only trade with capital they can afford to lose and should always conduct thorough due diligence.