How do you evaluate a trading signal?

A trading signal is worth further work only if it has a plausible reason to exist, clean point-in-time data, stable behavior across regimes, realistic performance after costs, and a clear role in a portfolio. A high backtest return is not enough.

Ask why the signal should predict anything after costs. Good rationales often involve compensation for risk, liquidity provision, behavioral pressure, market structure, or persistent constraints that other participants cannot arbitrage away instantly.

Make sure the signal uses only information that would have been available at the decision time. Timestamp errors, restated data, survivorship bias, and accidental future labels can turn a weak idea into a fake edge.

The evaluation should separate signal quality from strategy quality. A signal can be predictive but untradeable, or weak alone but useful as a context layer for a larger portfolio.

A useful signal must survive fees, slippage, turnover, capacity, position limits, and portfolio interactions. It also needs to add something that existing signals do not already capture.

Before promotion, decide whether the signal ranks markets, gates another strategy, changes sizing, or triggers execution. That role determines the right benchmark and the right failure test.

Alphora's catalogue structure emphasizes signal meaning, known weaknesses, historical behavior, implementation notes, and relationship to other signals so traders can evaluate more than a headline chart.