Every trading signal that Apex1819 generates carries a single headline number: the Alpha Score, a composite between 0 and 100 representing the aggregate strength of the opportunity. This number aggregates multiple independent analytical dimensions, each contributing a sub-score, that adapts to current market conditions. It is the primary metric by which the AI decides whether a signal is worth considering for execution.
The Alpha Score is designed to solve a specific problem: how do you combine many different analytical signals into a single actionable number that reflects both the strength and the breadth of the opportunity? The answer is a regime-weighted composite that adapts its weighting as market conditions change.
What the Alpha Score represents
The Alpha Score is not a prediction. It is a measurement of signal agreement across independent analytical systems. A score of 85 does not mean the trade will be profitable — it means multiple different analytical lenses, each looking at a different aspect of the opportunity, are pointing in the same direction with high collective confidence. When those lenses disagree, the score stays low regardless of how strong any single source looks.
ALPHA SCORE
A 0–100 composite score measuring the aggregate signal strength of a trading opportunity across multiple independent analytical dimensions. Weights across dimensions shift dynamically with the current market regime. Scores below the quality threshold are not traded. Scores above 85 are rare and represent maximum-conviction opportunities.
This design is deliberate. A single-source trading system — one that fires purely on RSI or purely on news sentiment — is vulnerable to the pathologies of that one source. RSI gives false signals in trending markets. News sentiment misses positioning data. The composite is designed to be robust to any single source being wrong, because the others have to largely agree before the score moves into tradeable territory.
How multiple signals combine
A proprietary scoring system combines all analysis into a single quality score. The analytical dimensions that contribute to every Alpha Score include: technical analysis, AI price forecasting, news sentiment, regime context, options flow, and additional analytical dimensions. Each brings a different perspective on the opportunity.
Each dimension's raw output is normalized to a 0–100 scale before it enters the composite. The weights are not fixed — they shift based on the current regime. The system emphasizes different analytical dimensions depending on what has historically worked best in the current type of market environment. The weights evolve over time as the learning system identifies which dimensions have been most predictive in each regime.
How the score adapts to market conditions
In trending markets, technical setups and momentum are more reliable — so they carry more influence. In volatile, disorderly markets, technical patterns lose predictive power — so the system shifts emphasis to other analytical dimensions. The system is not abandoning technical analysis; it is correctly recognizing that technical signals lose predictive power in disorderly markets.
These weight shifts are stored in the system's configuration and updated periodically by the learning system based on actual accuracy over the preceding period. They are not static values encoded in the codebase — they evolve. Over time, the weights for each regime reflect what has actually been accurate in that regime, not what a researcher guessed would be accurate.
What scores of 50, 70, and 90 mean in practice
Score ranges have explicit operational meanings. Below the quality threshold: the signal is not traded, period. That range means dimensions are disagreeing or the best-case reading is still marginal. Between 50 and 70: the signal is eligible for execution, but the compliance engine will apply its normal position-sizing logic and the position size will be at the lower end of the range. These trades exist but are conservative in size.
Between 70 and 85: a strong signal where broad agreement is present across most dimensions. These represent the majority of executed trades by volume. Above 85 is rare — occurring perhaps a few times per month — and represents moments of near-total agreement across all analytical dimensions in a supportive regime. When a 90-plus signal occurs, the system may approve full position sizing, and the trade explanation will reflect the strength of the conviction clearly. Chasing 90-plus signals is not a strategy; they happen when they happen.