AI & Quantitative5 min readUpdated Mar 2026

Composite Score

A single numeric score that combines multiple evaluation factors — technical strength, volume, gap, relative strength, watchlist priority, and AI conviction — to rank day trade candidates from best to worst before execution.

See Composite Score in real trade signals

Tradewink uses composite score as part of its AI signal pipeline. Get signals with full analysis — free to start.

Start Free

Explained Simply

The composite score solves a core problem in day trading automation: when you have 50+ screened candidates and can only take 3–5 trades, which ones do you pick? A single indicator is insufficient. The composite score aggregates multiple dimensions into one comparable number.

Typical components and their approximate weights:

  • Price action / technical signal: Breakout quality, VWAP relationship, RSI positioning (30–40%)
  • Volume: Relative volume vs. 20-day average; volume confirmation of moves (20–25%)
  • Gap: Stocks gapping up or down with volume get a bonus (5–10%)
  • Relative strength vs. sector: Is the stock outperforming its sector ETF? (10–15%)
  • AI conviction multiplier: The conviction score scales the composite score up or down (15–20%)
  • Watchlist bonus: User-added tickers receive a +15 point boost to prioritize monitored positions

Candidates are ranked by composite score descending. The top N candidates (configurable) proceed to the sizing and execution phase.

Why a Single Indicator Is Insufficient

Every technical indicator has failure modes that appear repeatedly in live markets. Volume alone can spike on news that turns out to be a rumor. RSI overbought conditions persist for weeks in strong trending markets. Gap-up stocks fade immediately when gaps are caused by low-conviction overnight news. A composite score is resilient to these individual failure modes because it requires multiple independent signals to align simultaneously. A stock showing high relative volume, a clean gap above resistance, RSI in the 55 to 65 zone (trending but not overbought), and sector outperformance is a far higher-quality candidate than a stock with only one of these characteristics. The composite score converts this multi-factor alignment into a single number that reflects the overall quality of the setup across all dimensions.

Component Weights and Their Rationale

The weights assigned to each composite score component reflect their historical predictive value for intraday follow-through. Technical signal quality — the strength and quality of the breakout, VWAP relationship, and momentum indicators — receives the highest weight (30 to 40%) because it most directly captures the price action thesis. Volume confirmation (20 to 25%) validates that institutional participation is present, not just retail activity. Gap magnitude (5 to 10%) provides a directional head start that momentum strategies can exploit. Relative strength versus the sector ETF (10 to 15%) confirms the stock is a leader within its sector context rather than a laggard being pushed by broad market moves. The AI conviction multiplier (15 to 20%) adjusts all scores based on the model's assessment of the qualitative narrative context surrounding the setup.

The AI Conviction Multiplier

The composite score's technical and volume components can be computed from quantitative data alone. The AI conviction multiplier is the final adjustment that incorporates context that indicators cannot capture: the quality of the news catalyst, the clarity of the narrative, the strength of the technical setup relative to the current market regime, and historical performance of similar setups. A conviction score of 80 out of 100 applied to a composite score of 75 produces a final rank of 60 (75 × 0.80), while the same raw score with a conviction of 40 produces a rank of 30. This ensures that technically strong setups with poor or unclear catalysts are ranked below slightly weaker setups with strong, clear catalysts — reflecting the reality that catalyst quality is as important as technical quality in day trading.

Watchlist Priority and Score Boosting

Composite scoring includes a structural bias toward user watchlist tickers. Stocks that a user has explicitly added to their watchlist receive a 15-point score boost before final ranking. This prioritization reflects a key advantage: a trader who has been monitoring a specific stock, understands its recent price history, and has conviction about its setup is better positioned to manage that trade than an algorithmically selected stranger. The watchlist boost ensures these pre-researched candidates are not crowded out by marginally higher-scoring candidates the system discovered through its scan universe. Users can further customize the boost amount or disable it entirely through their Tradewink preferences.

How to Use Composite Score

  1. 1

    Define Component Scores

    A composite score combines multiple factors into a single ranking. Common components: technical score (RSI, MACD, moving average position), fundamental score (earnings growth, P/E), momentum score (relative strength, price performance), and sentiment score (short interest, analyst ratings).

  2. 2

    Weight the Components

    Assign weights based on which factors are most predictive for your strategy. Example: technical 40%, momentum 30%, fundamental 20%, sentiment 10%. Test different weightings on historical data to optimize — the best weights depend on your timeframe and market conditions.

  3. 3

    Rank and Filter Using the Score

    Score all candidates and rank by composite score. Only trade the top 5-10 ranked stocks. Set a minimum threshold (e.g., score must be 7+ out of 10). This systematic approach removes emotion and ensures you only trade the highest-quality setups.

Frequently Asked Questions

Why use a composite score instead of a single best indicator?

No single indicator is reliable across all market conditions. Volume spikes occur for both legitimate breakouts and pump-and-dump setups. RSI overbought signals fail in strong trends. The composite score is robust because it requires multiple factors to align simultaneously, reducing false positives significantly compared to any single signal.

Can I see the composite score for signals in Tradewink?

Composite scores are used internally for ranking and are visible in the trade journal alongside conviction scores and final P&L. This allows retrospective analysis of whether high-scoring candidates performed better than expected.

How does the composite score change with market regime?

The composite score components are evaluated against the current market regime context. In a bearish trending regime, gap-up and momentum components receive reduced weight because breakouts have lower follow-through rates. In a choppy intraday regime, the minimum score threshold for execution is raised to reduce the number of trades taken. The AI conviction multiplier also adjusts implicitly — the model is trained to discount bullish setups in bearish regimes and vice versa.

How Tradewink Uses Composite Score

The DayTradeScreener computes composite scores for all screened tickers after gathering market data. The score is computed in `_compute_composite_score()` and combines volume ratio, gap magnitude, ATR percentage, RSI positioning, and S&P 500 heatmap alignment. After AI conviction scoring, the composite score is multiplied by the conviction ratio (conviction/100), producing the final rank. Watchlist-prioritized tickers are bubbled to the top via an additional score boost. The ranked list then feeds into the evaluate → size → execute pipeline.

Trading Insights Newsletter

Weekly deep-dives on strategy, signals, and market structure — written for active traders. No spam, unsubscribe anytime.

Related Terms

Learn More

See Composite Score in real trade signals

Tradewink uses composite score as part of its AI signal pipeline. Get daily trade ideas with full analysis — free to start.

Enter the email address where you want to receive free AI trading signals.