Build directional bias from Bias Score: composite per instrument

Bias Score is an aggregator that combines multiple positioning + macro signals into a single directional score per instrument. The goal isn’t calling tops/bottoms, but answering one question: “if I enter now, is my trade direction aligned with big money and macro flow?”

A common misconception: Bias Score is not a fixed 5-component formula across all instruments. Each instrument has its own bespoke set of confluence signals — what’s relevant for USD/JPY (BoJ stance, real yield differential) differs from what’s relevant for WTI Crude (OPEC compliance, EIA inventory, Cushing storage). The score is a weighted polarity blend over the signals that apply to that asset.

Panel layout

• Header: total assets covered + polarity filter (ALL / + BULLISH / − NEUTRAL / + BEARISH) + search (find a specific asset).
• Sort: by Score (strongest bullish first, or switch to strongest bearish).
• Per-asset card:
  • Asset name + ticker (e.g. “USD/IDR · Australian Dollar / US Dollar”).
  • Polarity badge: BULLISH (green) / NEUTRAL (yellow) / BEARISH (red).
  • Bias Score (number): magnitude of directional bias.
  • Conviction bar (%): how many of the confluence signals point the same direction as the score.
  • Confluence signal list: 4–6 rows with signal label + weight contribution (green = supports score direction, red = fights it).
• Click a card → drill into the breakdown (per-signal chart, 6-month score history, etc.).

Asset coverage

Coverage isn’t limited to FX majors. It includes:

• FX major + crosses: EUR/USD, GBP/USD, USD/JPY, AUD/USD, USD/CHF, NZD/USD, USD/CAD, etc.
• EM FX: USD/IDR, USD/MYR, USD/INR (selectively, depending on data availability).
• Commodities: WTI Crude, Brent, Gold, Silver, Copper.
• Indices: SPX, NASDAQ (selectively).

Confluence signals per asset are domain-specific:

Asset	Example confluence signals
USD/JPY	COT JPY short positioning, BoJ policy stance, US 10Y nominal yield, Real yield differential
USD/IDR	COT AUD net position (regional EM FX proxy), Iron + Copper trend (Indo export), China credit pulse, USD trend
WTI Crude	COT Money Manager net long crude, OPEC+ compliance, EIA crude inventory change, Cushing storage trend, USD trend
Gold	COT Money Manager net long, real yield 10Y, DXY, central bank reserves accumulation

The takeaway: always click the asset to see the breakdown — don’t assume “5 fixed inputs”. It varies per asset.

Workflow: daily scan (3 minutes)

1. Open Institutional Intel → tab Bias Score.
2. Click the + BULLISH filter. This shows every asset with a composite bullish bias. The list is sorted by score (strongest at top).
3. For the top 3 BULLISH:
   • Check the conviction bar. Conviction > 70% = most signals aligned (5 of 6, etc.). High score with low conviction (40–60%) = score pulled by 1–2 extreme signals, not consensus → trade with smaller size.
   • Click the card → open the breakdown. Identify the 1–2 strongest signals driving the score. If you agree with the logic (“BoJ really is dovish”, “OPEC compliance really is tight”), trade with more confidence.
4. Repeat for the + BEARISH filter.
5. For − NEUTRAL, skip: no institutional edge, let daily technicals dictate.
6. Treat the output as a directional filter (long-only / short-only candidates) for the day, not a direct entry trigger.

Reading confluence signals

Each row in the breakdown shows:

• Label: signal name (e.g. “Real yield differential”, “OPEC+ compliance”).
• Direction: + (supports bullish bias) or − (supports bearish bias).
• Weight contribution (visual bar): how much this signal contributes to the final score.

What to use for conviction:

• 5–6 signals aligned = strong consensus. Score is reliable.
• 3–4 aligned, 1–2 fighting = mixed but tilted. Score reliable with a caveat — examine the fighting signals; there may be a developing counter-thesis.
• 2 aligned, 3 neutral/fighting = weak score, barely actionable.

Common pitfalls

• Using Bias Score as an entry trigger. This is a directional filter, not a trigger. Entries still come from daily technicals/fundamentals. Bias Score answers “may I long?” not “long now?”.
• Trusting the score without examining the breakdown. Score 80 with 2 extreme + 4 neutral signals is far weaker than score 65 with 5 aligned signals. Always click the card to confirm consensus.
• Assuming the same formula across all assets. USD/JPY and WTI Crude have totally different confluence signals. The composite score is comparable, but the drivers aren’t — learn the relevant ones for the asset you’re trading.
• Refreshing too often. Signals like CFTC COT update weekly (Friday evening ET). Signals like policy stance update per meeting/speech. Hourly refresh won’t move the score meaningfully.
• Ignoring stance differential in FX. Many users see “Fed hawkish = USD bullish” without checking the counterparty CB. ECB more hawkish than Fed = EUR/USD can rally even as Fed hikes. Bias Score already accounts for the differential, but make sure you understand the logic when reading the breakdown.
• Trading low-conviction instruments at full size. Bias Score conviction 45% = 1 of 5 signals aligned, the rest mixed. This setup isn’t high-conviction; position size should reflect that (50% size, or skip).
• Forgetting score timeframe. Bias Score reflects swing positioning (days to weeks). Not relevant for intraday scalps; can go stale for multi-month positions (macro shift before the score updates).