Bidirectional Severity Scoring for Trade Events: How a -4 to +4 Scale Captures Disruptions and Restorations

Trade disruption intelligence is only as useful as the scoring framework behind it. Binary flags — disrupted or not — collapse the complexity of real-world trade events into a format that cannot support portfolio-level risk decisions. A port strike and a full naval blockade are not the same event. Neither are the lifting of sanctions and the partial reopening of a damaged terminal. Severity scoring must encode both magnitude and direction to be operationally useful.

Disruptis uses a bidirectional severity scale from -4.0 to +4.0, where negative values represent supply chain restorations and positive values represent disruptions. This structure reflects how trade corridors actually behave: they degrade and recover, often in overlapping sequences that unidirectional scales cannot represent. Understanding how this scale works — and why directionality matters — is foundational for anyone integrating disruption data into trading models, insurance pricing, or logistics planning.

Why Unidirectional Scales Fail

Most risk indices assign a score from 0 to some upper bound, where higher means worse. This works for static snapshots but breaks down in time-series analysis. Consider a scenario: a major container terminal suffers equipment failure (scored at, say, 3 on a 0–5 scale), and two days later partial operations resume. A unidirectional system either drops the score back toward zero — losing the restoration event as a distinct data point — or requires a separate taxonomy for "recovery events" that sits outside the primary scoring framework.

The problem compounds at the portfolio level. A trading desk monitoring 15 corridors needs to know the net disruption state across all of them. If your scoring system only counts negatives, you are forced to infer restorations from the absence of new disruption signals — a method that produces false positives when coverage gaps or reporting delays occur.

Anatomy of the -4 to +4 Scale

On the Disruptis scale, each event receives a severity score based on its assessed impact on trade flow through the affected corridor. The structure breaks down as follows:

• +1.0 to +2.0: Localized or partial disruptions — port congestion, minor weather delays, short-duration labor actions. Trade flow is impeded but not halted.
• +2.0 to +3.0: Substantive disruptions — extended strikes, targeted sanctions, infrastructure damage requiring rerouting. Freight rates and insurance premiums typically respond at this tier.
• +3.0 to +4.0: Severe or systemic disruptions — full corridor closures, armed conflict in shipping lanes, comprehensive trade embargoes. Physical commodity delivery is blocked or requires major alternative routing.
• -1.0 to -2.0: Partial restorations — reopening of one berth at a multi-berth port, easing of inspection requirements, provisional ceasefire in a conflict zone.
• -2.0 to -4.0: Full restorations or structural improvements — sanctions lifted, new infrastructure commissioned, trade agreements enacted that open previously restricted corridors.

Zero represents a neutral event — one that is classified and tagged but assessed as having no net directional impact on trade flow. For a deeper look at how raw news sources become scored, structured events, see how severity-weighted geographic intelligence is produced.

Directional Scoring in Practice

The practical value of bidirectional scoring shows up in three workflows:

Net corridor exposure. By summing severity scores across events in a given corridor over a rolling window, risk teams can compute a net disruption state. A corridor with a +3.2 blockade event followed by a -1.5 partial reopening has a net state of +1.7 — still disrupted, but trending toward recovery. This is directly actionable for freight operators deciding whether to reroute or wait. Logistics teams working with daily disruption feeds can layer this into rerouting decision frameworks.

Insurance pricing cycles. Underwriters pricing cargo or trade credit policies need to distinguish between a corridor that has been at +2.5 for three weeks and one that spiked to +3.5 and then recovered to +0.8. The cumulative exposure profiles are different. Bidirectional scoring makes that distinction computable rather than judgmental.

Trading signals. Commodity desks watching physical delivery corridors can treat restoration events as leading indicators. A -2.0 restoration on a corridor that carries 15% of global copper concentrate shipments is a supply signal that should feed directly into positioning models.

Scoring Transparency and Integration

A severity score is only trustworthy if its methodology is auditable. Disruptis publishes its scoring methodology so that downstream consumers — whether quantitative analysts building models or compliance teams documenting risk processes — can trace any score back to its classification logic and source inputs.

The scored events are delivered as structured Parquet files with severity, event type, commodity tags, geographic coordinates, and corridor mappings included. This schema, available for review in the data preview, is designed for direct ingestion into risk dashboards, trading systems, and actuarial models without manual transformation.

Bidirectional severity scoring is not an incremental improvement over binary or unidirectional approaches. It is a structural requirement for any system that claims to represent how trade corridors actually behave — as dynamic systems that break down and recover, often simultaneously across different segments. The -4 to +4 scale encodes that reality in a format that machines and analysts can both use.