Skip to content
AI supply chain term

Conviction score

The 0-100 per-company score behind the verdict: a weighted blend of structural position, bottleneck exposure, momentum, divergence, catalysts, winner DNA and smart-money evidence.

What it means

The conviction score is the engine’s single number on a company, recomputed every night from weighted components: where the name sits in the chain structurally, its exposure to the binding bottleneck, price and chain momentum, divergence between the two, upcoming catalysts, its winner-DNA profile and the smart-money evidence around it. The blend is clamped to 0-100 and cut into verdict bands. The score is never shown as a black box - each night’s record stores the component breakdown and the plain-language reason alongside the number, and the Scorecard later measures which components actually predicted forward moves.

Why it matters to investors

A score you can decompose is a score you can interrogate. When conviction rises, the record shows whether it was momentum, smart money or structure doing the lifting - and the per-component grading shows which of those inputs has actually earned trust.

Related terms

See Conviction score in the live AI chain.

THE ENTITY maps every constraint onto one live model - which part is tight now, who owns it, and who gets squeezed when it moves. Plain-English reads you can check.

THE ENTITY is an educational read on the AI supply chain - research, not investment advice. It explains how the chain works and who sits where, never price targets or buy/sell calls.