1. Source hierarchy
Where practical, Collectible prioritises manufacturer documentation, official technical material, regulatory or institutional records and directly attributable primary sources. Reputable secondary sources may be used where primary data is unavailable. Unverified statements remain marked as research rather than being silently promoted to fact.
2. Production & rarity
Production totals distinguish exact model counts from combined-series totals, planned production, market-specific allocations and estimates. A rarity label is only as reliable as the production count behind it and should not be read as an investment recommendation.
3. Listings, transactions & market benchmarks
Asking prices and completed transactions are different datasets. Collectible does not treat an advertised price as proof of realised market value. The Collectible Market Benchmark is a model-market reference derived from comparable observations; it is not an appraisal of a specific chassis. Reported auction sale prices can include buyer premiums depending on the source, and the recorded price basis is retained with each observation.
4. SCI score, risk & confidence
SCI is a proprietary analytical framework intended to summarise multiple collector-market factors. SCI Score, Risk and Data Confidence are separate. SCI-1.5.0 uses source trust, observation quality, vehicle comparability, recency, sample depth, independent-source diversity and source-backed Collector Evidence G as an explicit qualitative control. Collector Evidence G can support structural collector factors only when source-backed; generic editorial copy never creates factor coverage or score points. Market-performance factors remain observation-led. Missing evidence is not negative evidence: uncovered factors are excluded from the weighted SCI calculation and reduce Data Confidence instead. Historical sale performance requires a genuine transaction history, while a single observed sale remains evidence rather than becoming a market benchmark. Scores can change when methodology, source coverage or market evidence changes and every material methodology revision receives a new version.
5. Automated ingestion & no fabricated market data
Collectible can parse factual observations from approved public market-result pages and accept structured CSV imports. Automated extraction remains reviewable by administrators and does not republish source editorial copy. Collectible does not publish invented transaction histories or generate synthetic price charts merely to fill an interface. Where verified market history does not meet the current methodology threshold, the corresponding score remains in research.
6. Automated auction monitor
The Latest Auctions feed is an editorial discovery layer that periodically observes approved public auction-result and event pages. An item appearing in that feed is not automatically admitted into the SCI calculation. Market observations enter SCI only after they satisfy the separate transaction verification, comparability and publication controls described above.
7. Corrections & versioning
Research can be revised when better evidence becomes available. Material methodologies should be versioned and dated. Users can report factual errors or conflicting primary sources to office@supercarsharing.com.
8. Independent judgement
Collector significance, risk factors and outlook are analytical research. They can contain editorial judgement and are not guarantees. Users should independently inspect a specific vehicle, its title, condition, provenance, tax position, maintenance history and legal status before any transaction.