Data Sources & Methodology

1. Data sources and observations

RoyaleRecord uses normalized responses from official Clash Royale data interfaces when integration is enabled. It stores what it actually observes; it does not invent missing profiles, battles, rankings or statistics. An observation is a timestamped record received by RoyaleRecord, not proof of complete lifetime history.

2. Historical records and snapshots

Player and clan snapshots preserve eligible public values as they were observed. Current names, membership and progression may later change upstream. Ranking captures retain their track, location and capture time. Trophy Road and Ranked remain separate.

3. Deck and battle identity

A canonical deck identity is a deterministic hash of normalized card IDs, card roles and evolution state; card levels and battle outcomes are not part of deck identity. A battle receives a deterministic BattleHash from normalized battle time, mode and canonical participant sides. The unique hash permanently deduplicates the same battle when several player lookups encounter it.

4. Release 1 meta dataset

Population analytics use normalized, deduplicated battles with a valid BattleHash, reliable battle time, two complete one-player sides, usable canonical decks and a supported win, loss or draw result. Only modes positively classified as standard competitive 1v1 are eligible. 2v2, draft, boat battles, incomplete records and unknown or unverified modes remain stored but are excluded from Release 1 population analytics.

5. Results, draws and mirrors

Counts show wins, losses, draws and sample size. Decision win percentage is wins divided by wins plus losses; draws are preserved but excluded from that denominator. When there are no wins or losses, the percentage is unavailable. A mirror deck battle contributes to deck and card appearance statistics, but is excluded from deck matchups and card counter calculations.

6. Windows and trends

Release 1 supports 7-day, 30-day and 90-day windows; 30 days is the default public view. Trending compares the most recent complete seven UTC days with the immediately preceding complete seven UTC days. The current partial UTC day is excluded. Absolute sample change, percentage-point change and relative percentage change are distinct values.

7. Minimum public samples

• Deck statistics: 100 eligible deck appearances.
• Card statistics: 250 eligible card appearances.
• Deck matchup: 50 direct eligible matches between the pair.
• Card counter: 100 eligible opposing observations.
• Card synergy: 100 eligible shared-deck observations.

Aggregates below these thresholds may be retained internally, but RoyaleRecord does not rank or publish their unstable percentage as a qualified result. Public analytics show the window, Standard 1v1 dataset, sample size and Methodology v1 label.

8. Counters and synergies

Counter results are observed statistical associations between cards on opposing sides; they do not prove that one card mechanically causes another to lose. Synergy records how often a canonical card pair occurs on the same eligible deck and that side’s recorded wins, losses and draws. Co-occurrence and recorded results do not establish causality.

9. Limitations and version

Coverage depends on battles and profiles RoyaleRecord has actually observed. Upstream delays, missing fields, collection timing and minimum-sample rules can leave results unavailable. Release 1 uses Methodology v1. RoyaleRecord Score and subjective deck quality scores are disabled because no approved formula exists.