Cards
Card tiers are relative to the catalogue: S is the top 8%, A the top 25%, B the top 50%, and C the top 75%. The score combines attack, generation, defense, support, control, and economy.
Decks
Decks are found through beam search, followed by talent optimization. The relative index compares their score with an average random deck and the best deck found. S ≥ 90%, A ≥ 80%, B ≥ 70%, C ≥ 50%. It is not a win rate.
Theoretical analysis
A 10-turn model. The displayed gameplan and metrics are simulated against a neutral opponent (4 cards with 700 HP and 10% resistance in every element). No real meta can be observed: a deck’s score is the average of its scores against nine theoretical opponent profiles (neutral, fragile, tank, six decks resisting one element), and the gap against each profile gives its strengths and weaknesses. One active talent per card (chosen before the match); conventional allied deaths on turns 6 and 10. Ability order is heuristic. Tox costs and draw probabilities are configurable assumptions. Real matches can differ.
Official card data
Game version : 1.19
Card data from September 21, 2026
Analyzed on September 22, 2026 · UTC
Contact
For any request (data error, artwork removal, suggestion or partnership), write to contact@spamgame-wiki.com.
Privacy
The site requires no account and sets no cookies of its own: favorites, comparisons and compositions stay in your browser’s local storage. Google Analytics audience measurement (pages viewed, device, approximate country) is only enabled after your consent, which you can withdraw here or through the “Cookies” link in the footer.