Cards that stand out.

Source: official SPAM server. Snapshot: 12 Oct 2026, 21:01 (Paris). Statistics include public profiles recovered from the ranked ladder, once per player. The server supplies up to three most-played cards per profile. Unavailable profiles are excluded from the denominator and counted in the coverage.

Win rate = total wins / total games with the card. These are card appearances in player records, not unique matches: the same match may contribute from both sides. Only each profile’s three favourites contribute: other cards used by these players are missing. Rank filters use rank at collection time, not at match time.

The period, mode and version are not specified for these counters: they may include multiple modes and older patches. The server does not specify whether bot matches are excluded. Card values are checked against client 1.22. Missing data does not mean a 0% win rate.

Favourite pickrate. Here, pickrate measures presence among the three favourite cards: profiles whose top three includes the card / all sampled profiles × 100. For example, 10 of 40 profiles means 25%. One profile can count for three cards, so the rates can sum to 300%. This is not the percentage of all matches: the public histories checked are empty, and the server does not supply a global selection dataset. 0% means absent from the sampled favourites, not never played.

Collection started: 12 Oct 2026, 18:16 (Paris). 23,318 profiles read · 0 unavailable. Collection takes time and is not a simultaneous snapshot.

Tiers require at least 30 games and 2 profiles. Cards are ordered by descending win rate and split into S/A/B/C/D: approximately 10/20/40/20/10%. Ties use game count, then a stable ID. Search and element filters do not change tiers. This distribution describes this sample, not absolute strength. Below these thresholds, a card stays unranked.