The public attacker tier list available for this review was a version 7.2 snapshot, while later community update records reached v9.0.3. The names below are therefore a historical testing shortlist, not a current grade table. Verify each card’s present text, availability, and behavior before investing.
Historical high-priority watchlist
The old top group included Awakened Lion, Fox Sin, Awakened Eclipseborn Hawk, and Awakened Tormented Swordsman. A following high group included Awakened Bijuu Beast, Undead King, Blood Valkyrie, Wrathful Sin, Commandment of Truth, and Envious Sin.
Other previously discussed attackers included Shadow Ant, Shadow Commander, Soul Queen, Eminence Cat, and Deranged Elder Assassin. These placements describe an earlier community view. They do not claim those cards remain obtainable, unchanged, or superior to later additions.
Compare the damage pattern
Classify the attacker by front-loaded damage, steady output, scaling, stacking, burst windows, target behavior, or a special condition shown on the current card. Fast Story waves reward different timing from a long World Boss encounter. A unit that becomes powerful late may never reach that state in routine farming.
Read activation and round limits. Note summons, domains, revive effects, or one-per-party restrictions. Then select supports that actually interact with that pattern.
Test Story and farming value
Use a repeatable Story stage you already clear. Save the baseline deck, change one attacker, and compare completion time across several runs. Check whether the new unit clears ordinary waves reliably or relies on a rare sequence. Include setup time and failures, not only the fastest result.
For farming, consistency is usually more valuable than a theoretical peak. A slightly lower ceiling can win if it activates sooner and reduces failed or slow runs.
Test bosses and towers
For raids, World Boss, or another durable target, record sustained damage and whether scaling reaches full value. Confirm reward eligibility using the live activity rules. For towers, test at the current checkpoint or floor where your baseline begins to struggle; early floors may hide weaknesses.
Do not compare a fully invested showcase card with an undeveloped replacement. Match levels and removable items where practical, and record traits that cannot be equalized.
Read Ranked separately
Ranked may divide queues by scaling rules, apply duplicate restrictions, or change whether traits operate. Review the live rule panel before treating a PvE result as competitive proof. A card can be excellent in unrestricted play and ordinary under scaled conditions, or the reverse.
Judge the whole team: activation, counterplay, support synergy, and consistency across matches. One highlight does not create a reliable rank.
Decide where to invest
Promote a candidate to your core only after it improves a defined benchmark. Spend common leveling resources first. Add equipment that matches its role. Use scarce trait rerolls only when the card is stable in the deck and you understand replacement cost.
Keep one proven attacker while developing another. That preserves your farming loop if the experiment fails or an update changes the target.
Build your current list
Label owned attackers as Story/farming, boss, tower, Ranked, development, or collection. A card can hold more than one label. Add a date and one measurable reason, such as consistent clear time or sustained damage.
Recheck after balance changes and new packs. Continue with Support Rankings to complete the team and Deck Building for the one-slot test process.
Use a compact attacker scorecard
Give each tested attacker a note for activation speed, ordinary-wave reliability, long-fight output, targeting, support requirements, and resource investment. Use words such as low, medium, and high unless the game supplies a stable measurable value. Add the exact stage or boss used for the test.
If a result looks unexpectedly strong, repeat it without changing potions, Moon state, traits, or artifacts. If the advantage disappears, record the condition rather than promoting the card. If it persists, test the next harder target. This two-stage check prevents a lucky run or hidden modifier from becoming a false tier claim.
For a newly obtained attacker, begin with common leveling resources and a reversible item. Keep the established unit available until the replacement proves both damage and consistency. That makes experimentation safe even when a later update changes the candidate.