WATER
4Yaksha
ATK:
1900
DEF:
1500
View:
2159
Cannot be Special Summoned. During the End Phase of the turn this card is Normal Summoned or flipped face-up: Return it to the hand. When this card is Normal Summoned or flipped face-up: You can target 1 Spell/Trap Card your opponent controls; return that target to the hand.
Yaksha — Power Evaluation
Final Score: 51 ± 8 · Confidence: Medium
- Intrinsic: 51 · Deck: 51 · Meta: provisional
- Primary role: Board breaker · Secondary: Recovery/Recursion
Why it scores here
Yaksha's power is a tempo effect, not card advantage. Each Normal Summon lets it bounce one opposing Spell/Trap to hand, and it returns itself to hand at End Phase, so the effect is repeatable but card-neutral. Its presence in 222 of 231 recent decks (observed) confirms a practical role in the Shinobird home, but guardrail 3 means that frequency does not raise the score. The 1900 ATK earns nothing, and with no combo evidence, combo conversion stays low.
Axis scores
| Axis | Score | Evidence |
|---|---|---|
| Accessibility | 55 | Needs a Normal Summon each turn and runs a single copy on average (1.03) with no searcher evidence, though it is live in 96% of Shinobird decks (observed). |
| Combo & Conversion | 25 | Text shows only a bounce plus self-return with no board-building or follow-up, and no combo boards record it (inferred from text; no combo data). |
| Resource Economy | 45 | The bounce is tempo-positive but nets no card advantage, and the End Phase return keeps the body card-neutral (inferred from text). |
| Effect & Interaction | 55 | Targeted Spell/Trap bounce on each Normal Summon gives repeatable backrow removal, but it is limited to Normal Summon timing (verified text, inferred strength). |
| Material Utility | 25 | A Level 4 Water Fairy body, but it returns to hand each turn, which limits its use as material (inferred from text; no material data). |
| Resilience & Flexibility | 40 | It leaves the field at End Phase, dodging sorcery-style removal, but the summon trigger can be negated and it has no redundancy (inferred from text). |
| Deck-Building Efficiency | 80 | One copy, no garnet or lock cost, and a cheap Level 4 body that fits a generic Shinobird build (observed from deck copies and text). |
Calculation
Weight profile: 0.7 Board breaker + 0.3 Recovery/Recursion. Intrinsic ≈ 51, Deck ≈ 51, Meta = provisional. Modifiers: none. Final formula: 0.40*Intrinsic + 0.60*Deck
Strengths
- Repeatable backrow removal: one Spell/Trap bounce on each Normal Summon, and the body returns to hand each End Phase (verified text)
- Live in 222 of 231 recent decks (96%), confirming a practical role in the Shinobird home (observed)
- Cheap single-copy slot with no garnet or lock cost (observed)
Weaknesses
- Needs a Normal Summon every turn and cannot be Special Summoned (verified text)
- Tempo only: the bounce and self-return create no card advantage (inferred)
- No combo evidence, so no observed starter, material, or payoff role (data gap)
- Almost all decks are a single archetype home (96% Shinobird) (observed)
Counters / choke points
- Negating its summon triggers removes the Spell/Trap bounce (inferred from text)
- An opponent with no Spell/Trap on the field makes the bounce trigger dead (inferred from text)
Best use
Board breaker: repeatable backrow bounce with self-return. Decks: Shinobird. Suggested copies: 1.
Evidence & gaps
- Verified: Text: cannot be Special Summoned; on Normal Summon or flip, bounce 1 opponent Spell/Trap; End Phase return to hand; 231 recent decks, average 1.03 copies, 222 in Shinobird (96.1%)
- Observed: 96.1% Shinobird inclusion across the attached decks; Zero recorded combo boards, so no starter, searched, material, or payoff stats; One Shinobird guide, which runs it in the key-card section
- Inferred: Repeatable backrow removal each turn it is Normal Summoned; Tempo-only value with no card advantage
- Data gaps: no combos supplied — combo conversion is text-only inference no banlist token — assumed 3 copies (unlimited) no usage statistics (StatisticsCard) — archetype inclusion % inferred from deck sample only






