Harpie Deck #767370 — Yu-Gi-Oh! OCG

Deck — Deck Power Evaluation

Final Contextual Deck Power: 25/100 (±8 · Confidence: Low)

• Primary strategy: Midrange
• Secondary strategies: Toolbox, Stun
• Format: ocg

Legality

Main 52 / Extra 17 / Side 0 — status under the rules at evaluation time: Illegal. 4 violation(s).

Identity & game plan

• Win condition: Train-line Xyz boards (Superdreadnought Rail Cannon Gustav Rocket, Juggernaut Liebe, Flying Launcher) backed by Machina and Infinitrack bodies; Harpie Lady - Elegant Egotist and Mulcharmy draw-engine pressure with Kaiju and K9 side pieces
• Turn 1 (going first): Open a searcher (Night Train Blue Traveler, Infinitrack Harvester, Ancient Gear Box, or Urgent Schedule), set up a Level 10 Machine or Infinitrack body, then Xyz into a Train boss; hand-trap backup from Ash Blossom, Maxx C, or Droll & Lock Bird
• Turn 2 (going second): Use Ash Blossom, Droll & Lock Bird, Maxx C, and Ghost Sister to disrupt the opponent's first summon or search, then rebuild through Mulcharmy and Train recursion
• Recovery: Recover through graveyard Level 10 Machines (Night Train, Derricrane, Exceptional Schedule banish) and Infinitrack Trencher or Tunneller revival
• After siding: Not evaluable: the deck has no Side Deck (side_count 0)

Why it scores here

The deck is scored as a system, not as an average of its cards. Its searcher density (14 text-searcher cards, about 81 percent chance of at least one in a 5-card opening hand) gives it a reasonable chance to find a starter, but the archetype-core check shows only one of its 52 Main Deck cards is a true Harpie core piece, so the engine is split across Train, Infinitrack, Mulcharmy, Machina and K9 with about 3 percent exact-name purity. Interaction is reasonable (Ash Blossom, Maxx C, Droll and Ghost Sister), and the late Extra Deck (Rocket at 5000 ATK, Juggernaut Liebe, Flying Launcher) gives real closing power, but the Extra Deck is 2 over the OCG limit and contains two Forbidden cards, so it cannot be played legally. Popularity or archetype share did not earn points, and no tournament or win-rate evidence exists. The deck is capped low as an illegal, theoretical, structural-only evaluation.

Dimension scores

How the score was computed

Strategy profile: Midrange. Structural ≈ 45, Going First ≈ 46, Going Second ≈ 40, Meta = provisional, Tournament Reliability ≈ 35. Modifiers: banlist_exposure (-6); engine_conflict (-4). Final formula: 0.80*Structural(45) + 0.20*TournamentReliability(35) + ContextModifiers(-10) = 33, then Illegal cap applied -> 25; Meta provisional

Probability & combo summary

• Opening a starter: P(≥1) ≈ 0.81 · key two-card combo ≈ — · full-brick estimate ≈ — (all Inferred)
• Combo output: best PowerIndex —, median —, board-per-starter —
• Turn order: 0 going-first lines vs 0 going-second lines

Strengths

• Dense searcher text (14 cards) gives a high chance of finding a starter
• Layered hand traps and removal text (Ash Blossom, Maxx C, Droll and Ghost Sister, plus 15 removal-text cards)
• High-ATK late Extra Deck (Gustav Rocket 5000 ATK, Juggernaut Liebe 4000 ATK)

Weaknesses

• Illegal in OCG: Extra Deck size 17 over the 15 maximum, 2 Forbidden cards and Gustav Max over its copy limit
• No Side Deck (side_count 0) so no post-side plan
• Engines split across five or more archetype families with about 3 percent exact-name purity

Critical choke points

• Extra Deck legality: cannot be played as built until it is cut to 15 and the Forbidden cards are removed
• Single Xyz starter lines depend on Level 10 Machine bodies, so loss of the Train core breaks the main route

Recommended changes

Evidence & confidence

• Verified: Legality: Illegal (Extra size 17 over 15; Forbidden Kashtira Arise-Heart and Varudras; Gustav Max 3 copies over Limited 1); Main Deck size 52 is within the 40-60 range; Card text and stats for all listed cards from the evidence pack; Searcher-text count 14; removal-text 15; hand-trap 4
• Inferred: Opening-hand searcher probability approx 0.81 (hypergeometric); Going-second hand-trap chance approx 0.40 (hypergeometric); Structural scores for D1 through D14 and the 45 Midrange blend; Engine split and purity implications
• Provisional: Meta Power (no dated metagame sample); Format fit D13 (no tournament link)
• Data gaps: No recorded combos supplied, so combo floor, median and ceiling are text-only inference No tournament link, so archetype meta-share and pilot match record are absent No win-rate telemetry for going-first, going-second, matchup, brick or OTK rates No Side Deck, so post-side performance cannot be assessed Meta Power is Provisional unless a dated metagame sample is supplied Illegal deck, so the score is capped and theoretical/structural-only