Recommended lineups & decks

Layer 1 is your SKILL CARD deck — not picked in advance, but gathered gradually inside a produce run through lessons and events, mixing default/produce/pIdol/support-sourced cards. Layer 2 is the P-ITEM, a PASSIVE piece of gear that triggers on a condition (start of turn, playing a certain card type...) and carries an activation limit. Layer 3 is the SUPPORT CARD, which must already be attached BEFORE you enter produce, matching the training direction you picked.

• These 3 layers aren't independent — they all need to push toward the same signature effect (recommendedEffect) of the P-idol you're raising.
• The P-idol's signature P-item is usually the centerpiece, with skill cards and support cards built to amplify that same effect rather than each layer pulling a different way.
• Misaligning the layers — say, a P-item that pushes Concentration while the support card leans hard into Motivation — is the most common reason a loadout looks strong on paper but scores nothing.

SENSE is built around 好調 Good Condition (a % score boost while it's active) and 集中 Concentration (score that stacks with each card played) — leaning toward burst scoring in a few turns. LOGIC is built around 好印象 Good Impression (passive score every turn) and やる気 Motivation (raises genki/元気) — leaning toward steady, snowballing gains across many turns. ANOMALY is a newer plan with its own separate axis; it's recent enough that you should only treat it at a general level rather than trusting detailed combos for it.

• Skill cards, P-items, and support cards each lock to one specific plan (except FREE cards, which work in any plan) — slotting the wrong plan means a dead slot, one that doesn't synergize with anything else in the loadout.
• Before picking up any card or P-item during produce, always ask whether it pushes the exact axis your P-idol needs — Good Condition/Concentration for Sense, or Good Impression/Motivation for Logic.
• The strongest loadout is the one with the fewest off-plan pieces, not the one with the rarest pieces.

The more cards you stuff into your deck, the lower your odds of drawing the exact card you need on the exact turn you need it — a strong ACTIVE card sitting inert among 20 useless ones is as good as not owning it. Worse, TROUBLE cards (bad cards forced into your deck by certain events/effects) surface more often as the deck grows thicker, because the good cards get diluted across a bigger pool.

• Prioritize keeping ACTIVE cards that actually fit your plan, and skip extra cards whenever produce offers a choice — thinner and focused almost always beats thick and mixed.
• ACTIVE cards cost stamina to play, while MENTAL cards recover stamina or grant buffs — balance the two to the game's rhythm: enough MENTAL to survive to the midterm exam without burning too many recovery turns, enough ACTIVE to close out the final exam with points.
• Managing stamina across the whole produce run matters as much as card count — running out mid-run forces a rest, costing a turn you'd otherwise spend picking up cards or training.

Support cards split across 3 stats — Vocal / Dance / Visual — with rarities SSR / SR / R. You get them from gacha (shared banner with P-idols), by trading contest coins, or from free giveaways. The milestone that matters most isn't rarity, it's LEVEL: leveling one card to Lv25 unlocks its ABILITY — below Lv25, that card is contributing little more than raw stats.

• Pouring level-up resources into one card to hit Lv25 beats spreading them thin across several half-leveled cards — one card with its ability unlocked usually outperforms several half-finished ones combined.
• Attach support cards that match your chosen training direction — leaning Vocal, Dance, or Visual — rather than defaulting to whichever card is rarest; an SR already at Lv25 in the right direction usually beats an unleveled, off-direction SSR.
• Contest coins are a reliable way to target the exact card you need instead of leaving it entirely to gacha luck.

CONTEST is an asynchronous PvP mode — you fight with a MEMORY (a saved record of an idol that finished produce), while the opponent is an AI simulation of another player's team, not someone playing live. Because of that, contest loadouts get scored by a fixed formula, not by a gut feeling of "higher stats means stronger":

Contest power = 3×(Vocal + Dance + Visual) + 24×Stamina + P-item power + card power + 36×number of customs.

The coefficients say it plainly: 1 point of Stamina is worth 24 power, while 1 point of stat (Vocal/Dance/Visual) is only worth 3. Undervaluing Stamina when building a contest memory is the most common mistake — a memory loaded with stats but short on Stamina is usually much weaker than a balanced one.

Loadouts don't carry over cleanly either: inside the H.I.F編 chapter, shared Legend cards (the type usable across multiple chapters) simply can't be used, so the power balance between the 3 plans flips compared to the base chapter — a loadout that dominated the regular chapter can drop noticeably once ported straight into H.I.F編 without recalculating.