Card Draft Theory
Every Minor or Major gain pops a 4-card offer. You pick one, the other three shuffle back. That’s the mechanic. The theory — when to take, when to pass, how to coordinate drafts with a partner — is what separates drafting by feel from drafting for leverage.
Read this after Major vs. Minor. That chapter sets the archetype-level bias; this one operates below it.
The 4-card offer
When you gain a Minor Power, 4 Minor Power cards reveal from the top of the Minor deck. You take 1; the other 3 return to the bottom of the deck.
When you gain a Major Power, 4 Major Power cards reveal from the top of the Major deck. You take 1 (and pay the Forget cost); the other 3 return.
In multiplayer, the offer is pooled by default — any spirit can take any revealed card, but only the acting spirit pays the gain cost. This creates coordination opportunities (and alpha-gaming traps — see Alpha-Player Problem).
The two axes of card evaluation
Every offer card is evaluated on two dimensions, simultaneously:
1. Spirit fit
Does the card’s elements, cost, and effect type fit your spirit?
- Elements: the card adds elements to your per-turn total; you care about elements your innates threshold.
- Cost: the card must play within your energy curve.
- Effect type: damage, fear, defend, push, gather, etc. — matching your spirit’s needs.
A 0-cost Moon Minor for Shadows is A+. A 3-cost Fire-only Minor for Shadows is D because the cost breaks the curve.
2. Offer variance
What are the other 3 cards? If all 4 are weak fits, the best of 4 may still be worth taking. If 1 is a perfect fit and 3 are irrelevant, take the fit.
Experienced drafters think about expected-next-offer too: if you pass this mediocre card, the next gain-growth produces a fresh 4-offer. Sometimes the draft is a ratchet — take the decent card now because variance may hand you a worse offer next turn.
The decision tree
For each gain:
graph TD
Q1{Is any offer a top-tier fit?} -->|Yes| Take[Take it]
Q1 -->|No| Q2{Is the best-of-4 at least B-tier?}
Q2 -->|Yes| Q3{Do I have energy + CP for another gain soon?}
Q3 -->|Yes| Consider[Consider passing; next gain may be better]
Q3 -->|No| Take
Q2 -->|No| Q4{Am I critically short on this effect type?}
Q4 -->|Yes| Take
Q4 -->|No| Pass[Take the least-bad or pass if possible]
“Pass” here means: take a weak card to clear the offer, or take a gain-option that doesn’t advance card count (rarely available).
Multiplayer draft coordination
The 4-offer is pooled. Any spirit can claim a revealed card. This means:
Pre-gain discussion
Before Spirit A’s growth triggers a gain, a quick table check:
- “Anyone want a gain this turn? I could pick different growth.”
- “If the offer has X, call dibs.”
During-gain claim
When 4 cards reveal:
- Spirit A (gainer) announces their top pick.
- If Spirit B or C wants a different revealed card MORE than their own upcoming gain, they may swap — but only the gainer executes the gain mechanically.
Gain-pooling patterns
- Spirit A has a Major gain but nothing fits. Spirit B needs a Minor but doesn’t have a gain this turn. Spirit A can “waste” their Major gain on a Minor-shape offer for Spirit B’s benefit — if the table agreed.
- Multiple claims on one card: communicate before the reveal.
The alpha-gaming risk is real here. See Alpha-Player Problem for how to keep this coordination fair.
Forget math
Majors require a Forget. Evaluate:
- Which card am I forgetting? The weakest Unique, typically. Check the spirit chapter for your “forget-candidate.”
- Is the forget candidate still useful? Some Uniques degrade in the matchup (Wildfire’s Flash-Fires vs. England), making them free Forgets. Others are load-bearing (Earth’s Guard the Healing Land), making the Forget painful.
- Net value: Major gained − forgotten card utility. If negative, the Major isn’t worth it.
Draft cadence
Most spirits gain 3–5 cards over a game (excluding Major Power Shoppers who gain 5–7). Your draft cadence is:
- T1: if a gain-growth option exists, usually take it.
- T2–T3: 1–2 more gains, depending on spirit’s draft-bias.
- T4–T6: Major consideration (if Mixed or Major-heavy bias); more Minors (if Minor-heavy).
- T7+: closer draft — the one that ends the game.
Over-gaining (5+ cards on a Minor-heavy spirit) produces a bloated hand you can’t play. Under-gaining (2 cards by T6) leaves you without options. Target 3–4 total for most spirits.
Pass to shuffle
If an offer is uniformly bad AND you have future gains coming, refusing this one shuffles the 4 back to deck bottom — improving the next gain’s offer statistically. Rarely a primary strategy but valid as a tie-breaker.
Pool-size awareness
The Minor deck has ~101 cards; Major ~80. Your offers draw from this pool. Late-game, the deck shrinks as cards get into hands and stay there. Card variance changes — the 4-card offer late in a long game is often worse than a T2 offer because the “best” cards are already in play.
Card-specific grades
Each spirit chapter has a Card Priority Ratings table (e.g., Shadows’s table) listing the top Minors, Majors, and Uniques for that spirit. Refer to those when drafting at the table.
Stat insight
Per mindwanderer (2026-Q1):
- Spirits that pass Major offers more than 30% of the time have higher win rates than spirits that take the best-of-4 every time.
- Over-drafting Majors (4+ total) correlates with loss rates above 60%.
- Sample variance is high; treat as directional, not prescriptive.
Common mistakes
Taking “the cool Major” without threshold-fit. Vigor of the Breaking Dawn is a beautiful card; with 0 Sun elements, it’s dead in hand.
Passing a decent Minor “because the next offer might be better.” Variance makes next offers often worse. Take the B+ you have.
Drafting a Major that perfectly fits a future state you’ll reach T6 — but you’re on T2 with 1E. You paid the Forget for a 4-turn-delayed card.
Ignoring the offer-pool in multiplayer. Your Major gain might be better used to let a partner grab a better card. Coordinate.
Cross-references
- Major vs. Minor — archetype bias.
- Energy Curves — cost fit.
- Alpha-Player Problem — draft-coordination etiquette.
- SICK Card Database — browse + filter all power cards.
Last revised: 2026-04-19