QA

Every booster pack contains exactly 16 cards in specific slots:

The foil slot uses weighted random selection (weights vary by set):

What are Belts?

Belts are shuffled pools of cards that ensure proper distribution and prevent immediate duplicates. Each belt contains all eligible cards for that slot, shuffled randomly.

How Belts Work:

1. Cards are organized into belts by slot type (leader, base, common A, common B, uncommon, rare/legendary, foil)
2. Each belt is shuffled once at the start
3. Cards are drawn sequentially from the belt
4. When a belt is exhausted, it's reshuffled and reused
5. This prevents the same card from appearing in consecutive packs

Common Belts (A & B):

• Belt A: Vigilance, Command, Heroism aspects
• Belt B: Aggression, Cunning, Villainy aspects
• Packs alternate drawing from A and B (positions 0,2,4,6,8 vs 1,3,5,7)
• Ensures all 6 aspects appear in every pack's 9 commons

After the base pack is generated, an upgrade pass can replace certain cards with premium versions:

The Quality Score is a percentage representing overall pack generation health:

• Calculation: (metrics within expected / total metrics) × 100
• Excellent (90-100%): Nearly all metrics pass validation
• Good (75-89%): Most metrics pass, minor variances acceptable
• Acceptable (60-74%): Some metrics failing, may need attention
• Needs Attention (<60%): Significant issues with pack generation

The score combines structural checks (pass/fail) and rate metrics (statistical tolerance).

The Quality subtab in each set shows automated validation of pack generation. Here's what each metric measures:

Structural Validation:

• Pack has exactly 16 cards: Every pack must contain exactly 16 cards
• Pack has 1 leader: Exactly one leader card per pack
• Pack has 1 base: Exactly one base card per pack
• Pack has 9 commons: Nine common cards (from A/B belts)
• Pack has 1 foil: Exactly one foil card of any rarity
• All 6 aspects in commons: The 9 commons must cover all 6 aspects
• No same-treatment duplicates: No card appears twice with the same treatment

Rate Validation:

• Legendary Rate: Sets 1-3: ~14.3% (6:1), Sets 4+: ~16.7% (5:1)
• Hyperspace Leader: ~17% of leaders should be Hyperspace (1 in 6 packs)
• Hyperspace Base: ~25% of bases should be Hyperspace (1 in 4 packs)
• Hyperspace Common: ~33% of packs should have a Hyperspace common (1 in 3)
• Hyperfoil: ~2% of foils should be Hyperspace Foil variants (1 in 50)
• Showcase Leader: ~0.35% of leaders should be Showcase (~1 in 288 packs)

Foil Distribution (Sets 1-3 / Sets 4+):

• Common foils: Expected ~78% / ~75%
• Uncommon foils: Expected ~17% / ~17%
• Rare foils: Expected ~5% / ~4%
• Legendary foils: Expected ~0.3% / ~0.3%
• Special foils: Expected 0% / ~4% (Sets 4+ only, same rate as Rare)

Z-Score (Standard Score):

Measures how many standard deviations an observed value is from the expected value.

• Formula: Z = (observed - expected) / standard_error
• |Z| < 1.96: Within 95% confidence (normal)
• 1.96 < |Z| < 2.58: Between 95-99% confidence (outlier)
• |Z| > 2.58: Beyond 99% confidence (extreme outlier)

Wilson Score Confidence Interval:

Used for binomial proportions (rate metrics). More accurate than simple intervals for small samples or extreme probabilities.

• Provides upper and lower bounds for the true rate
• If expected rate falls within the confidence interval, the metric passes
• Handles edge cases (0% or 100% observed) better than simpler methods

Chi-Squared Goodness-of-Fit Test:

Tests whether an observed distribution matches an expected distribution.

• Used for foil rarity distribution (multiple categories)
• p-value > 0.05: Distribution matches expected (pass)
• p-value < 0.05: Significant deviation from expected (fail)
• Reports degrees of freedom (df) = number of categories - 1

All rate metrics use Z-tests to compare actual vs expected generation rates:

In a typical booster box (24 packs):

• Legendaries: ~3-4 non-foil + ~0.5 foil = ~4 total (Sets 1-3), ~4-5 total (Sets 4+)
• Rares: ~20-21 non-foil + ~2 foil = ~22 total
• Hyperfoils: ~0.5 (about 1 per 2 boxes)
• Showcase Leaders: ~0.08 per box (~1 per 12 boxes, all sets)
• UC Upgrades: ~4-5 packs will have 2 rare/legendary

In a sealed pool (6 packs):

• Legendaries: ~1 total (0-2 range is normal)
• Rares: ~5-6 total
• All 6 aspects: Guaranteed in commons of every pack
• Card+Foil pairs: 95-99% chance of at least one pair

Card Treatments:

• Base: Normal version of a card (no special treatment)
• Hyperspace: Blue border alternate art version
• Foil: Shiny/holographic version (any variant can be foil)
• Hyperspace Foil: Foil version of Hyperspace variant
• Showcase: Borderless premium art (leaders only, all sets)

Duplicate Prevention:

• Same card with same treatment cannot appear twice in one pack
• Different treatments of same card ARE allowed (e.g., normal + foil version)
• Belt system prevents immediate duplicates across consecutive packs

Aspect Balance:

• All 6 aspects (Vigilance, Command, Aggression, Cunning, Heroism, Villainy) guaranteed in every pack
• Achieved through alternating common belt draws (A/B pattern)
• Belt A: Vigilance, Command, Heroism (positions 0,2,4,6,8)
• Belt B: Aggression, Cunning, Villainy (positions 1,3,5,7)
• 9th common position ensures at least one duplicate aspect for deckbuilding