EV Statistics: Spiritforged
Riftbound: League of Legends TCG • Monte Carlo simulation • model v1
Expected Value by Product
| Product | Price | Mean EV | Median EV | ROI | Std Dev | P5 | P25 | P75 | P95 | P99 | Simulated |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Pack | $6.88 | $7.00 | $2.74 | +1.7% | $41.43 | $1.56 | $1.86 | $4.27 | $21.45 | $62.61 | 1,008,000 |
| Box | $171.35 | $167.96 | $127.41 | -2.0% | $194.79 | $77.88 | $99.9 | $168.06 | $384.36 | $784.72 | 42,000 |
| Case | $1118.84 | $1007.76 | $862.98 | -9.9% | $480.40 | $648.16 | $752.49 | $1089.79 | $1782.16 | $3550.77 | 7,000 |
Mean EV is skewed upward by ultra-rare chase cards; the median tells you what a typical rip returns. Percentiles show the spread: P5 is a bad rip, P95 is a great one.
Hit Odds (per Booster Box)
| Rarity | Expected / Pack | Expected / Box | Expected / Case | P(≥1 per box) |
|---|---|---|---|---|
| R | 2.6062 | 62.55 | 375.30 | 100.0% |
| Epic | 0.2802 | 6.73 | 40.35 | 100.0% |
| Overnumber | 0.0140 | 0.34 | 2.01 | 33.5% |
| Alt-Art Rune | 0.0313 | 0.75 | 4.51 | 75.2% |
| Ultimate Rare | 0.0000 | 0.00 | 0.00 | 0.0% |
| Signature Overnumber | 0.0014 | 0.03 | 0.19 | 3.2% |
| Alt-Art Champion | 0.0669 | 1.61 | 9.63 | 100.0% |
Active Pull-Rate Model
| Rarity | Finish | Rate / Pack | Odds | Sources |
|---|---|---|---|---|
| C | normal | 7.0000 | 7.00 per pack | Riot official pack anatomy (14 cards: 7C / 3UC / 2 foil R+ / 1 rare / 1 token) |
| UC | normal | 3.0000 | 3.00 per pack | Riot official pack anatomy (14 cards: 7C / 3UC / 2 foil R+ / 1 rare / 1 token) |
| Epic | foil | 0.2500 | 1 : 4 packs | Launch-window community calibration (site engine v1 rates); Spiritforged pull rates — 72-pack community aggregation (TCGOdds) |
| Rune | foil | 0.1000 | 1 : 10 packs | Launch-window community calibration (site engine v1 rates) |
| Alt-Art Champion | any | 0.0670 | 1 : 15 packs | Launch-window community calibration (site engine v1 rates); Spiritforged pull rates — 72-pack community aggregation (TCGOdds) |
| Alt-Art Rune | any | 0.0313 | 1 : 32 packs | Launch-window community calibration (site engine v1 rates) |
| Overnumber | any | 0.0146 | 1 : 69 packs | Launch-window community calibration (site engine v1 rates); Spiritforged pull rates — 72-pack community aggregation (TCGOdds) |
| Signature Overnumber | any | 0.0014 | 1 : 727 packs | Launch-window community calibration (site engine v1 rates) |
Validation Report
Latest run: PASSED • 33 metrics checked against research data • 0 outside tolerance • 10,000 boxes, 1,000 cases, 100,008 packs
| Metric | Observed | Simulated | Delta | Tolerance | Result |
|---|---|---|---|---|---|
| C (normal) per pack | 7 | 7 | +0% | ±10% | PASS |
| UC (normal) per pack | 3 | 3 | +0% | ±10% | PASS |
| ALT_ART_CHAMPION (any) per pack | 0.066667 | 0.066745 | +0.12% | ±60.00000000000001% | PASS |
| ALT_ART_RUNE (any) per pack | 0.03125 | 0.031737 | +1.56% | ±60.00000000000001% | PASS |
| EPIC (foil) per pack | 0.25 | 0.279698 | +11.88% | ±60.00000000000001% | PASS |
| OVERNUMBER (any) per pack | 0.0125 | 0.013709 | +9.67% | ±60.00000000000001% | PASS |
| RUNE (foil) per pack | 0.1 | 0.098222 | -1.78% | ±60.00000000000001% | PASS |
| SIG_OVERNUMBER (any) per pack | 0.001375 | 0.00131 | -4.73% | ±60.00000000000001% | PASS |
| ALT_ART_CHAMPION (any) per pack | 0.069444 | 0.066745 | -3.89% | ±89.44271909999159% | PASS |
| EPIC (foil) per pack | 0.25 | 0.279698 | +11.88% | ±75% | PASS |
| OVERNUMBER (any) per pack | 0.027778 | 0.013709 | -50.65% | ±141.42135623730948% | PASS |
Research Sources
| Source | Type | Sample | Confidence | Data Points |
|---|---|---|---|---|
| Riot official pack anatomy (14 cards: 7C / 3UC / 2 foil R+ / 1 rare / 1 token) — Riot Games | official | — | 100% | 2 |
| Launch-window community calibration (site engine v1 rates) — Cards EV Lab research | community aggregate | — | 60% | 6 |
| Spiritforged pull rates — 72-pack community aggregation (TCGOdds) — TCGOdds | community aggregate | 72 packs / 3 boxes | 50% | 3 |
Pull rates are never hardcoded: every observation is stored with its source and confidence, merged into a weighted model, then verified by simulating 100,000+ packs and comparing the results back against the research data.