Experimental balance lab · first screen
How do the rare cards change faction strength?
1,520 completed games. All 27 cards implemented in an isolated test engine, with separate low-cost and full-package comparisons.
First findings to investigate
Largest upward changes in the full-package Intermediate screen: Hearthbound +6.3 pp; Hourglass Covenant +6.3 pp; Revelwild +6.3 pp. Largest downward change: Prismforge -9.4 pp. Each faction has only 32 games per configuration. These are fixed-deck screening signals; no stat adjustment is justified from this sample alone.
Orabella triggered 5 times in the 720 full-package Intermediate games, restoring Charge 5 times and moving Allies 3 times. Her corrected ability works in natural simulated matches as well as targeted rules tests. This does not isolate her individual contribution from Revelwild’s other two rares.
Method and limits
Two 720-game screens: 45 unordered faction pairs including 9 mirrors, both companion selections, both starting seats and all four base/enhanced configurations. Full package adds three singleton rares; low-cost screen adds one. Each added card replaces a listed base card; all decks stay at 30 cards. Same seed within each paired comparison, normal mulligans. Advanced check: 80 games on fresh seeds across Hearthbound–Skyrunners, Veilborn–Revelwild, Ironmaw–Prismforge, Hourglass–Prismforge and Vigilant Array–Elderwake.
Four games per matchup/configuration per screen. Intermediate faction summaries have 32 non-mirror games per configuration, Advanced summaries only 4 or 8. Draws count as half a win. AI policies, fixed replacements and limited ability exposure constrain the results. Rare choices use deterministic policies: highest damage heal/repair, strongest ready Aevora target, original relative order on Blueprint bottoms. A small fair-information adapter lets AI use Terminus, Orabella and optional rare choices; this is not a new published AI version. Policies do not fully anticipate every rare combo or delayed threat. These are screening signals, not statistical balance claims, optimal deck searches or human win-rate predictions. No 2v2 testing in this batch. Mirrors are excluded from faction strength tables. “Change” compares a faction’s rare-enhanced deck against base opponents to its own base-deck results, in percentage points.
Low-cost rare only · Intermediate · 720 games
| Faction | Base vs base | Rare vs base | Change | Rare vs rare |
|---|---|---|---|---|
| Hearthbound | 53.1% 32 games | 53.1% 32 games | +0.0 pp | 53.1% 32 games |
| Skyrunners | 50.0% 32 games | 50.0% 32 games | +0.0 pp | 53.1% 32 games |
| Veilborn | 50.0% 32 games | 50.0% 32 games | +0.0 pp | 50.0% 32 games |
| Ironmaw | 37.5% 32 games | 43.8% 32 games | +6.3 pp | 43.8% 32 games |
| Hourglass Covenant | 53.1% 32 games | 53.1% 32 games | +0.0 pp | 50.0% 32 games |
| Revelwild | 37.5% 32 games | 37.5% 32 games | +0.0 pp | 40.6% 32 games |
| Prismforge | 56.3% 32 games | 50.0% 32 games | -6.3 pp | 46.9% 32 games |
| Vigilant Array | 53.1% 32 games | 53.1% 32 games | +0.0 pp | 53.1% 32 games |
| Elderwake | 59.4% 32 games | 59.4% 32 games | +0.0 pp | 59.4% 32 games |
Individual matchup results
Rates below are for the first named faction. Each cell contains only four games; draws count as half a win.
| Matchup | Both base | First rare | Second rare | Both rare |
|---|---|---|---|---|
| Hearthbound vs Skyrunners | 25.0% | 25.0% | 25.0% | 25.0% |
| Hearthbound vs Veilborn | 25.0% | 25.0% | 25.0% | 25.0% |
| Hearthbound vs Ironmaw | 75.0% | 75.0% | 75.0% | 75.0% |
| Hearthbound vs Hourglass Covenant | 50.0% | 50.0% | 50.0% | 50.0% |
| Hearthbound vs Revelwild | 75.0% | 75.0% | 75.0% | 75.0% |
| Hearthbound vs Prismforge | 25.0% | 25.0% | 25.0% | 25.0% |
| Hearthbound vs Vigilant Array | 100.0% | 100.0% | 100.0% | 100.0% |
| Hearthbound vs Elderwake | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Veilborn | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Ironmaw | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Hourglass Covenant | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Revelwild | 75.0% | 75.0% | 75.0% | 75.0% |
| Skyrunners vs Prismforge | 50.0% | 50.0% | 75.0% | 75.0% |
| Skyrunners vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Skyrunners vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Ironmaw | 100.0% | 100.0% | 100.0% | 100.0% |
| Veilborn vs Hourglass Covenant | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Revelwild | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Prismforge | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Elderwake | 75.0% | 75.0% | 75.0% | 75.0% |
| Ironmaw vs Hourglass Covenant | 0.0% | 25.0% | 0.0% | 25.0% |
| Ironmaw vs Revelwild | 100.0% | 100.0% | 100.0% | 100.0% |
| Ironmaw vs Prismforge | 50.0% | 75.0% | 50.0% | 75.0% |
| Ironmaw vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Ironmaw vs Elderwake | 50.0% | 50.0% | 50.0% | 50.0% |
| Hourglass Covenant vs Revelwild | 25.0% | 25.0% | 25.0% | 25.0% |
| Hourglass Covenant vs Prismforge | 75.0% | 75.0% | 75.0% | 75.0% |
| Hourglass Covenant vs Vigilant Array | 50.0% | 50.0% | 50.0% | 50.0% |
| Hourglass Covenant vs Elderwake | 0.0% | 0.0% | 0.0% | 0.0% |
| Revelwild vs Prismforge | 25.0% | 25.0% | 50.0% | 50.0% |
| Revelwild vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Revelwild vs Elderwake | 50.0% | 50.0% | 50.0% | 50.0% |
| Prismforge vs Vigilant Array | 50.0% | 50.0% | 50.0% | 50.0% |
| Prismforge vs Elderwake | 50.0% | 50.0% | 50.0% | 50.0% |
| Vigilant Array vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
Card exposure and ability counters
Deployments/casts can exceed games through reshuffling. Zero logged triggers is a coverage gap, not evidence of no value. Some direct tactics are covered by casts and targeted checks rather than separate effect counters.
| Card | Deployments / casts | Effect counters |
|---|---|---|
| Kindleseed Apothecary | 34 | heal: 4 |
| Zephyr Needle | 20 | freeMove: 1 |
| Threadblind Acolyte | 22 | filter: 2 |
| Ashfang Whelp | 42 | draw: 35 |
| Minutehand Scribe | 28 | insight: 2 |
| Mirthspark Firewalker | 10 | insight: 12 |
| Facetwisp Caretaker | 28 | No separately logged ability triggers |
| Orison Gatelet | 30 | No separately logged ability triggers |
| Mosswake Larvalith | 32 | No separately logged ability triggers |
All three rares · Intermediate · 720 games
| Faction | Base vs base | Rare vs base | Change | Rare vs rare |
|---|---|---|---|---|
| Hearthbound | 37.5% 32 games | 43.8% 32 games | +6.3 pp | 50.0% 32 games |
| Skyrunners | 40.6% 32 games | 40.6% 32 games | +0.0 pp | 40.6% 32 games |
| Veilborn | 62.5% 32 games | 59.4% 32 games | -3.1 pp | 53.1% 32 games |
| Ironmaw | 46.9% 32 games | 46.9% 32 games | +0.0 pp | 43.8% 32 games |
| Hourglass Covenant | 28.1% 32 games | 34.4% 32 games | +6.3 pp | 28.1% 32 games |
| Revelwild | 53.1% 32 games | 59.4% 32 games | +6.3 pp | 59.4% 32 games |
| Prismforge | 59.4% 32 games | 50.0% 32 games | -9.4 pp | 50.0% 32 games |
| Vigilant Array | 65.6% 32 games | 62.5% 32 games | -3.1 pp | 65.6% 32 games |
| Elderwake | 56.3% 32 games | 56.3% 32 games | +0.0 pp | 59.4% 32 games |
Individual matchup results
Rates below are for the first named faction. Each cell contains only four games; draws count as half a win.
| Matchup | Both base | First rare | Second rare | Both rare |
|---|---|---|---|---|
| Hearthbound vs Skyrunners | 75.0% | 75.0% | 75.0% | 75.0% |
| Hearthbound vs Veilborn | 0.0% | 0.0% | 25.0% | 25.0% |
| Hearthbound vs Ironmaw | 50.0% | 75.0% | 50.0% | 75.0% |
| Hearthbound vs Hourglass Covenant | 100.0% | 100.0% | 100.0% | 100.0% |
| Hearthbound vs Revelwild | 50.0% | 50.0% | 50.0% | 50.0% |
| Hearthbound vs Prismforge | 0.0% | 0.0% | 50.0% | 25.0% |
| Hearthbound vs Vigilant Array | 0.0% | 25.0% | 25.0% | 25.0% |
| Hearthbound vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
| Skyrunners vs Veilborn | 25.0% | 25.0% | 25.0% | 25.0% |
| Skyrunners vs Ironmaw | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Hourglass Covenant | 75.0% | 75.0% | 50.0% | 50.0% |
| Skyrunners vs Revelwild | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Prismforge | 0.0% | 0.0% | 0.0% | 25.0% |
| Skyrunners vs Vigilant Array | 50.0% | 50.0% | 50.0% | 50.0% |
| Skyrunners vs Elderwake | 50.0% | 50.0% | 50.0% | 50.0% |
| Veilborn vs Ironmaw | 75.0% | 75.0% | 75.0% | 75.0% |
| Veilborn vs Hourglass Covenant | 50.0% | 50.0% | 50.0% | 50.0% |
| Veilborn vs Revelwild | 100.0% | 100.0% | 75.0% | 75.0% |
| Veilborn vs Prismforge | 50.0% | 50.0% | 25.0% | 25.0% |
| Veilborn vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Veilborn vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
| Ironmaw vs Hourglass Covenant | 75.0% | 75.0% | 75.0% | 75.0% |
| Ironmaw vs Revelwild | 25.0% | 25.0% | 50.0% | 50.0% |
| Ironmaw vs Prismforge | 100.0% | 100.0% | 75.0% | 75.0% |
| Ironmaw vs Vigilant Array | 25.0% | 25.0% | 25.0% | 25.0% |
| Ironmaw vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
| Hourglass Covenant vs Revelwild | 25.0% | 25.0% | 0.0% | 0.0% |
| Hourglass Covenant vs Prismforge | 50.0% | 50.0% | 50.0% | 50.0% |
| Hourglass Covenant vs Vigilant Array | 25.0% | 50.0% | 25.0% | 25.0% |
| Hourglass Covenant vs Elderwake | 25.0% | 25.0% | 25.0% | 25.0% |
| Revelwild vs Prismforge | 50.0% | 50.0% | 50.0% | 50.0% |
| Revelwild vs Vigilant Array | 75.0% | 75.0% | 75.0% | 75.0% |
| Revelwild vs Elderwake | 50.0% | 75.0% | 50.0% | 75.0% |
| Prismforge vs Vigilant Array | 50.0% | 0.0% | 25.0% | 0.0% |
| Prismforge vs Elderwake | 75.0% | 50.0% | 75.0% | 50.0% |
| Vigilant Array vs Elderwake | 75.0% | 50.0% | 75.0% | 50.0% |
Card exposure and ability counters
Deployments/casts can exceed games through reshuffling. Zero logged triggers is a coverage gap, not evidence of no value. Some direct tactics are covered by casts and targeted checks rather than separate effect counters.
| Card | Deployments / casts | Effect counters |
|---|---|---|
| Hearthroot Matriarch | 19 | heal: 6; insight: 6 |
| Reclaim the Hearth | 0 | No separately logged ability triggers |
| Kindleseed Apothecary | 34 | No separately logged ability triggers |
| Auric Cloud Regent | 5 | No separately logged ability triggers |
| Windward Procession | 0 | No separately logged ability triggers |
| Zephyr Needle | 29 | No separately logged ability triggers |
| Maw Beyond the Map | 5 | No separately logged ability triggers |
| Veil of Reversal | 0 | No separately logged ability triggers |
| Threadblind Acolyte | 24 | filter: 10 |
| Thronetusk Colossus | 20 | ready: 11 |
| Banner of Predators | 0 | No separately logged ability triggers |
| Ashfang Whelp | 33 | draw: 32 |
| Aevora, Keeper of Unlived Hours | 12 | grant: 1 |
| The Hour Nobody Remembers | 0 | No separately logged ability triggers |
| Minutehand Scribe | 22 | No separately logged ability triggers |
| Orabella, Carnival Crown | 12 | trigger: 5; refund: 5; move: 3 |
| One More Impossible Night | 0 | No separately logged ability triggers |
| Mirthspark Firewalker | 18 | insight: 19 |
| Cathedrax, Living Observatory | 32 | renown: 20; heal: 2 |
| Blueprint of Tomorrow | 0 | No separately logged ability triggers |
| Facetwisp Caretaker | 39 | No separately logged ability triggers |
| Last Horizon Executor | 12 | No separately logged ability triggers |
| Terminus Protocol | 1 | No separately logged ability triggers |
| Orison Gatelet | 28 | No separately logged ability triggers |
| Itharyn, Continental Warden | 8 | No separately logged ability triggers |
| World Before Footsteps | 0 | No separately logged ability triggers |
| Mosswake Larvalith | 28 | No separately logged ability triggers |
All three rares · Advanced focused check · 80 games
Only the five named cross-faction pairs above. Do not compare these aggregate rates to the all-matchup Intermediate tables.
| Faction | Base vs base | Rare vs base | Change | Rare vs rare |
|---|---|---|---|---|
| Hearthbound | 25.0% 4 games | 25.0% 4 games | +0.0 pp | 50.0% 4 games |
| Skyrunners | 75.0% 4 games | 75.0% 4 games | +0.0 pp | 50.0% 4 games |
| Veilborn | 0.0% 4 games | 0.0% 4 games | +0.0 pp | 0.0% 4 games |
| Ironmaw | 50.0% 4 games | 75.0% 4 games | +25.0 pp | 100.0% 4 games |
| Hourglass Covenant | 75.0% 4 games | 50.0% 4 games | -25.0 pp | 100.0% 4 games |
| Revelwild | 100.0% 4 games | 100.0% 4 games | +0.0 pp | 100.0% 4 games |
| Prismforge | 37.5% 8 games | 0.0% 8 games | -37.5 pp | 0.0% 8 games |
| Vigilant Array | 25.0% 4 games | 25.0% 4 games | +0.0 pp | 50.0% 4 games |
| Elderwake | 75.0% 4 games | 50.0% 4 games | -25.0 pp | 50.0% 4 games |
Individual matchup results
Rates below are for the first named faction. Each cell contains only four games; draws count as half a win.
| Matchup | Both base | First rare | Second rare | Both rare |
|---|---|---|---|---|
| Hearthbound vs Skyrunners | 25.0% | 25.0% | 25.0% | 50.0% |
| Veilborn vs Revelwild | 0.0% | 0.0% | 0.0% | 0.0% |
| Ironmaw vs Prismforge | 50.0% | 75.0% | 100.0% | 100.0% |
| Hourglass Covenant vs Prismforge | 75.0% | 50.0% | 100.0% | 100.0% |
| Vigilant Array vs Elderwake | 25.0% | 25.0% | 50.0% | 50.0% |
Card exposure and ability counters
Deployments/casts can exceed games through reshuffling. Zero logged triggers is a coverage gap, not evidence of no value. Some direct tactics are covered by casts and targeted checks rather than separate effect counters.
| Card | Deployments / casts | Effect counters |
|---|---|---|
| Hearthroot Matriarch | 6 | No separately logged ability triggers |
| Reclaim the Hearth | 0 | No separately logged ability triggers |
| Kindleseed Apothecary | 4 | No separately logged ability triggers |
| Auric Cloud Regent | 2 | No separately logged ability triggers |
| Windward Procession | 2 | No separately logged ability triggers |
| Zephyr Needle | 4 | No separately logged ability triggers |
| Maw Beyond the Map | 4 | No separately logged ability triggers |
| Veil of Reversal | 0 | No separately logged ability triggers |
| Threadblind Acolyte | 2 | No separately logged ability triggers |
| Thronetusk Colossus | 3 | ready: 1 |
| Banner of Predators | 2 | renown: 1 |
| Ashfang Whelp | 3 | No separately logged ability triggers |
| Aevora, Keeper of Unlived Hours | 2 | grant: 1; freeChallenge: 1 |
| The Hour Nobody Remembers | 1 | resolved: 1 |
| Minutehand Scribe | 4 | No separately logged ability triggers |
| Orabella, Carnival Crown | 3 | trigger: 2; refund: 2; move: 1 |
| One More Impossible Night | 6 | No separately logged ability triggers |
| Mirthspark Firewalker | 6 | insight: 16 |
| Cathedrax, Living Observatory | 8 | renown: 3 |
| Blueprint of Tomorrow | 13 | resolved: 13 |
| Facetwisp Caretaker | 5 | No separately logged ability triggers |
| Last Horizon Executor | 4 | No separately logged ability triggers |
| Terminus Protocol | 0 | No separately logged ability triggers |
| Orison Gatelet | 4 | No separately logged ability triggers |
| Itharyn, Continental Warden | 3 | No separately logged ability triggers |
| World Before Footsteps | 0 | No separately logged ability triggers |
| Mosswake Larvalith | 4 | No separately logged ability triggers |
Full-package deck substitutions
Three one-for-one replacements: two Allies and one Tactic. These fixed presets are not optimized decks.
| Faction | Add one copy each | Remove one copy each |
|---|---|---|
| Hearthbound | Hearthroot Matriarch Reclaim the Hearth Kindleseed Apothecary | Cindertusk Briar Verdict Lampkin |
| Skyrunners | Auric Cloud Regent Windward Procession Zephyr Needle | Thundercrest Lantern Signals Kitelet |
| Veilborn | Maw Beyond the Map Veil of Reversal Threadblind Acolyte | Riftcrown Stolen Horizon Wispweaver |
| Ironmaw | Thronetusk Colossus Banner of Predators Ashfang Whelp | Ridgebreaker Shatter the Line Shardsprinter |
| Hourglass Covenant | Aevora, Keeper of Unlived Hours The Hour Nobody Remembers Minutehand Scribe | Eonbreaker The Coming Fracture Yesterday’s Echo |
| Revelwild | Orabella, Carnival Crown One More Impossible Night Mirthspark Firewalker | Crescendo Hulk Painted Rebuke Paintspark |
| Prismforge | Cathedrax, Living Observatory Blueprint of Tomorrow Facetwisp Caretaker | Facetbreaker Prismatic Lance Shardling Surveyor |
| Vigilant Array | Last Horizon Executor Terminus Protocol Orison Gatelet | Orison, Last Directive Intercept Protocol Spark Surveyor |
| Elderwake | Itharyn, Continental Warden World Before Footsteps Mosswake Larvalith | Oruun, the Walking Horizon Seismic Roar Mossback Forager |
All 27 cards: rules-check status
| Faction | Card | Status |
|---|---|---|
| Hearthbound | Hearthroot Matriarch | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Hearthbound | Reclaim the Hearth | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Hearthbound | Kindleseed Apothecary | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Skyrunners | Auric Cloud Regent | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Skyrunners | Windward Procession | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Skyrunners | Zephyr Needle | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Veilborn | Maw Beyond the Map | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Veilborn | Veil of Reversal | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Veilborn | Threadblind Acolyte | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Ironmaw | Thronetusk Colossus | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Ironmaw | Banner of Predators | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Ironmaw | Ashfang Whelp | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Hourglass Covenant | Aevora, Keeper of Unlived Hours | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Hourglass Covenant | The Hour Nobody Remembers | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Hourglass Covenant | Minutehand Scribe | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Revelwild | Orabella, Carnival Crown | Corrected to trigger after an Encore resolves. Targeted refund, cap, once-per-turn and movement checks passed; included in package simulations. |
| Revelwild | One More Impossible Night | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Revelwild | Mirthspark Firewalker | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Prismforge | Cathedrax, Living Observatory | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Prismforge | Blueprint of Tomorrow | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Prismforge | Facetwisp Caretaker | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Vigilant Array | Last Horizon Executor | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Vigilant Array | Terminus Protocol | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Vigilant Array | Orison Gatelet | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Elderwake | Itharyn, Continental Warden | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Elderwake | World Before Footsteps | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
| Elderwake | Mosswake Larvalith | Implemented in isolated engine; targeted rule checks passed; included in full-package simulations. |
Follow-up before release
Investigate large changes with more seeds and alternative substitutions; improve rare-specific AI choices where ability exposure is low; then run human playtests and separate 2v2 tests. No automatic stat changes are justified by this small screen alone.