Six custom Commander decks, validated by Monte Carlo simulation, real-precon calibration, Oracle templating lint, wiki theme audit and instrumented playtests. Sim run 2026-08-24 · dashboard generated 2026-08-27.
Dot = win rate, whisker = 95% CI. The hairline is the 25% four-player baseline; distance from it is the signal, not the absolute number.
| Deck | Win rate | 95% CI | Wins/games | Avg win round | Avg elim round |
|---|---|---|---|---|---|
| Ranger's Run | 30.1% | 29.8–30.4 | 20,075/66,670 | 13.6 | 11.0 |
| Bitterman's War | 26.2% | 25.8–26.5 | 17,452/66,670 | 13.4 | 11.2 |
| Rhino Squad | 25.3% | 25.0–25.7 | 16,880/66,670 | 12.9 | 11.2 |
| Hold The Line | 23.8% | 23.4–24.1 | 15,843/66,670 | 13.3 | 11.7 |
| The Final Tier | 23.3% | 22.9–23.6 | 15,513/66,670 | 14.0 | 10.8 |
| Arena Eternal | 19.4% | 19.1–19.7 | 12,913/66,670 | 13.4 | 10.9 |
The nerf/buff round that followed the first sim run. Spread tightened from 32.4–16.9 to 30.1–19.4.
| Deck | Before | After | Change |
|---|---|---|---|
| Ranger's Run | 32.4% | 30.1% | -2.3 |
| Bitterman's War | 30.0% | 26.2% | -3.8 |
| Rhino Squad | 24.1% | 25.3% | +1.2 |
| Hold The Line | 22.0% | 23.8% | +1.7 |
| The Final Tier | 22.8% | 23.3% | +0.5 |
| Arena Eternal | 16.9% | 19.4% | +2.5 |
Same decks, upgraded engine: players now hold reactive instants and cast them in response — removal on a big attacker or a scary suit-up, prevention when lethal comes in. Reactive cards the old engine called dead finally do their job.
| Deck | Before | After | Change |
|---|---|---|---|
| Ranger's Run | 30.2% | 30.1% | -0.1 |
| Bitterman's War | 26.2% | 26.2% | +0.0 |
| Rhino Squad | 25.4% | 25.3% | -0.0 |
| Hold The Line | 23.7% | 23.8% | +0.1 |
| The Final Tier | 23.3% | 23.3% | -0.0 |
| Arena Eternal | 19.3% | 19.4% | +0.0 |
Each deck 1-vs-3 against real WotC precons. Shaded band = the 20–35% bracket-2 target. No deck sits above it — the library errs weak, never strong.
Precon self-consistency (expect ≈25% each): Fae Dominion 18.0% · Spirit Squadron 29.7% · Land's Wrath 27.2% · Masters of Evil 27.6%. Parser coverage of the precons: Fae Dominion 97% cards / 99% sentences, Spirit Squadron 77% cards / 89% sentences, Land's Wrath 66% cards / 82% sentences, Masters of Evil 64% cards / 84% sentences. — misses weaken the precons, so custom rates are, if anything, overestimates.
| Deck | Win rate | Games | Verdict |
|---|---|---|---|
| Ranger's Run | 26.9% | 1,999 | in band |
| Bitterman's War | 23.5% | 1,997 | in band |
| Hold The Line | 20.9% | 1,998 | in band |
| Rhino Squad | 18.3% | 1,998 | below band - weaker than precon level |
| The Final Tier | 15.8% | 1,996 | below band - weaker than precon level |
| Arena Eternal | 13.2% | 1,997 | below band - weaker than precon level |
The same pods replayed with cautious, standard and aggressive players. The strongest and weakest decks stay the same everywhere — the rankings are findings, not artifacts of one heuristic.
| Deck | cautious | standard | aggressive |
|---|---|---|---|
| Ranger's Run | 28.5% | 30.6% | 31.5% |
| Bitterman's War | 26.6% | 26.9% | 27.3% |
| Rhino Squad | 26.9% | 25.3% | 23.2% |
| Hold The Line | 20.5% | 25.1% | 25.2% |
| The Final Tier | 23.3% | 22.4% | 23.3% |
| Arena Eternal | 19.8% | 18.1% | 18.1% |
The same pods with politically-motivated targeting: grudgeful players retaliate, archenemy tables converge on a runaway board, diplomats do both and spare harmless opponents. The tick is the apolitical baseline.
| Deck | standard | grudgeful | archenemy | diplomat |
|---|---|---|---|---|
| Ranger's Run | 30.6% | 31.6% | 40.3% | 42.4% |
| Bitterman's War | 26.9% | 27.2% | 24.9% | 24.6% |
| Rhino Squad | 25.3% | 23.9% | 17.3% | 16.9% |
| Hold The Line | 25.1% | 24.2% | 22.1% | 20.6% |
| The Final Tier | 22.4% | 23.0% | 24.4% | 24.5% |
| Arena Eternal | 18.1% | 18.2% | 17.1% | 16.2% |
Each deck solo against a do-nothing 40-life opponent. Shaded band = the ~8–11 bracket-2 window. Everyone is inside it: Arena Eternal's weakness is surviving interaction, not speed.
| Deck | Avg kill turn | Median |
|---|---|---|
| Rhino Squad | 8.7 | 8 |
| Hold The Line | 8.7 | 8 |
| Arena Eternal | 8.9 | 8 |
| Bitterman's War | 9.3 | 9 |
| The Final Tier | 9.5 | 9 |
| Ranger's Run | 10.2 | 10 |
200 games per deck.
Pure draw simulation, no AI. Ranger's Run is the outlier — a five-pip commander over a basics-heavy base — which is why its games feel bimodal at the table: colour-screwed, or the commander connecting.
| Deck | By turn 5 | Avg castable turn | Screw rate |
|---|---|---|---|
| Arena Eternal | 77% | 4.6 | 6.4% |
| Rhino Squad | 72% | 4.9 | 4.9% |
| Bitterman's War | 72% | 4.6 | 5.0% |
| Hold The Line | 69% | 5.1 | 6.4% |
| The Final Tier | 54% | 6.2 | 5.5% |
| Ranger's Run | 29% | 7.4 | 6.4% |
Each deck's wins split by the blow that ended the game. Commander damage matters most for the voltron decks; "pings & abilities" is the library's turret-heavy signature.
| Deck | combat damage | commander damage | pings & abilities | alternate win |
|---|---|---|---|---|
| Ranger's Run | 47.6% | 29.9% | 22.4% | 0.1% |
| Bitterman's War | 75.6% | 8.6% | 15.8% | 0.0% |
| Rhino Squad | 89.3% | 2.7% | 8.0% | 0.0% |
| Hold The Line | 89.1% | 1.5% | 9.3% | 0.0% |
| The Final Tier | 77.1% | 6.7% | 16.2% | 0.0% |
| Arena Eternal | 87.2% | 5.3% | 7.5% | 0.0% |
Row deck's win rate in games where the column deck was at the table. Blue = better than its overall rate, red = worse. Each pairing covers 40,002 games.
| Deck ↓ vs → | Ranger | Bitterman | Rhino | Hold | Final | Arena |
|---|---|---|---|---|---|---|
| Ranger's Run | — | 29.9% | 30.5% | 30.1% | 29.2% | 30.8% |
| Bitterman's War | 26.3% | — | 26.5% | 25.6% | 26.4% | 26.1% |
| Rhino Squad | 23.7% | 24.9% | — | 25.4% | 25.6% | 27.1% |
| Hold The Line | 22.9% | 23.5% | 23.8% | — | 24.6% | 24.1% |
| The Final Tier | 22.7% | 23.1% | 22.6% | 23.9% | — | 24.0% |
| Arena Eternal | 18.7% | 19.3% | 19.2% | 19.6% | 20.0% | — |
Share of decided games won by each seat position, hairline at the fair 25%. Going first is a real disadvantage in this engine (no draw on the first turn) — worth remembering when reading table results.
| Seat | Wins | Share |
|---|---|---|
| Seat 1 (first) | 20,513 | 20.8% |
| Seat 2 | 25,648 | 26.0% |
| Seat 3 | 26,027 | 26.4% |
| Seat 4 (last) | 26,488 | 26.8% |
Δwin = win rate when the card is seen in the first six rounds minus when it isn't (the six-round window removes the long-games-draw-more-cards bias). Top three pull the deck up, bottom three drag it down. Shared ±26-point scale. Full tables: quake/balance/report.md.
| Deck | Card | Δwin (pts) |
|---|---|---|
| Ranger's Run | Ranger, Slayer of Shub-Niggurath | +22.6 |
| Ranger's Run | Rune of Black Magic | +11.5 |
| Ranger's Run | Rune of Elder Magic | +7.8 |
| Ranger's Run | Phantom Swordsman | -4.1 |
| Ranger's Run | Zombie | -4.3 |
| Ranger's Run | Scrag | -4.4 |
| Bitterman's War | Black Widow Guardian | +15.1 |
| Bitterman's War | Force Field | +13.8 |
| Bitterman's War | Tank Commander | +9.4 |
| Bitterman's War | Quad Damage | -5.1 |
| Bitterman's War | Invulnerability | -5.2 |
| Bitterman's War | Medic | -5.4 |
| Rhino Squad | Voss, Strogg Commander | +14.0 |
| Rhino Squad | Matthew Kane, Reactivated | +6.1 |
| Rhino Squad | Makron | +5.8 |
| Rhino Squad | Strogg Gateway | -3.1 |
| Rhino Squad | Orbital Dispersal | -3.3 |
| Rhino Squad | Bidwell, Squad Technician | -4.3 |
| Hold The Line | Mobile Command Post | +17.4 |
| Hold The Line | Cyclops Heavy Walker | +9.7 |
| Hold The Line | Desecrator Hovertank | +6.8 |
| Hold The Line | GDF Engineer | -4.8 |
| Hold The Line | Orbital Strike | -5.2 |
| Hold The Line | GDF Field Ops | -5.8 |
| The Final Tier | Major, Squad Leader | +12.9 |
| The Final Tier | Hunter, Lord of Battle | +11.2 |
| The Final Tier | Cadaver, Rotting Contender | +8.3 |
| The Final Tier | Kamikaze | -4.3 |
| The Final Tier | Orbb, All-Seeing | -4.4 |
| The Final Tier | Respawn Protocol | -6.5 |
| Arena Eternal | Volkerh, the Green Rot | +25.5 |
| Arena Eternal | The Vadrigar, Arena Masters | +7.0 |
| Arena Eternal | The Lurching King | +4.6 |
| Arena Eternal | Super Nailgun | -2.9 |
| Arena Eternal | Ghost Walk | -3.0 |
| Arena Eternal | Super Shotgun | -3.4 |
Lift = a pair's actual win rate minus what the two cards' solo effects predict. Positive pairs are the deck's real engines; negative pairs fight each other for the same games. Top three and bottom two per deck, pairs seen together in 300+ games.
Outliers from the full run — the fastest kills, the biggest comebacks, and the longest grinds, with what the winner saw early.
Spearman rank correlation between each creature's in-game hit points and its card power+toughness. Hairline = the ρ ≥ 0.7 gate. quake3, quakewars and champions have no HP bestiary to rank against (arena roster / multiplayer sets), so the audit covers the three campaign sets.
| Set | Spearman ρ | Creatures matched |
|---|---|---|
| Quake II (QAO) | 0.89 | 27 |
| Quake 4 (QST) | 0.86 | 14 |
| Quake (QDD) | 0.73 | 22 |
Oracle templating lint + EDHREC salt cross-reference.
Templating. 10 of 898 rules sentences have no close printed-Oracle analog — all reviewed as intentional custom-mechanic composites; two genuine wording misses were fixed.
Human playtests, side by side with the sim's expectation. Nothing logged yet — this panel fills itself in as games are recorded.
0 games logged. To record a game:
quake/balance/playtests/TEMPLATE.md to a new file (one per game) and fill in winner, ending round, MVP and dead cards.quake/balance/playtests/SCHEDULE.md so table time covers all 15 pod combinations.python common/sim/run.py --ingest-playtests, then python common/sim/balance_dashboard.py --collection quake — the comparison table appears here.For Arena Eternal A/B nights, QCH-ARENA-ETERNAL-V1 in TTS Saved Objects is the pre-balance version.