Runs all 9 × 9 directional matchups, including mirrors. Results describe the neutral simulator policy—not player skill.

No simulation is running.

Run the lab to generate matchup, turn-length, utilization and stall telemetry.

Paired mirror testing of the current initiative rules against the second-player +1 opening-card compensation. No card balance changes.

Issue #83: why Manifestations die before they act. Runs all 81 matchups × 10 seeds under each candidate rule (Strike Back, Arrival Ward, Arrival Shield, Bender Arts, Ascension) and reports survival, Card Depletion, duel length, first-player edge and deck spread. Simulator only: live duel rules are unchanged.

Finalist calibration for paired mirror matches. Compares baseline, second-player +1 Essence, and graduated first-player Round-2 attack caps while recording role/physical-seat pairing, per-element lifecycle, sequencing, first-removal correlation, and representative traces. Live duel rules and card balance are unchanged.

Runs deterministic paired seeds with the decks swapped between Side A and Side B. Mirror controls isolate first-player/seat advantage before card balance conclusions are trusted.