Divisions 1945 — Omega War

1.1 The spawning instruction (route record + the ARQ5 deferral)

This cycle is the route of record. The arsenal-1945 Stage-1 final YAML answered the route gate continue-divisions-cycle: "a second staged mini-cycle (own scope page → sweep + packet → canonical, e.g. research/divisions-1945.md) determining which divisions players can build decks from and what units go in each; /game-genre-map after that." The arsenal Stage-2 final YAML then handed this cycle its hardest question — the ARQ5 phase & deck-mapping gate was answered rework-again with the user's note recorded verbatim in research/arsenal-1945.md (§2 ARQ5 round 2, §4 status banner):

"We should punt this to the divisions research and then we can determine how to split up how divisions should deploy units over the course of the battle and how we construct decks over the course of a potentially short or long term campaign."

That deferral makes this cycle the owner of phase/deck construction: the arsenal artifact carries only research inputs and an unadopted two-axis proposal (see §1.3).

1.2 Approved fantasy anchors that constrain this study scope evidence

• The five-front set and scope ladder are approved (research/game-fantasy.md, Theaters & Fronts, approve-set-and-ladder): 1. Central Germany (Saxony–Thuringia zonal line) — base game; 2. North German Plain & Baltic — DLC #1; 3. Czechoslovakia–Austria–Hungary — DLC #2; 4. Balkans–Adriatic — DLC #3; 5. Scandinavia (Norway · Denmark) — DLC #4. Each is a real Aug-1945 contact line or standoff. Map representation, token granularity, and pacing remain deferred to the campaign design pass.
• Pillar 1 — Command weight: "decisions feel consequential because losses persist" — persistent veterancy and casualties across the campaign. Division persistence mechanics need the historical replacement/refit reality this study gathers (DRQ6).
• Pillar 2 — The plan comes together: deck-as-roster with build-order escalation is the payoff engine, and "the arsenal history shelved is the reward shelf." Decks are authored from divisions; this study supplies the divisions.
• Layer-handoff guardrail: the operational fantasy must not become "an entirely other game" (the SD2 Army General complaint) — division identity must read in both the tactical battle and the campaign map.
• War start: spark 14 July 1945; the shooting war opens August 1945 — the order-of-battle snapshot date for this study.
• Tone register B + clean-Wehrmacht guardrail (F-A4): applies with full force to how re-armed German formations are named and described (Scandinavia front puts this "at maximum sharpness").

1.3 Arsenal-1945 anchors (confirmed artifact) scope evidence

All from research/arsenal-1945.md (canonical, 2026-06-10). Equipment evidence is [carried] — this cycle reuses it and never re-collects it.

• Per-faction equipment tables + plausibility tiers (accepted/adopted): §3 tables for US, UK/Commonwealth, USSR, re-armed Germans, France, exiles, satellites; ARQ2 tier vocabulary (in-theater / in-production / production-ready / prototype-in-window / paper). Deck-candidate divisions link to these tables rather than re-deriving equipment.
• §4A card-class taxonomy (ACCEPTED): the WARNO-card-style unit-type frame (line squad → recovery vehicle, per faction) is the canonical unit-type frame — but "To&E counts and per-division card counts are explicitly deferred to the historical-divisions cycle." DRQ2 picks that up.
• §4B zone dial + one-wonder-slot (ADOPTED): the three-zone historical-accuracy dial and Option C (exactly one wonder card per deck, Phase-C only, named documented artifacts) are design rules of record. Division/deck proposals here must respect them.
• Era-band E0–E2 campaign-availability proposal (UNADOPTED input): E0 = standing kit Aug–autumn 1945; E1 = in-production winter 45–spring 46; E2 = reward shelf toward mid-46 — carried as a research input only, deferred with the phase model. The design proposal gated in this cycle (§2.2) must resolve whether to adopt, amend, or replace it.
• ARQ4b — US Pacific rotation: documented roster pressure (~1M Category II earmarks incl. 13 infantry + 2 armored divisions; First Army HQ Downfall-bound). DRQ4 deepens this to named-division granularity.
• Gaps 1–16 are live caveats, especially gap 1 — German depot-stock inventory totals in Western zones (no quantified source found; re-armed-German scale capped at medium confidence). The re-armed-division shortlist in DRQ3 inherits that cap until new evidence moves it.

1.4 Idea-brief anchors scope evidence

• Company-scale battles: the tactical layer fights at company scale (CoH × Broken Arrow); divisions are the deck/identity container above it.
• Faction split: UK/Commonwealth, US, France, re-armed German formations, plus western exile formations (Polish II Corps, 1st Polish Armoured Division, Czechoslovak Armoured Brigade) versus the USSR and its satellites (Polish People's Army, Romanian/Bulgarian forces, the extrapolated Hungarian contingent — reliability footing per the arsenal §1.4/Claim 4.4 ruling).
• "Divisions persist and evolve across the campaign" — the brief's persistence pillar at division granularity; DRQ6 gathers its historical basis.
• Formation-token granularity (company vs battalion) is explicitly open — owned by the campaign design pass, not this cycle. This study describes historical structures; it does not decide token granularity.

1.5 User scope decisions for this cycle (this session, 2026-06-11)

Two decisions taken in chat before this page was written; recorded here as the scope they are:

• All five fronts at equal full depth — not Central-Germany-only with later-front headline depth (this supersedes the arsenal cycle's by-front depth rule, AR-A3, for this cycle; see DR-A2). User, verbatim: "it's fine if it goes stale, bc we'll have a starting point… it's history, we can pick and choose but I'm sure the overall research will have value."
• The punted ARQ5 becomes research + a gated design proposal — following the wonder-framework precedent from the arsenal cycle (research the evidence, then put a concrete design proposal behind its own Stage-2 gate, adopted only if the user answers adopt). DRQ5 is the research half; the Phase/Deck-Construction Design Proposal (§2.2) is the design half, opt-in via its own gate.

2.1 Research questions

2.2 Deliverable shape (canonical artifact contract)

• Per-front order-of-battle tables — all five fronts, army → corps → division, both blocs, with citations (DRQ1).
• Per-faction To&E patterns mapped to the accepted §4A card classes, counts recorded as schema inputs (DRQ2).
• Named-division deck-candidate shortlists per faction, each linked to the arsenal equipment tiers and tagged for identity/veterancy/reliability (DRQ3).
• Roster-dynamics timetable — who leaves/stays/arrives 1945–46, per faction, division granularity (DRQ4).
• Phase/Deck-Construction Design Proposal — the design half of the ARQ5 deferral, gated separately at Stage 2 (wonder-framework precedent): a concrete proposal resolving in-match deployment bands (how divisions deploy units over the course of the battle), short- vs long-campaign deck evolution (how decks are constructed over a potentially short or long campaign), and the carried era-band E0–E2 question (adopt / amend / replace). Adopted only if its gate is answered adopt; otherwise it remains a recorded proposal.
• Gaps register — what the sweep could not establish, with confidence caps; arsenal gaps 1–16 carried forward where touched.
• Citations throughout, with confidence tags ([high]/[medium]/[carried]; inference labeled), matching the arsenal artifact's evidence discipline.

2.3 Non-goals

• No unit stats or balance design — division identity, structure, and evidence only.
• No unit/division data schema authoring — this study produces the schema's research inputs, not the schema (AR-A5 carried as DR-A4).
• No campaign-map or formation-token design — token granularity (company vs battalion), map representation, and campaign pacing belong to the campaign design pass (idea-brief deferral, §1.4).
• No genre-convention mapping — that is /game-genre-map, the next step on the standing route.
• No art/asset sourcing (insignia, photos, silhouettes).
• No re-collection of arsenal equipment evidence — reused as [carried] where valid.
• The Phase/Deck-Construction Design Proposal is the only design output, and it is opt-in via its own Stage-2 gate — everything else in the artifact is research.

Availability is a plan, not verified coverage. Method note carried from all three prior cycles: research subagents have no web access in this environment; the sweep runs in the main session. Stopping rule per the depth chosen at the coverage checkpoint below. scope evidence

7. Compile