feedback1 WB: 战术决策 review: independent R11 draws marked assumed (the slot pattern of the 4 drafts left open); twin battle passives stacking marked assumed in META

docs/DATA.md

@@ -324,7 +324,7 @@ Types: `BAND_INITIAL` 41, `BOND` 23, `EQUIP` 115, `ENEMY_GAIN` 129 (悬赏), `BU
 | `bountyDrafts[kind]` | `{"initial":{"events":[…],"slots":10,"pick":"slot","groups":[{"cards":[…6],"seen":[1,3,5,9,17]},…],"rule":{"series":[10,…,15,20],"tiers":[1,1,1,2,2,3],"perSeries":2,"prefer":["enemyeffect_10_6","enemyeffect_20_6"]},"count":6},"boss":{"events":[…],"slots":6,"pick":"seen","groups":[{"cards":[…9],"seen":[7,…,21],"weights":[12,13,…]},…],"count":6},"hunter":{"events":[…],"slots":7,"pick":"slot","groups":[{"cards":[…7],"seen":[1,6,21,22],"weights":[…]},{"cards":[…6],"open":1,…},…],"rule":{"size":7,"onePerSeries":[10,…,15],"maxSeries16":2,"giants":[…],"cards":[…]},"count":6}}` | the official 悬赏决策 card lists, read from 66 screenshots of 22 official co-op 绝境 / 终极 matches (player feedback after 0.1.0 report #2; build-data `BOUNTY_INITIAL_SETS` / `BOUNTY_BOSS_GROUPS` / `BOUNTY_HUNTER_GROUPS`, readings in `test/fixtures/official-bounty-drafts.json`, `seen` = its match numbers). A draft = one group (`pick` `slot`: one of `slots` events, uniform — a seen group or, for an unseen event, a list built by `rule`; `seen`: a group by the matches it came in), then `count` different cards of its `cards` drawn one by one by `weights` (1 + the drafts of the group the card showed in; a rule-built card 1). `initial`: 9 seen sets of 6 (all six shown), the 10th by `rule` — tiers I I I II II III, ≤ `perSeries` a series, the III one of `prefer` (the two-battle cards no draft showed) [ASSUMED]; `boss`: the 6 groups of 9, 9, 9, 9, 8 and 6 cards (one per bossInitial event; 9 where three named bosses came; the three single-draft groups completed only with the base cards they lack [ASSUMED] — match 4's base-only group may instead be a draft of the 泥岩 group), the 鼠王 group in 14 of 22 matches — every one on the dark grey board — so picked by `seen` [ASSUMED: the per-match cause is open]; `hunter`: the 7 seen lists of 7, `slots` 7 so one of them uniform, so no list is built from nothing [ASSUMED] (which of the 15 events R11 fires is open: by the blocks of effectChoiceInfoDict it would be bounty_hunter_8..15, beside artifact_paid_4 / 5 and hardbuff_select; with 8 events an unseen list would come in about 1 R11 bounty draft in 8); `open` = cards still unseen, drawn by `rule`: a giant for a list without one, else tier I / II cards, one per `onePerSeries` series, ≤ `maxSeries16` of 16_9..12 [ASSUMED] — the only way 12_8 / 16_2 / 16_6 (in no draft) come, about 1 R11 bounty draft in 15 |
 | `schedule[modeId]` | `{"spRounds":[3,6,9],"rounds":{"3":{"families":[{"family":"bounty","weight":50},…],"cards":6,"supplyTiers":[1,4],"bountyDraft":"initial","events":{"bounty":[…],"supply":[…],"shop":[…],"tactic":[…]},"assumed":true}}}` | research 01 A4 defaults: pick a family by weight; `supplyTiers` = item tier window for 道具补给; 机密商店 draws `shopDraft` at its `rounds` (R11), else any normal shop item of tiers I–VI per card (duplicates allowed); `bountyDraft` = the `bountyDrafts` kind of the round's 悬赏决策 (R3 / 险境 R6 `initial`, R9 `boss`, R11 `hunter`); 绝境 / 终极 R11 = 悬赏决策 14 / 机密商店 4 / 战术决策 4 (the official co-op R11s of 22 matches, no 道具补给; weights = those counts [ASSUMED]; solo takes the same weights [ASSUMED: no solo screenshot — extrapolated from co-op, the data has solo R11 悬赏决策 events `bounty_hunter_*_s`]); `events` = the official event ids of that kind (`enemy_initial_*`, `bossInitial_*`, `bounty_hunter_*`), usable as the phase header |
 | `shopDraft` | `{"rounds":[11],"slots":[{"6":1},{"6":1},{"5":1},{"coin":1},{"3":1,"4":2,"5":3,"coin":2},{"3":1,"4":2,"5":3,"coin":2}],"coin":"chess_item_1_03_e_a","itemWeights":{"chess_item_6_09_e_a":5,…},"seen":[2,4,5,8],"count":6}` | the official 机密商店 (R11 of matches 2, 4, 5, 8; build-data `SHOP_DRAFT`; the user: "机密商店按官方改成可以重复吧"): six slots, each drawn on its own — with replacement, so an item can come twice (official: 盟约之币 ×2, 变形同构体 ×2) — a tier (or `coin` = 盟约之币) by the slot's weights, then an item of that tier by `itemWeights` (1 + the official cards it showed on; unlisted 1). Every official shop: exactly two tier VI, ≥ 1 V, ≥ 1 盟约之币, no other tier I / II; the slot split and the weights [ASSUMED]; solo 3 of the 6 [ASSUMED]; only at `rounds` (R11 = 绝境 / 终极, the rounds of the screenshots) — the earlier 机密商店 of 标准 R3 / R9 and 险境 R3 / R6 / R9 (no screenshot; the data's shop events sit in three blocks) keep the previous draw, any normal shop item of tiers I–VI per card, with replacement [ASSUMED]; free (price 0) |
-| `tacticDraft` | `{"rounds":[11],"kinds":["ally"],"weights":{"allybuff_select_4":4,…},"seen":[7,9,18,20],"count":6}` | the official 战术决策 (R11 of matches 7, 9, 18, 20; build-data `TACTIC_DRAFT`; the user: "战术决策也按官方改成可以重复吧"): six cards, each drawn on its own — with replacement, so a card can come twice (official: 补给 ×2 in match 7; the other three drafts show six different cards). All 24 official cards are `ally` cards (no 排斥 / 责罚 / 裁决, no 模拟战场演变; four drafts without a debuff: about 0.04 % for the previous draw, 6 different cards of the 26 ally + 9 debuff cards), and 列装 / 财富 / 补给 / 整备 / 升华 made 12 of them (uniform: about 4.6), so at `rounds` (R11 = 绝境 / 终极, the round of the screenshots) only the cards of `kinds`, each by `weights` (1 + the official cards it showed on; unlisted 1) [ASSUMED: the weights; no terrain card at R11 — the maps of the 4 matches are unknown]; solo 3 [ASSUMED]; other rounds (标准 R3 / R9, 险境 R3 / R6 / R9, 险境 solo R9; no screenshot) every card uniform, terrain cards only for the match stage, with replacement too [ASSUMED]. A card acting only on dead bonds is never offered (`Match.bondLive`) |
+| `tacticDraft` | `{"rounds":[11],"kinds":["ally"],"weights":{"allybuff_select_4":4,…},"seen":[7,9,18,20],"count":6}` | the official 战术决策 (R11 of matches 7, 9, 18, 20; build-data `TACTIC_DRAFT`; the user: "战术决策也按官方改成可以重复吧"): six cards, each drawn on its own — with replacement, so a card can come twice (official: 补给 ×2 in match 7; the other three drafts show six different cards). All 24 official cards are `ally` cards (no 排斥 / 责罚 / 裁决, no 模拟战场演变; four drafts without a debuff: about 0.04 % for the previous draw, 6 different cards of the 26 ally + 9 debuff cards), and 列装 / 财富 / 补给 / 整备 / 升华 made 12 of them (uniform: about 4.6), so at `rounds` (R11 = 绝境 / 终极, the round of the screenshots) only the cards of `kinds`, each by `weights` (1 + the official cards it showed on; unlisted 1) [ASSUMED: the weights; no terrain card at R11 — the maps of the 4 matches are unknown]; each card independent, no slots like `shopDraft` [ASSUMED, open: all 4 drafts hold a 驰援, a 盟誓 and two of 列装 / 财富 / 补给 / 整备 / 升华 — about 3 % under independent draws, a pattern spotted after the fact]; solo 3 [ASSUMED]; other rounds (标准 R3 / R9, 险境 R3 / R6 / R9, 险境 solo R9; no screenshot) every card uniform, terrain cards only for the match stage, with replacement too [ASSUMED]. A card acting only on dead bonds is never offered (`Match.bondLive`). `assumed` lists the assumed parts |
 | `pools[poolId]` | `{"kind":"equip","items":[…]}` / `{"kind":"chess","tier":2,"rule":"shopEligible"}` | server-side reward pools referenced by effects (`pool_equip_*`, `pool_chess_shop_N_reward`, `pool_char_later`, …), all [ASSUMED]; `rule:"shopEligible"` = visible/non-hidden entries, `maxTier:"shopLevel"` = ≤ current shop level, `weighted` = `[[id, weight]]`. `pool_equip_vict` (维多利亚, every 25 layers: "获得一件带有随机特殊效果的维式重锤") and `pool_equip_rockr` (洛洛's 定制品; user playtest #4 — the research 04 §8 guess 有限加速器 / 激光发射器 / 护盾无人机 / 双模机械臂 / 蜂鸣器 is superseded) = the 4 special 维式重锤, uniform |
 
 ## 9. `enemies.json` — `{ [enemyKey]: Enemy }` (249 keys that can appear)