diff --git a/docs/REVIEW_LOG.md b/docs/REVIEW_LOG.md index f53aa0d..375e736 100644 --- a/docs/REVIEW_LOG.md +++ b/docs/REVIEW_LOG.md @@ -7048,3 +7048,34 @@ from a neighbouring assertion. *(Committed as a5c4378 under the subject "R-294", which was already taken by the derived source. The number here is the correct one; the commit subject is wrong and is left alone rather than rewritten.)* + +## R-297 — the step-5 figures now resolve against the rule, all thirty-two of them + +R-294 built the mechanism and R-295 derived the counterfactual; this places the markers. +Every figure in CLEAN GROUND's step-5 material now carries a citation resolving against +`tools/step5-split.mjs`, which enumerates rather than stores. + +**Thirty-two markers across fourteen lines**, covering the countdown rows, the two-case +table and its blended row, the Act Four reasoning, the counterfactual pair and the POW×5 +targets themselves. `refused.*`, `cut.*`, `accepted.*`, `hadHeRefused.*`, `blended.*`, the +63/37 shares, the targets and the thing's own 75 — nothing printed in that material is now +unheld. + +**No prose changed.** Stripping every HTML comment from the result and diffing against HEAD +gives a byte-identical file, which is the check worth having when editing somebody else's +document: the markers are invisible to a reader and the sentences are exactly as c0 wrote +them. + +**Proved live rather than assumed.** In a worktree, raising Okonkwo's POW from 11 to 13 makes +Braithwaite the lowest in the room; **eleven of the thirty-two markers go red at once**, each +naming the figure, the path and the fact that nothing can be re-recorded because the number is +enumerated from `rules.mjs` on this build. That is the whole point of the exercise: the +document is now wrong in a way the build can see, in the same breath as the rule moving. + +**Why thirty-two and not a representative few.** A figure without a marker is a figure nothing +holds, and the three that went bad in this document — 74.3%, 3.8%, "nearly quadruples" — were +all in prose rather than tables. The prose is where it drifts, so the prose is marked. + +Numbering: R-294 exists twice in commit subjects (fb6e2cf mine, a5c4378 c0's) and once in this +log. c0 took R-296 and left the wrong subject standing on purpose — rewriting a pushed subject +is worth less than the entry that points at it. diff --git a/docs/scenarios/CLEAN_GROUND.md b/docs/scenarios/CLEAN_GROUND.md index 311017b..cf82335 100644 --- a/docs/scenarios/CLEAN_GROUND.md +++ b/docs/scenarios/CLEAN_GROUND.md @@ -308,8 +308,8 @@ Advance one step per act, plus one whenever the agents spend Coherence on the fa | **3** | It takes an NPC. The NPC's file is still good, so the copy holds | | **4** | The copy is fluent. It is inside the cordon and helping | | **5** | It reaches for **the lowest POW in the room** — opposed POW×5 (it is **75**, not 100), permanent. **Unless Ashcroft accepted THE OFFER in Act Two, in which case it takes him instead and he resists at Difficult.** GM ESSENTIALS has the resolution; Act Four beat 3 has the reasoning | -| **5 — they hold** | ⚠ **THE LIKELIEST RESULT, and the one this document used to leave blank.** If the agent's band beats the thing's — or ties it, because a tie goes to whoever is being acted upon — **the reach simply fails and the person is still themselves.** **48.1%** against Okonkwo, **52.4%** against Braithwaite, **74.3%** against an Ashcroft who refused. This is **not** the unsettled row below and must not be played as it | -| **5 — unsettled** | **If neither side succeeds, the reach fails and the row is spent.** Common enough to plan for: **11.3%** against Okonkwo, who is the target at six players; **10%** against Braithwaite, who is the target in the four-player cut; **14.5%** against an Ashcroft who accepted. **THE OFFER raises it by about a quarter, not fourfold** — see below. It does not retry; it is an outcome, not a gap | +| **5 — they hold** | ⚠ **THE LIKELIEST RESULT, and the one this document used to leave blank.** If the agent's band beats the thing's — or ties it, because a tie goes to whoever is being acted upon — **the reach simply fails and the person is still themselves.** **48.1%** against Okonkwo, **52.4%** against Braithwaite, **74.3%** against an Ashcroft who refused. This is **not** the unsettled row below and must not be played as it | +| **5 — unsettled** | **If neither side succeeds, the reach fails and the row is spent.** Common enough to plan for: **11.3%** against Okonkwo, who is the target at six players; **10%** against Braithwaite, who is the target in the four-player cut; **14.5%** against an Ashcroft who accepted. **THE OFFER raises it by about a quarter, not fourfold** — see below. It does not retry; it is an outcome, not a gap | | **6** | It goes south at dusk wearing whatever face it has. **Read the Close's "leave it filed" note: in a one-shot this lands tonight, not as a sequel** | ### Step 5 has three outcomes, and the one nobody expects is a person holding @@ -320,19 +320,19 @@ Advance one step per act, plus one whenever the agents spend Coherence on the fa | What actually happens at step 5 | How often | Takes them | **They hold** | Unsettled | |---|---|---|---|---| -| **Ashcroft refused** → it reaches for **Okonkwo, POW×5 55** | **63%** of games | 40.7% | **48.1%** | 11.3% | -| **Ashcroft accepted** → it reaches for **Ashcroft at 42** | 37% | 48.7% | **36.8%** | 14.5% | -| **Across all games** | | 43.6% | **43.9%** | 12.5% | +| **Ashcroft refused** → it reaches for **Okonkwo, POW×5 55** | **63%** of games | 40.7% | **48.1%** | 11.3% | +| **Ashcroft accepted** → it reaches for **Ashcroft at 42** | 37% | 48.7% | **36.8%** | 14.5% | +| **Across all games** | | 43.6% | **43.9%** | 12.5% | ⚠ **Ashcroft at his full 85 is NOT on this table, and that is not an omission.** If he refused he is the hardest file in the room and the thing does not reach for him at all — it reaches for -the easiest. The **21.9% / 74.3%** pair printed in Act Four is a *counterfactual*: what it would +the easiest. The **21.9% / 74.3%** pair printed in Act Four is a *counterfactual*: what it would cost him if it tried, quoted only to price what accepting THE OFFER sold. **No table ever rolls it.** Desk pass 8's fix list got this wrong and a second reader repeated the error straight out of it, which is why this note is here. - **If the target is Braithwaite instead** — a table that cut Okonkwo, or a GM who reads the - room differently — it is 37.6% / **52.4%** / 10.0% at his POW×5 60. + room differently — it is 37.6% / **52.4%** / 10.0% at his POW×5 60. - **Across the whole population of games they hold very slightly more often than they are taken**, and a GM who has been assuming the countdown lands has been assuming wrong. Desk pass 8 rolled it and found the document had nothing to say. @@ -360,18 +360,18 @@ which is the one expression it has never rehearsed. #### WHEN IT COMES TO NOTHING — the unsettled row, which is a different outcome -**And it will, about one time in seven against an Ashcroft who accepted — 14.5%, against -11.3% for the Okonkwo the thing would otherwise have reached for.** The mechanism is the +**And it will, about one time in seven against an Ashcroft who accepted — 14.5%, against +11.3% for the Okonkwo the thing would otherwise have reached for.** The mechanism is the Difficult that makes Dominic easy to take: a resisting side at 42 succeeds far less often, so *both* sides missing becomes likelier. Accepting THE OFFER makes him both the likeliest person to be taken **and** a slightly likelier reason the beat resolves into nothing. A GM whose Ashcroft accepted should expect this outcome, not be surprised by it. > ⚠ **This document said "nearly quadruples" for five versions, comparing 14.5% against -> 3.8%.** The 3.8% is the unsettled rate against Ashcroft at his undiminished 85 — **a +> 3.8%.** The 3.8% is the unsettled rate against Ashcroft at his undiminished 85 — **a > contest that never happens**, because if he refuses the thing reaches for the lowest POW > in the room instead. Against what a table actually rolls the factor is **1.29**. The trade -> THE OFFER describes is real and the *taken* rate genuinely moves 40.7% to 48.7%; only this +> THE OFFER describes is real and the *taken* rate genuinely moves 40.7% to 48.7%; only this > one consequence was overstated, and it was overstated by three times. The rule returns *unsettled*, which is not the agent winning. Do not re-roll it and do not quietly hand the understudy the win. **Play the failure, because it is the most frightening thing the creature does all night:** @@ -957,7 +957,7 @@ them. **Anyone who examines him properly finds this, with no roll.** for somebody in front of the whole cordon. The tell arrives as an event rather than a finding, which is worse for them and perfectly playable. - **Who it reaches for: the lowest POW in the room.** For the cast as written that is - **Okonkwo, POW×5 55%** — and Ashcroft is the *hardest* at 85%. It has practised for thirty + **Okonkwo, POW×5 55%** — and Ashcroft is the *hardest* at 85%. It has practised for thirty years and it is not stupid: it takes the easiest file. The party's muscle being the easiest thing in the room to take is a better scene than the anchor being it. - **Exception — if Ashcroft accepted THE OFFER in Act Two, it reaches for him instead**, @@ -968,10 +968,10 @@ them. **Anyone who examines him properly finds this, with no roll.** refuses about five times in eight** — desk passes 6 and 7 both did, on honest rolls. The accepting figures are the dramatic ones and they are the minority case. **Read the refusing numbers first and treat the accepting ones as the branch you may not get.** - - **What it cost, exactly.** Refusing, Ashcroft is taken **21.9%** of the time — the + - **What it cost, exactly.** Refusing, Ashcroft is taken **21.9%** of the time — the safest person in the room. Accepting, **48.7%**. Accepting does not make him *a bit* more vulnerable; **it moves him from the hardest file at the table to the easiest, past - Braithwaite's 37.6%**, and the countdown reaches for the easiest. The offer was never a + Braithwaite's 37.6%**, and the countdown reaches for the easiest. The offer was never a trap, it was a trade, and this is the half of it nobody quoted him. - *Enumerated rather than sampled — all 10,000 roll pairs through `opposedContestFor`, so these are exact and there is no seed involved. Reproduce with the rule itself.*