From 97f5cbbcae4c9216271c27f3c464cfcd7e6b8cf2 Mon Sep 17 00:00:00 2001 From: slaguru666 <111923774+slaguru666@users.noreply.github.com> Date: Sun, 13 Sep 2026 14:41:39 +0100 Subject: [PATCH] Desk playtest 5: the four-player cut, played for the first time MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Seed 8815, four players — Ashcroft, Bhattacharya, Renshaw, Braithwaite. Every rating pulled from roster.mjs before rolling, which is the direct consequence of pass 4 having typed six of sixteen from memory. The cut is the most measured thing in this project and had never been played. EXPOSURE prices its fights, check-firstblood holds its swing, check-attackers holds what each of the four contributes -- and no pass in four had run a session with Pollard and Okonkwo off the table. THE RESULT IS NOT ABOUT COMBAT. The fights were never the problem. What breaks is the party's eyes: POLLARD'S SPOT 53 IS THE HIGHEST IN THE ENTIRE ROSTER AND HE IS THE FIRST PERSON THE SCALING NOTES DROP. At five players the party's eyes go from 53 to 40; at four they stay at 40 with a 35 alongside. This run missed the stump (79 vs 40) and the six at the back (67 vs 35), both Pollard's lane in the six-player game, both clues the scenario leans on. check-rollable cannot see this, and the reason is worth keeping: it asks whether every named skill is reachable at 25% or better, and Spot 40 clears 25 comfortably. It is a rollability check, not a competence check, and the cut is where the difference bites. The scaling note's only stated cost of dropping Pollard is that "the cordon loses its shield", which is about a fight, in a scenario whose first two acts are almost entirely looking at things. Second finding: FOUR OF THE FIVE PARTY-SPLIT ROWS NAME PEOPLE WHO ARE NOT THERE. The table is introduced as the fallback for a table that will not choose its own splits -- and at four players the fallback does not exist, which is exactly when it is most needed. Timing runs the other way from pass 4: ~3:15, twenty-five minutes SHORT, because four people ask fewer questions. Nothing in the Pacing Note says the cuts are player-count dependent, so a GM following it at a small table finishes early. The unsettled countdown fired for a second pass running. Recorded with the caveat that both passes had Ashcroft accept THE OFFER, so both used the Difficult branch -- 14.5% unsettled against 3.8% if he refuses. Not two draws from the same distribution, and the document prints only one of those figures. Five fixes listed, none applied. Also listed: what five passes have still never tested -- the Swinburne fumble, the re-cut-the-peg fumble, a failed Xenology, and any fight at all. npm run check: 17 guards, exit 0. Co-Authored-By: Claude Opus 5 --- docs/scenarios/CLEAN_GROUND_PLAYTEST_5.md | 228 ++++++++++++++++++++++ 1 file changed, 228 insertions(+) create mode 100644 docs/scenarios/CLEAN_GROUND_PLAYTEST_5.md diff --git a/docs/scenarios/CLEAN_GROUND_PLAYTEST_5.md b/docs/scenarios/CLEAN_GROUND_PLAYTEST_5.md new file mode 100644 index 0000000..690d0ec --- /dev/null +++ b/docs/scenarios/CLEAN_GROUND_PLAYTEST_5.md @@ -0,0 +1,228 @@ +# CLEAN GROUND — desk playtest 5 (the four-player cut, played) + +**Scenario version:** v0.14, at commit ae06df8 +**Date:** 13 September 2026 +**Method:** full session, GM chair, **four players — the cut**: Ashcroft, Bhattacharya, +Renshaw, Braithwaite. Seeded roller, **seed 8815**, banding from `rules.mjs`. Earlier +passes were 4271, 9314, 5580/7731, 6142. + +**Every skill rating in this pass was pulled from `roster.mjs` before rolling**, which is +the direct consequence of pass 4 having typed six of its sixteen from memory and got three +outcomes wrong. See that document's second post-pass. + +**Why this pass.** The four-player cut is the most *measured* thing in this project and has +never been **played**. EXPOSURE prices its fights, `check-firstblood` holds its swing, +`check-attackers` holds what each of the four contributes, and the scaling notes tell a GM +how to build it — and no pass in four had ever run a session with Pollard and Okonkwo off +the table. + +**Verdict: the cut runs, and the fights were never the problem. What breaks is quieter — +the party's eyes. Dropping Pollard takes the best Spot in the entire roster off the table +and nothing in the document says so, because the guard that watches this only asks whether +a roll is possible, not whether it is any good.** + +--- + +## The rolls, in order + +``` +seed 8815 + Bhattacharya Research — Corrigan's paperwork 15 vs 63 success + Braithwaite Medicine — the body 60 vs 53 fail + Renshaw Persuade — Swinburne 60 vs 63 success + Ashcroft Spot — the stump, no Pollard 79 vs 40 fail + Ashcroft Listen — the valley, crossing 35 vs 63 success + Renshaw Persuade — the column 17 vs 63 success + Braithwaite Spot — the six at the back 67 vs 35 fail + Ashcroft Insight — it is offered, not remembered 98 vs 63 fail + Bhattacharya Anomaly Lore — the depot entry 16 vs 63 success + Renshaw Insight — Ivy on the almanac 14 vs 63 success + Ashcroft Listen — the lorry, Pollard's lane 15 vs 63 success + Braithwaite Xenology (Baseline) — the understudy 38 vs 53 success + Renshaw Insight — a colleague who is not 47 vs 63 success + Ashcroft First Aid — whoever went down 40 vs 40 success + + COUNTDOWN STEP 5 — opposedContestFor + understudy 75 rolled 91 failure + Ashcroft 85 -> 42 Diff rolled 84 failure + → winner none · settled FALSE +``` + +A friendly seed, deliberately not re-rolled after pass 4's hostile one. **A cut that only +survives good dice is not a cut**, and this run's failures land in the two places that +matter most: the body, and the six at the back. + +--- + +## ⚠ FINDING 1 — the cut cannot see + +This is the pass's real result and it has nothing to do with combat. + +| | Spot | Listen | +|---|---|---| +| Ashcroft | 40 | **63** | +| Bhattacharya | 40 | — | +| Renshaw | 40 | **63** | +| Braithwaite | **35** | — | +| *Pollard (dropped at five)* | ***53*** | *53* | +| *Okonkwo (dropped at four)* | *35* | — | + +**Pollard's Spot 53 is the highest in the entire roster, and he is the first person the +scaling notes drop.** At five players the party's eyes go from 53 to 40. At four they stay +at 40 with a 35 alongside. Every Spot clue in the scenario becomes materially harder in the +configuration the document recommends for a smaller table. + +- This run: **Ashcroft missed the stump** (79 vs 40) and **Braithwaite missed the six at + the back** (67 vs 35). Both are Pollard's lane in the six-player game, both are at 53 + there, and both are clues the scenario leans on. +- ⚠ **The guards cannot see this and it is worth understanding why.** `check-rollable` + asks whether every skill the acts name is reachable at 25% or better by somebody in the + declared cast. Spot 40 clears 25 comfortably, so the build is green and stays green. It + is a **rollability** check, not a **competence** check, and the cut is the case where + the difference bites. +- The scaling note says of dropping Pollard only that *"the cordon loses its shield"* — + which is about Act Four's fight. **The real cost is that the party loses the only pair of + eyes above 40, in a scenario whose first two acts are almost entirely looking at + things.** + +**Fix direction:** say it in the scaling note, and give the smaller table the compensation +it is owed. The cleanest is already half-written elsewhere — the column's **special** +result now hands over something portable (v0.14), and the same logic applies here: at four +players, the clues Pollard would have spotted should be **given, not rolled for**. The +lorry, the six at the back, the stump. A four-player table that has to roll Spot 40 three +times to assemble the loop will assemble it late or not at all. + +--- + +## ⚠ FINDING 2 — four of the five split rows name people who are not there + +The party-seams table is written for six and is **unusable at four**: + +| Act | Split as written | At four players | +|---|---|---| +| One | Braithwaite + **Pollard** · Renshaw + Bhattacharya | ✗ | +| Two | Renshaw + Ashcroft · **Okonkwo + Pollard** | ✗ — the whole second half is gone | +| Three | Bhattacharya + Braithwaite · **Pollard** + Ashcroft | ✗ | +| Four | Do not split | ✓ | +| Close | **All six**, at the stump | ✗ — wording only, but it is wrong | + +The table is introduced as *"let the table choose; this is the fallback"* — and the +fallback is the thing a GM reaches for when the table will not choose. **At four players +the fallback does not exist**, which is the moment it is most needed, because four players +splitting 2+2 leaves two pairs with no overlap at all. + +This is not a hard problem. It needs one extra column or four extra words per row. + +--- + +## Act-by-act + +**Act One — ran ~40.** Faster than six players, as expected: four people ask four +questions, not six. Bhattacharya's Research landed and Corrigan's unfiled casual labour +came out early, which is the strongest opening this scenario has had in five passes. +Braithwaite failed Medicine on the body and the coroner's note carried it instead — +Renshaw's Persuade on Swinburne succeeded, so H01 came from the drawer as written. + +- The new fumble branch for Swinburne (v0.14) did **not** fire. Still untested. + +**Act Two — ran ~40 against a 45 budget, and this is the first time an act has come in +UNDER.** With four players the line's rest has three people in it, because Ashcroft is +aside for THE OFFER. Twenty-five minutes of conversation between Ivy and three agents ran +out at about twenty. + +- ⚠ **Minor: the rest clock is sized for six.** It is not a defect — the old ones standing + up early is fine, and no GM will complain about an act finishing early. But a GM watching + a 25-minute budget with three players talking will feel they are doing it wrong. One + clause would fix it: *at four players the line rests twenty minutes, because there is + less to say.* +- THE OFFER fired and Ashcroft **failed** the Insight (98) again — accepted believing it + was a real memory. Two passes running. + +**Act Three — ran ~35 with the seam cut taken.** Bhattacharya's Anomaly Lore succeeded, so +no re-cut-the-peg chase; that clause remains untested after five passes. Renshaw read Ivy, +and **Ashcroft's Listen 63 picked up the lorry** — Pollard's clue, inherited exactly as the +scaling note promises. That part of the scaling works and should be said more loudly. + +**Act Four — ran ~45.** Braithwaite's Xenology succeeded, so the tell arrived as a quiet +word in the GM's ear at beat 2, which is what the text wants. Renshaw's Insight on the +colleague-who-is-not also landed, so the cut was actually *better informed* here than pass +4's six-player table. + +### The unsettled countdown fired again — second pass running + +Understudy 91 (failure) against Ashcroft's 84 on the Difficult branch (failure). **Neither +succeeded. Unsettled.** Two passes, two unsettled outcomes on a 14.5% branch — that is +luck, not a pattern, and it should not be read as one. + +- Worth noting all the same: because Ashcroft accepted THE OFFER in both passes, **both + runs used the Difficult branch, which is the branch with the highest unsettled rate.** + 14.5% against 3.8% if he refuses. So the two are not independent draws from the same + distribution, and a scenario where Ashcroft usually accepts will see this outcome roughly + four times more often than the headline figure suggests. +- The read-aloud landed for a second time and needs nothing. + +**Close — ran ~25.** On budget. Four players at the stump is a tighter scene than six and +arguably a better one: nobody is spectating. + +--- + +## Timing + +| Act | Budget | Ran | Δ | +|---|---|---|---| +| One | 45 | ~40 | −5 | +| Two | 45 | ~40 | −5 | +| Three | 50 | ~35 | −15 *(seam cut)* | +| Four | 45 | ~45 | — | +| Close | 25 | ~25 | — | +| **Play** | **210** | **~185** | **−25** | +| **Wall clock** | **3:40** | **~3:15** | **−25** | + +**Four players runs about twenty-five minutes short**, which is the opposite problem and a +much better one to have. It also means **the four-player table can afford to keep the +northern seam** — the cut that pass 4 made compulsory is not needed here at all. + +- ⚠ **Nothing in the Pacing Note says the cuts are player-count dependent.** At six they are + the default; at four, taking them all leaves a 3:15 session in a four-hour slot. A GM + following the current instruction at a small table will finish forty-five minutes early. + +--- + +## Player-hat verdicts + +| Archetype | Verdict | Score | +|---|---|---| +| Methodical veteran | Ashcroft inheriting the Listen lane worked invisibly, which is the compliment | 8/10 | +| Roleplay-first talker | Three-way conversation with Ivy was better than six-way. Everyone got a turn | 9/10 | +| Quiet tactician | Spot 40 twice in two acts, both missed. "We are the party that cannot see" | 5/10 | +| Lore-digger | Research landing in Act One set the whole night up. Best opening yet | 9/10 | + +--- + +## Fix list + +1. **⚠ Say that the cut cannot see, and compensate it.** *(Roster-level notes)* Pollard's + Spot 53 is the roster's best and he is dropped first. State the cost plainly, and + instruct that at four players the clues in his lane — the stump, the six at the back, + the lorry — are **given rather than rolled for**. `check-rollable` will not catch this + because it tests the 25% floor, not competence. + +2. **⚠ Make the split table work at four players.** *(Roster-level notes)* Four of five + rows name Pollard or Okonkwo. Add a four-player column, and fix "All six, at the stump" + to "everyone". + +3. **Tell the Pacing Note that the cuts are for six.** At four players the session runs + ~3:15 with them all taken. Say: at four, keep the northern seam; take cuts only if the + table is slow. + +4. **Size the line's rest to the table.** Twenty-five minutes is a six-player number. + Twenty at four, and say why — there are three people talking, because Ashcroft is aside. + +5. **Record that the Difficult branch raises the unsettled rate fourfold.** *(Act Four)* + 14.5% against 3.8%. A GM whose Ashcroft accepted THE OFFER should expect the reach to + come to nothing far more often than the countdown's headline suggests, and the document + currently prints only the one figure. + +**Still untested after five passes**, and worth stating so it is not mistaken for tested: +the Swinburne fumble branch, the re-cut-the-peg fumble, a failed Xenology in Act Four, and +any fight at all — no pass has yet had the party shoot anybody.