# Findings so far

Part of the [shape exploration](README.md). Written 2026-10-02, after the pool was mapped ([04-assess.md](04-assess.md))
and nine small indie successes were placed on the same axes ([map.html](map.html),
[reference-games.md](reference-games.md)).

Each finding carries one of three grades:

| Grade | Meaning |
|---|---|
| **Confident** | stated by a source, or a plain observation that doesn't depend on how Claude cut or placed things |
| **Likely** | a pattern visible in the material and supported from a second side, but resting on Claude's placements or on a reading of intent |
| **Assumption** | an inference nobody has checked; treat it as a question |

Nothing here is a decision about the game.

## The main finding: core and supporting loops are spaced on purpose

**The user's finding (2026-10-02):** core and supporting loops "are spaced on purpose, they are part of the
tension/release cycle, but it's more like a swing between pillars, so that the player can take a rest from the
difficulty of each pillar in the other."

**Grade: likely.** The spacing is there to see; that it is on purpose, and that its job is rest, is a reading.

**What the maps show.** In six of the nine games the core loop has a partner loop far from it on the
cerebral–kinetic axis: one is done with the hands, the other with the head. The gap is in axis units, where the whole
axis from Cerebral to Kinetic is 2.

| Game | One pillar | The other pillar | Gap |
|---|---|---|---|
| 20 Minutes Till Dawn | move and shoot | pick an upgrade | 1.12 |
| Brotato | dodge while the weapons fire | shop between waves | 0.88 |
| Forager | gather | craft and build the base | 0.84 |
| Minami Lane | arrange the street | finish the missions | 0.72 |
| Islanders | place a building | end up with a pretty island | 0.66 |
| Kabuto Park | catch bugs | fill the collection | 0.62 |
| A Short Hike | roam the island | collect golden feathers | 0.20 |
| Stacklands | stack cards | feed everyone before the moon ends | 0.20 |
| SUMMERHOUSE | place pieces | soak in the mood | 0.28 |

**What supports the reading.**

- Guide [07](../../guides/game-design/07-keeping-players-engaged.md): a game has several pillars, each with its own
  intensity, and good pacing swaps between them before the player tires of one.
- The user's own observation on Noita ([brief.md](brief.md)): the tension–release cycle sits "not in difficulty, but
  viable choices. Sometimes you just shoot with your loadout, but sometimes you stop and design it."
- In Brotato and 20 Minutes Till Dawn the two pillars alternate on a fixed beat (a wave, then the shop; a stretch of
  dodging, then an upgrade), which is hard to explain as an accident.

**What limits it.**

- The positions, and so the gaps, are Claude's placements. The user checked the games they have played and found
  nothing to disagree with; the rest rests on that trust.
- No developer of these games was read saying the spacing was intended.
- **A Short Hike became a third exception on 2026-10-03,** when the user moved its loops on the map: roaming went
  up to the horizontal axis (0.04 toward Kinetic) and the feathers down near it (0.16 toward Cerebral). Its gap was
  0.75 before the move.
- **The two exceptions matter.** Stacklands, one of the best sellers on the list (30,383 reviews), has no swing in pace: every loop is
  thinking. SUMMERHOUSE has none either. So a swing between a fast and a slow pillar is common in this sample, not
  required. Stacklands may swing in another way (between calm stacking and the deadline of the moon), which this axis
  doesn't show.

## The user's leaning from this (2026-10-02, not a decision)

> "This tells me that I need to lean into the excitement part a bit, so that the player can take a break from the
> simulation / discovery part."

Claude's notes on it:

- **It matches the brief.** The user had already fixed "casual during play, with moments to stop and think"
  ([brief.md](brief.md)). The pool came out almost all thinking because the brainstorm collected ways to act on the
  simulation; the casual, hands-on pillar was never collected.
- **What the sample shows is a hands pillar, which is not the same as Excitement.** On Quantic Foundry's chart
  Excitement means fast-paced action and thrills. The peaceful games on the list get their swing from something
  milder: gathering (Forager), which sits 0.40 toward Kinetic, with no violence and no time pressure. So the leaning can be followed without leaving "peaceful" or the
  user's doubt about high-tension play.
- **Which pillar is the rest depends on the game.** In Brotato the shop is the rest from dodging. In a game whose
  demanding part is understanding a simulation, the hands pillar would be the rest.
- **Grade.** It follows from a finding graded "likely", and Stacklands shows a successful game without such a pillar.

## Confident

1. **Every game on the list has one dominant activity that can be named in a verb** (roam, place, stack, dodge, catch,
   gather). This holds however finely the games are cut into loops.
2. **Far apart on the map means independent, not opposed.** Quantic Foundry's own words: motivations that are farther
   apart "don't suppress each other". A game can serve two far-apart motivations.
3. **Games do appear under more than one motivation.** Animal Crossing and Pokémon are on both the Design list and the
   Completion list in Quantic Foundry's data, although those two are 2.36 squares apart on the map.
4. **Short to the first sale, long afterwards.** Brotato went on sale after 3 months and 20 Minutes Till Dawn after 2;
   both were then developed for about a year more in early access. Luck be a Landlord spent two years there.
5. **Small games have been made in the 4 to 8 month window and sold far above the "hit" line.** A Short Hike (4
   months, 22,778 reviews), Minami Lane (about 6, 6,786), Islanders (7, 16,787).
6. **One game on the list has a published cost.** Kabuto Park: 9 months, about €65,000, break-even near 30,000 copies,
   reached within the first month.
7. **A toy with no goals has sold.** SUMMERHOUSE: no score, no win, 3,382 reviews. One case.
8. **Two pairs on the list are repeat developers.** Minami Lane and Kabuto Park share a programmer; Islanders and
   SUMMERHOUSE share a designer.
9. **The list says nothing about modest success.** Every game on it is far above 1,000 reviews; the lowest has 3,382.
   The 200 to 300 review tier, which the user's Steam plan is judged by, is not represented.
10. **The model has no motivation for nurturing, cuteness or relaxing.** Shapes built on those can't be judged by the
    map.

## Likely

1. **Small successful games reach across the map.** All nine have loops on both sides of at least one axis (SUMMERHOUSE only through a side loop).
   Supported independently by point 3 above. How far they reach is not reliable: cutting a game into loops and
   placing each near the motivation it resembles spreads any game.
2. **Two halves can be one small game when one half is short and exists to feed the other.** Brotato and 20 Minutes
   Till Dawn pair a kinetic half with a thinking half and were built in two to three months. This bears on the user's
   worry that prep-and-run "might be TWO games".
3. **The peaceful games on the list sit where this game's pool sits.** A Short Hike, Minami Lane, SUMMERHOUSE and
   Kabuto Park are left of the vertical axis; Stacklands is almost entirely inside the pool's area. Both sets of
   positions are Claude's.
4. **This game's pool is almost all on the thinking side.** Of 36 marks, six are below the horizontal axis and none
   is further than 0.32 toward Kinetic. Read against the main finding: as it stands, the pool offers many slow pillars
   and hardly a fast one. The shapes that would supply one are the runs, the front and the character as centre.
5. **The toy is Discovery, and its two near neighbourhoods are far from each other** (Design on one side; Strategy and
   Completion on the other). The distances are measured on the picture; that the toy is Discovery is the user's own
   "surprise comes from discovery".
6. **Peaceful costs little on the map.** The only motivation the user's limit rules out is Destruction.

## Assumptions

1. **That a loop can be placed on a map made from players' motivations at all.** Everything on the map page rests on
   this. The user's check of familiar games is the only test so far.
2. **"One core, two supporting, one side."** The count comes from how finely Claude cut each game, which was a free
   choice. Only "one dominant activity" survives a different cut.
3. **The pool's two anchors, Power as an add-on and difficulty as a dial** ([03-consolidate.md](03-consolidate.md)).
   Claude's grouping; the user has not confirmed it.
4. **A wide mix widens the audience when its parts are optional to each other and narrows it when both are forced.**
   Inferred from how Quantic Foundry built its game lists; not stated by the source.
5. **That any of these patterns has to do with success.** All nine games are successes picked by Claude and the user.
   No failed game was placed, so nothing here separates what successful games do from what all games do.
6. **That the decline of the Strategy motivation touches light planning games.** The decline is sourced (strategy/09);
   the games behind the Strategy motivation are heavy ones, and whether a casual game with a little planning is
   affected is unknown.
7. **Exact distances and left–right positions.** Differences under about 0.1 on the axes, or 0.3 squares on the map,
   mean nothing. The World–Player position of a loop is the softest judgement of all.
8. **Development times that have no source:** SUMMERHOUSE, and Forager's "about two years". Stacklands is under 4
   months by the user's word (2026-10-02), with no written source on file.
9. **Time spent per loop.** Not estimated, on purpose. The user: it "depends on playstyle too much, and less so on the
   game itself".

## Not known yet

- How competitive each market is, its success rate, and how many games land in the modest tier (stage 4b).
- Loop times and complexity (stage 5).
- Which of the factors in [audience-factors.md](audience-factors.md) the audience fixes, since no audience is chosen.