Formal Specification
Mossi Traditional Archery and Target Game — Formal Specification
Burkina Faso · dexterity-throw · Confidence: low
← Back to game pageImplementation Suitability
Suitability: adaptation-requiredFidelity: probabilistic-proxyGap-fills: 3
# Reemdo — AI-optimized specification
> **Reconstructed source.** The entry carries a provenance warning: its rules were written from
> general knowledge, not read from the two works it cites. Tiendrébéogo (1963) and Skinner (1989)
> are real books, but the entry does not evidence that either describes this competition.
## General idea
A Mossi target-archery contest with a **progressive-distance structure**: competitors shoot a
fixed budget of arrows, then move back and shoot again, so the same nominal action gets steadily
harder across the match. The scoring is two-tier — a direct hit on the suspended calabash scores
2, a graze scores 1 — which means the target has an effective "near miss" band rather than a
binary outcome, and that band grows in importance as distance increases. There are no decisions
in the game: every competitor shoots the same number of arrows from the same line. It is a pure
skill-measurement contest, and formalizing it produces a simulation rather than a game.
## Formal model
- **Abstraction taken:** Archery accuracy is collapsed into a per-arrow outcome distribution over
{hit, graze, miss}, conditioned on distance. The entry specifies the round structure, arrow
budget, and point values exactly, and supplies no accuracy model and no relationship between
distance and success rate.
- **Player/agent count:** 2-8 competitors (entry §Players).
- **Turn model:** Sequential shooting within each round; rounds advance in lockstep for all
competitors.
## Formal rules
R1. Equipment (entry §Materials): small wooden bows, reed arrows or blunt shafts, and a suspended
dry calabash target — or a straw bundle (*Baa-kando*).
R2. Arrow budget (entry §1): each competitor shoots **3 arrows per round** from a designated firing
line.
R3. Scoring (entry §2): a **direct hit** on the calabash scores **2 points**; a **grazing shot**
scores **1**. A miss scores nothing [gap-fill: the entry does not define what counts as a graze
— a strike on the target's edge, a shot passing within some distance, or a hit that fails to
penetrate. The band's width is the single most important free parameter in the model and the
entry fixes none of it].
R4. Distance progression (entry §3): competitors **advance to farther distance markers after each
round**. The Objective names **15 m, 25 m, and 35 m**.
R5. Round count (entry §4): the competitor with the **highest cumulative score after 4 rounds**
wins.
R6. Round/distance mismatch: R5 specifies **four** rounds while R4 supplies only **three**
distances [gap-fill: the entry does not say what distance the fourth round is shot from. The
minimal reading is that the fourth round repeats 35 m, since the alternative — extrapolating a
45 m marker — would be inventing a distance the source does not give].
R7. Tiebreak: **unstated** [gap-fill: R5 compares cumulative scores and provides no procedure for
equal totals. With a maximum of 24 points over 12 arrows and coarse 2/1/0 scoring, ties are
likely rather than exceptional].
## Edge cases
- Tie at the end: **unaddressed by the entry** — see R7. This is the most probable unhandled
outcome in the game: 12 arrows scoring 2, 1, or 0 each produce a small integer range across which
8 competitors will frequently collide. No shoot-off, no countback on hits-versus-grazes, and no
shared victory is invented.
- A player with no legal move: not applicable — every competitor has exactly one action, `shoot`,
available for each of their 3 arrows in every round. Arrows are supplied per round rather than
from a depleting personal stock.
- Repetition / cycles: none possible. R5's round counter advances unconditionally and R2's arrow
budget is fixed, so the match is a straight-line sequence of exactly `competitors × 4 × 3` shots
with no branching and no loops.
- Simultaneous claims on one position: not applicable — shooting is sequential (R2) and the target
is not consumed or occupied by a hit. Two arrows striking the same spot is possible and harmless,
since each is scored independently.
- Elimination mid-round and turn-order repair: not applicable — no competitor is eliminated. All
competitors shoot all four rounds regardless of score, and the rotation is fixed throughout.
- Resource exhaustion: not applicable in any binding sense. Arrows are the only consumable and are
replenished to 3 each round (R2); the calabash target is struck repeatedly without being
destroyed, which the entry implies by scoring later shots normally.
- Termination guarantee: **guaranteed deterministically.** R2 and R5 together fix the total shot
count at exactly `competitors × 12`, independent of every outcome. No probabilistic argument is
required and no clock is needed — the round counter alone carries the guarantee. As with several
entries in this batch, termination and resolution are distinct: the match always ends, but under
R7 it may end without identifying a single winner.
## State representation
- Persistent (match) state: `score[p]` (monotone non-decreasing under R3), `current_round` ∈
{1..4}, `current_distance` (per R4 and R6), `current_competitor`, `arrows_left_this_round`.
- Per-shot ephemeral state: `outcome` ∈ {hit, graze, miss}, drawn against the distance-conditioned
distribution.
- Observable information: fully observable — every shot and its result are visible to all, and
scores are public.
## Action space
- `shoot(aim)`: the active competitor; precondition: `arrows_left_this_round > 0`. This is the only
action in the game. Since R2 fixes the budget and R4 fixes the distance, the competitor chooses
nothing about *when* or *from where* — only execution varies, which is why the game has no
decision structure.
## Playability notes
- Single-player-vs-AI: poor as an adversarial problem and honest as a solo challenge. Competitors
never interact, never block, and never respond to one another, so a match is a set of independent
scoring sequences compared at the end; an AI opponent contributes nothing that a score
distribution would not.
- Reducing n > 4 to 2 agents: lossless mechanically, precisely because there is no interaction —
removing competitors removes only the size of the field.
- Physical-skill games: bow draw, aim, and release are the entire contest and collapse into one
outcome draw per arrow (R3). Wind, which materially affects a reed arrow at 35 m and would be the
main source of round-to-round variance in the real event, is unmodelled.
- What is lost: the occasion and the lineage the entry names — harvest festivals, warrior training
assemblies, and the cavalry-and-archery tradition of the Mogho Naba. Also lost is the rural
variant the entry records, in which competitors shoot at a **rolling melon or woven straw ring**;
that is a moving-target game and genuinely different from the suspended-calabash version, and it
is not modelled here.
## Implementation hints
Make the distance-to-accuracy relationship an explicit, tunable curve and label it as an addition —
the entry supplies none, and this curve alone determines whether the progression in R4 produces a
meaningful difficulty ramp or a flat one. R3's graze band is the second required parameter: model
the target as two concentric radii, an inner hit radius and an outer graze radius, so a single
shot-dispersion model yields both tiers naturally. Resolve R6 visibly rather than silently — the
fourth round's distance is genuinely unspecified and repeating 35 m is an assumption, not a fact.
The representational trap is treating this as a game with strategy: there is none, so effort spent
on opponent modelling is wasted, and the useful work is entirely in the shot model. Suggested first
milestone: the four-round loop with a two-radius dispersion model, verifying that exactly
`competitors × 12` arrows are shot per match and that mean scores decline as distance rises.
Download as JSON
Download the full game corpus entry for this game as structured JSON.
Download JSONCite this page
Mossi Traditional Archery and Target Game (Burkina Faso). (n.d.). Games of Countries. https://gamesofcountries.com/games/bf-burkina-faso/reemdo-moore
@misc{gamesofcountries-bf-burkina-faso-reemdo-moore,
title = {Mossi Traditional Archery and Target Game (Burkina Faso)},
year = {n.d.},
publisher = {Games of Countries},
url = {https://gamesofcountries.com/games/bf-burkina-faso/reemdo-moore}
}