Sentiment Pricing

Each answer pool is a V2-style pair (KUSD × AnswerToken) governed by the invariant x · y = k. The marginal price of the answer token is p = x / y (KUSD per token).

When you buy Δx KUSD worth of answer tokens, creator/protocol/platform buy fees are 0%, but the V2-style AMM retains a 0.3% pool fee. Output is: Δy = y − (x · y) / (x + Δx · 0.997). The price moves nonlinearly as larger buys displace the reserves.

Buy example:
  x (KUSD reserve)      = 1000
  y (token reserve)     = 9000
  p = 1000 / 9000       = 0.111 KUSD/token

  Buy Δx = 100 KUSD (10% of reserve)
  Δy ≈ 818 tokens out
  New x = 1100, new y ≈ 8182
  New p ≈ 0.134 KUSD/token  (+20.7% price impact)

The displayed percentage for answer i is: s_i = p_i / (p_1 + p_2 + ... + p_n) × 100%. All pools start with the same virtual KUSD seed and the same token seed, so all answers begin at equal weight (e.g., 50/50 for a two-answer market).

As participants buy one answer more than others, its reserves shift, its price rises, and its share grows. This is pure demand-weighted sentiment measurement — no external input is required or trusted.

Three contract-level controls bound single-actor displacement:

• maxBuyPriceImpactBps = 2500 (25%) — one buy cannot move the price of one answer pool by more than 25%.
• maxBuyKusdBpsOfVirtualSeed = 5000 (50%) — one buy cannot inject more than 50% of the initial virtual seed into a single answer pool.
• tradeCooldownSeconds — optional per-market cooldown between successive trades from the same address.

These controls do not eliminate manipulation — they bound it and make it measurable. Because impact limits are explicit protocol parameters, the maximum single-trade price displacement is governance-defined, not dependent on liquidity depth alone.

A sentiment share above 60% on one answer tells you: more KUSD has flowed into this answer relative to alternatives. It does not tell you that this answer is correct, likely, or even that the crowd understands the question accurately.

Useful reads: watch for rapid price shifts near major events. A sudden 10-point swing in a two-answer market may reflect new information entering the crowd faster than any oracle could capture. That real-time responsiveness is the unique value of on-chain sentiment markets.