Core Concepts

Dual-Market Genesis

Every FS market begins as two linked surfaces:

1. Probability Market

Optimised for forecasting, belief expression, liquidity dynamics, and trader behaviour.

2. Reality Market

Optimised for finality, correctness, adversarial resistance, and rapid convergence when information becomes public.

They share:

• The same bounds

• The same resolution precision

• The same market ID

The Probability Market expresses opinions. The Reality Market determines truth. The two are inseparable parts of a single market lifecycle.

Continuous Probability Forecasting

The Probability Market is neither an Bonding Curve nor an Order Book. It uses:

• A Dirichlet state vector (A(α))

• Bernstein basis coefficients

• A concave state potential A

• Deterministic, reversible sells

• Pure participant-funded liquidity

• No LPs, no buckets, no discrete shares

Unique Properties

• Expressive beliefs: Any shape, not just binary or multiple choice.

• Continuous domain: Markets are real-valued.

• Exit positions: Position values are reappraised with every change to the market consensus and can be sold at any time.

• Neutrality: Consensus emerges from aggregated belief curves.

UI Freedom

FS does not prescribe a user interface. Developers can:

• Translate simple toggles into narrow distributions

• Offer Gaussian presets

• Let users paint arbitrary curves

• Offer domain-specific UIs (sports, finance, elections)

• Hide everything behind simple sliders

FS requires only the coefficients. Everything else is implementation detail.

The Predictive Liquidity Pool

functionSPACE uses a model that behaves closest to a parimutuel in that all collateral sits in one shared pool, but unlike a parimutuel you can enter, exit and adjust positions at any time.

The defining characteristic is simple:

Liquidity is not split across outcomes.
All trades feed into one predictive pool.

Every participant on a Topic contributes to the same market surface instead of isolated buckets.
As more people trade, the pool grows and the market consensus is harder to move.
There is no separate LP class. Liquidity comes directly from participants expressing beliefs.

This avoids the thinness, fragmentation and subsidy requirements that show up in Order Book or LMSR based models.

How Traders Make Money in functionSPACE

Traders earn based on three dimensions that operate together:

1. Be accurate

Payout is based on how much density your position has near the resolved value compared to the market.
If your forecast tightly covers the outcome, your share of the pool rises.
If you were vague or wrong, it falls.

It is accuracy-weighted, not binary win-or-lose.

2. Be early (relative to pool size)

“Early” is not a timestamp. It is a ratio.

A trade has more influence when the pool is small.
As the pool grows, a trade of the same size has less impact.
This rewards predictions/forecasts that arrive early in relative size, not just early in time.

A newcomer who trades meaningfully against a now-large pool can still shift the market and still be “early.” This prevents markets from becoming stale or locked-in.

3. Add meaningful collateral

Large, high-conviction trades matter because they shift the curve.
Collateral is not just margin but the weight behind the belief.

Traders who stake more on accurate predictions earn proportionally larger claims on the pool.

Truly organic liquidity

Because all collateral goes into one unified pool:

• there is no LP/informed-trader adversarial relationship

• every trader contributes liquidity simply by participating

• the pool always has the capital needed for payouts

• the system stays solvent without external injections

This is “predictive liquidity” — liquidity that grows as more people express and refine their beliefs.

Entering and Exiting

Opening a position adds its shape to the market.
Closing a position removes it.

Redemption is simply the inverse of entry:

• you sell your position back to the protocol

• your influence on the curve is removed

• you receive value based on how aligned your position is with the current market

There are no capital traps because the rules for entering and exiting are symmetric.

Settlement: Closed-Budget

When the outcome resolves:

1. Payouts are proportional to:

   • the ticket’s influence when minted

   • the accuracy of its density near the outcome

2. The entire pool is redistributed to ticket holders.

3. No external payout capital is required.

functionSPACE probability markets are a closed economic game with transparent incentives

Reality Markets: Resolution Layer

The Reality Market is a capital-driven, adversarial mechanism designed for the final stage of prediction: determining what actually happened.

It is prediction-like (agents assert values), but its purpose is not speculation. Its sole purpose is resolution.

Properties

• Propositive assertions: Any x* allowed

• Continuously disputable: Open to challenge by placing a "no" Bounty top up

• Capital requirements: Token-denominated closure/continuance capital

• Finalisation: Capital-weighted median finalisation

• Slashing: MAD-band slashing to penalise inaccurate assertions

• Robustness: Clearance ratio + dwell period for robust finality

• Timing: Fixed-time expiry markets prevent premature resolution

The result is a single numerical truth x*

Composability & Flexibility

FS is the mechanism. Developers decide how to integrate it. Developers can run:

• Probability Markets only

• Reality Markets only

• Full PM + RM

• Custom UIs with hosted RM

• Hosted PM with custom RM

• Curated market directories

• White-label prediction layers