Skip to main content
Key terms used throughout the Belief Index documentation, explained for a finance-literate audience. Terms are grouped by category for easy reference.

​
Index Terms

A family of rules-based indices, published by Belief Systems, that measure how prediction markets price themed categories of real-world event risk. Each index is organized as a “series” tracking a specific theme, valued at periodic valuation windows from observable market data and published as a chain-linked index level.
A named, themed basket of prediction markets with assigned weights – the fundamental unit of a Belief Index. Each series tracks a specific theme such as monetary policy expectations, election outcomes, or economic indicators. Examples: “Belief U.S. Monetary Policy Dovish Stance Expectations Index,” “Belief U.S. Presidential Election Republican Expectations 2028 Index.” Analogous to a single named index within a provider’s index family.
The lifecycle classification of a series, set at publication and immutable thereafter. Every series is either Fixed (composition locked at publication; lifecycle bounded by market resolutions) or Perpetual (composition may evolve through formal reconstitution events; lifecycle open-ended). Displayed on each series detail page. See Perpetual Series.
A series whose market composition is locked at publication and never changes. Its lifecycle is bounded – the series runs until all constituent markets resolve, at which point it reaches its terminal index value. The default classification for event-bounded themes (e.g., a specific election or calendar year of macro data).
A series whose market composition may evolve over time through formal reconstitution events. Its lifecycle is open-ended – the series runs until manually archived. Suitable for thematic exposure to a domain that evolves (e.g., geopolitical conflict risk). Reconstitution adds new markets – and, under the published maintenance criteria, may remove them on a rules-based (never discretionary) basis – at a published review cadence. See Perpetual Series and Index Governance.
The specific set of prediction markets, tracked outcomes, and weights that define a series. The composition specifies exactly which markets are included, which side (YES or NO) is tracked, and how much weight each market carries. Compositions are versioned – each change creates a new version for auditability.
The valuation of a series’ underlying basket, computed at each valuation window: a probability-weighted aggregate of constituent prices. Published in two forms – the raw index value and the rebased Index Level – both defined below. It is a theoretical measurement of a basket, not the net asset value of any fund.
The probability-weighted aggregate of all market prices in a series, bounded between 0 and 1. Computed as the weighted average of each market’s midprice (or settlement price, if resolved). A raw index value of 0.65 means the weighted-average implied probability across all markets in the basket is 65%. See Index Calculation Methodology.
A rebased representation of the raw index value, starting at 100 at inception. Computed as: 100 x (raw_nav / inception_raw_nav). Makes it easy to track percentage performance over time – an Index Level of 95 means a 5% decline from inception; 110 means a 10% gain. Analogous to the level of a price index like the S&P 500 or FTSE 100.
The first successful computation for a series. The raw index value at inception becomes the denominator for the Index Level calculation, setting the base at 100. All subsequent performance is measured relative to inception.
A periodic checkpoint at which every series is revalued and its index level published – currently every 30 minutes (subject to change). See How It Works.
A computation that relies on outdated price data because one or more underlying market price fetches failed. When a fetch fails, the system uses the last known good price as a fallback and flags the computation as stale. Stale computations are always transparently marked so data users know the data quality.
The final, definitive index value of a series after all underlying markets have resolved. It is not subject to further change and is never marked as stale, since all component prices are settlement prices ($1 or $0). No further computations occur after a series reaches its terminal value.
A formal event, initiated by Belief Systems, that modifies a Perpetual series’ composition by adding one or more new constituent markets. Analogous to an index provider adding a company to the S&P 500. Reconstitution is methodology-driven, forward-looking, and explicitly authorized. Markets are added under the published inclusion criteria and removed only on a rules-based (never discretionary) basis under the maintenance criteria. See Perpetual Series and Index Governance.
A planned, bounded pause on publication for a series while a composition change is written and verified. The most recent pre-gate index value remains authoritative during the pause; publication resumes once the change is committed and verified. The gate typically lifts within roughly 48 hours of activation. Other series are unaffected.
The mechanism that preserves index-level continuity across a reconstitution. The index level immediately before the composition change is recorded as the chain-link anchor, and a new inception_raw_nav is computed so the first post-reconstitution index level equals the anchor. The reported index history is continuous – the chart shows no jump. See Index Calculation Methodology – Chain-Linking.

​
Market Terms

A market where participants trade contracts that pay $1 if a specified event occurs and $0 otherwise. The trading price reflects the crowd’s implied probability of the event. For example, a contract trading at $0.70 implies the market believes there is a 70% chance the event will occur. Belief Index sources its data from a single prediction market venue.
A digital token representing one side of a binary prediction market. A YES token pays $1 if the event occurs; a NO token pays $1 if it does not. These are ERC-1155 tokens on the Polygon blockchain, and they are the instruments a Belief Index series measures.
The specific side (YES or NO) of a prediction market that a series composition follows. For example, a series might track the YES outcome for “Will the Fed cut rates?” – meaning the series measures the YES side of that market. The tracked outcome determines which order book is used for pricing.
The arithmetic mean of the best bid and best ask prices: (best_bid + best_ask) / 2. Used as the fair value estimate for each market in the index computation. The midprice is a theoretical value – it represents what a trade might execute at, but does not guarantee execution at that price. See Index Calculation Methodology.
A Central Limit Order Book – a list of buy orders (bids) and sell orders (asks) for a given market, sorted by price. The best bid is the highest price a buyer will pay; the best ask is the lowest price a seller will accept. Belief Index reads the top of book (best bid and best ask) from the venue’s CLOB to compute midprices.
The settlement of a prediction market when its underlying event occurs (or definitively does not occur). At resolution, the winning outcome token pays $1 and the losing token pays $0. Resolution is final and binary – there are no partial outcomes.
The final price of an outcome token after market resolution: $1.00 for the winning outcome, $0.00 for the losing outcome. Settlement prices replace midprices in the index computation once a market resolves.
The difference between the best ask and the best bid in an order book. A narrow spread (e.g., $0.01) indicates a liquid market with strong price discovery. A wide spread (e.g., $0.05) indicates a less liquid market where the midprice is a less reliable estimate of fair value.
The difference between the expected execution price and the actual execution price when trading. Slippage increases with order size relative to market depth. Index levels are theoretical midprice measurements and do not account for it – see Risk Factors.

​
Blockchain Terms

The public blockchain network where the underlying prediction markets settle. Polygon is operated by Polygon Labs and hosts the outcome-token and settlement contracts a Belief Index series reads from.
A token standard on Ethereum-compatible blockchains that supports multiple token types within a single smart contract. Prediction market outcome tokens use the ERC-1155 standard, allowing both YES and NO tokens for a given market to exist within one contract.
The public smart contract on Polygon through which prediction markets settle. When a condition resolves, the oracle reports its payout to that contract. Belief Systems reads payoutDenominator and payoutNumerators from it to confirm a constituent has genuinely resolved, rather than trusting a venue API. See On-Chain Verification.
Recorded on a public blockchain. On-chain data is transparent (anyone can read it), immutable (cannot be altered after recording), and independently verifiable. When we say a resolution is confirmed “on-chain,” it means anyone can confirm the same fact by querying the Polygon blockchain directly.
A program that runs on a blockchain and executes automatically when predefined conditions are met. Outcome tokens and the settlement contracts behind them are implemented as smart contracts. Smart contracts are transparent (their code is publicly readable) but may contain undiscovered bugs.