Heavymath: Rethinking Prediction Markets with Percentage-Based Odds

--

Abstract

Prediction markets have long promised to harness collective intelligence for forecasting real-world events. Yet most existing platforms force users into rigid binary choices — yes or no, up or down — discarding the nuance that makes predictions valuable. Heavymath introduces a fundamentally different approach: percentage-based odds that let participants express exactly how confident they are, from 0% to 100%. Combined with a novel dual-boundary market split algorithm, fully on-chain settlement, and a dealer-driven market creation system, Heavymath creates prediction markets that are more expressive, fairer, and more transparent than anything that exists today.

The Problem with Binary Prediction Markets

Traditional prediction markets present a false dichotomy. When you bet on whether Bitcoin will hit $100,000, you are forced to choose “yes” or “no.” But what if you believe there is a 70% chance? On most platforms, you place the same bet as someone who believes there is a 99% chance. The market cannot distinguish between moderate confidence and near-certainty.

This binary constraint creates several problems:

• Loss of information — The market aggregates less signal from participants because nuance is discarded at the point of entry.
• Unfair odds — Participants with different levels of conviction end up in the same pool, subsidizing those who are less certain.
• Limited expressiveness — Markets cannot capture the full spectrum of belief, reducing their value as forecasting tools.
• Opaque settlement — Many platforms settle off-chain, requiring trust in a centralized operator.

Heavymath was built to solve these problems from first principles.

A New Primitive: Percentage-Based Predictions

On Heavymath, every prediction is a number between 0 and 100. This percentage represents the participant’s belief in the probability of a positive outcome. A prediction of 80% means “I believe there is an 80% chance this happens.” A prediction of 20% means the opposite.

This is not just a UI change — it is a structural shift in how the market operates. Each participant’s percentage and stake amount together define their position. The market aggregates all positions to determine which predictions end up on which side of the outcome, and how much each side has at stake.

The result is a prediction market that captures far more information from its participants. Instead of a single yes/no price, Heavymath markets produce a rich distribution of beliefs across the probability spectrum.

The Dual-Boundary Market Split

The core innovation in Heavymath is the market split algorithm — a mechanism that fairly divides participants into two sides based on their percentage predictions and stake amounts.

When a market’s deadline passes, the algorithm scans all occupied percentage points and identifies two boundaries: a negative boundary and a positive boundary. Predictions at or below the negative boundary form the negative side (betting against the outcome). Predictions at or above the positive boundary form the positive side (betting for the outcome). Predictions between the two boundaries fall in a “middle zone” and receive a full refund — the market determined that their position was too ambiguous to place on either side.

The boundaries are not arbitrary. They are calculated using a bidirectional capital-adequacy constraint: each side must have enough capital to make the other side’s odds worthwhile. If one side is overweight, boundary stakes are trimmed proportionally so that neither side is unfairly disadvantaged.

This mechanism has several desirable properties:

• Fairness — No participant is forced onto a side where their odds are worse than what they expressed.
• Capital efficiency — Only the capital needed to balance both sides is locked in. Excess is refunded.
• Expressiveness — The full range of beliefs (0–100%) is preserved, not collapsed into binary buckets.
• Determinism — The algorithm is pure math with no discretion, executed entirely on-chain.

How a Market Works: End to End

1. Market Creation. A dealer — someone who holds a Dealer License NFT — creates a market by specifying a description, category, deadline, and optional oracle configuration. Dealers earn a share of settlement fees, creating an incentive to curate high-quality markets.

2. Prediction Phase. Before the deadline, any user can place a prediction by choosing a percentage (0–100) and staking USDC. Predictions can be updated or withdrawn at any time before the deadline. There are no order books, no counterparties to match, and no liquidity pools to seed. Users simply state their belief and stake.

3. Market Lock. After the deadline passes, the market is locked. The split algorithm runs, determining the two boundaries and each participant’s side. Middle-zone predictions are refunded. Boundary predictions may be partially trimmed. The remaining capital forms the distributable pool.

4. Resolution. The market is resolved with a binary outcome: positive or negative. Resolution can happen manually by the dealer, automatically via an on-chain oracle, or through an asynchronous Chainlink Functions request. The resolution method is configured at market creation and cannot be changed.

5. Settlement. Winners — those whose side matches the outcome — claim their proportional share of the distributable pool, minus a small settlement fee (default: 1%). The fee is split between the dealer and the platform. All settlement math happens on-chain, verifiable by anyone.

The Dealer System: Decentralized Market Curation

Anyone can become a dealer by minting a Dealer License NFT. The license grants permission to create markets within specific categories. This creates a curated ecosystem where market quality is maintained by the people who have the most domain expertise — sports analysts create sports markets, finance professionals create finance markets, and so on.

Dealers earn a share of the settlement fee from every market they create, aligning their incentives with market quality and participation. A dealer who creates popular, well-defined markets earns more. A dealer who creates vague or unpopular markets earns nothing.

The permission system uses a category and subcategory hierarchy with wildcard support, allowing fine-grained control over which dealers can create which types of markets.

Built-in Communication: Email Notifications via Signic

One of the most overlooked problems in decentralized applications is communication. When a market locks, resolves, or a user’s prediction is refunded, how do they find out? Most platforms rely on users checking back manually — a terrible experience that leads to missed outcomes and unclaimed funds.

Heavymath solves this through integration with Signic, a decentralized email platform. Every wallet address automatically corresponds to a free email account. When a market event occurs — a lock, a resolution, a refund — participants receive an HTML email notification with all the relevant details: which side they are on, what their effective stake is, whether they won or lost, and how to claim their funds.

This is not a bolt-on notification service. It is built into the indexing infrastructure, triggered directly by on-chain events, and delivered through a decentralized email stack. No email signup, no notification preferences, no push tokens. If you have a wallet, you have an inbox.

Sports: The First Vertical

While Heavymath supports markets on any topic — crypto, politics, economics, current events — sports is the first deeply integrated vertical. The platform provides live data for ten major sports: soccer, basketball, American football, baseball, hockey, rugby, Formula 1, MMA, handball, and volleyball.

Dealers can create markets directly from upcoming fixtures, with oracle-based resolution that automatically settles markets based on game results. Users can browse leagues, teams, and upcoming games, then place predictions with full context — standings, head-to-head records, and historical data — all within the same interface.

This vertical integration turns Heavymath from a generic prediction market into a purpose-built sports forecasting platform, while the underlying infrastructure remains flexible enough to support any market category.

Radical Transparency

Every aspect of Heavymath’s operation is verifiable on-chain:

• Market creation parameters, deadlines, and oracle configuration are public from the moment a market is created.
• Every prediction — percentage, amount, and timestamp — is recorded on-chain.
• The market split algorithm is deterministic and can be independently verified by reading the smart contract.
• Settlement fees, dealer shares, and payout calculations use fixed formulas with no discretionary components.

There is no hidden order book, no market maker taking the other side, and no admin key that can change the rules after bets are placed. The smart contracts are the rules, and they apply equally to everyone.

Fee Model

Heavymath uses a simple, transparent fee model. A settlement fee (default: 1%) is charged on the distributable pool — the total capital remaining after middle-zone refunds and boundary trimming. This fee is split between the dealer who created the market and the platform.

There are no fees on placing predictions, updating predictions, or withdrawing before the deadline. Users only pay when they win. The fee is deducted from the payout, so there is never a scenario where a user pays more than they receive.

Looking Ahead

Heavymath is in active development with a clear roadmap:

• Additional oracle integrations — enabling automated resolution for a wider range of market types.
• Mobile applications — native apps with Wallet Connect integration.
• Governance — transitioning platform parameters (fee rates, dealer permissions) to community governance.
• More verticals — expanding beyond sports into structured markets for elections, economic indicators, and cultural events.

The percentage-based prediction primitive is general enough to support any market where participants have a probabilistic view of the outcome. As the platform grows, the same core algorithm and settlement logic scales to every new category and chain.

Try Heavymath

Heavymath is live on testnet. You can explore markets, place predictions, and experience the percentage-based odds system today — no real funds required. Visit the testnet application at dev.heavymath.com to try it yourself.

Conclusion

Prediction markets work best when they capture the full range of human belief. Binary markets throw away information. Heavymath preserves it.

By combining percentage-based odds with a fair, deterministic market split algorithm, decentralized settlement, a dealer incentive system, built-in email notifications, and deep sports integration, Heavymath creates prediction markets that are more expressive, more transparent, and more useful than what exists today.

The future of forecasting is not yes or no. It is a number between 0 and 100.