About SettleRisk

We build the infrastructure that makes prediction markets trustworthy.

Our Mission

Prediction markets are one of the most powerful mechanisms for aggregating human knowledge into actionable forecasts. They have the potential to transform how societies make decisions — from policy design to corporate strategy to scientific research. But there is a fundamental problem that threatens this promise: resolution risk.

When a market's resolution rules are ambiguous, when a single oracle can fail, when "end of day" doesn't specify a timezone — traders face hidden risks that no amount of fundamental analysis can uncover. These risks lead to disputes, frozen capital, delayed settlements, and eroded trust in the entire ecosystem.

SettleRisk exists to quantify and price these risks. We provide the analytical infrastructure that traders, market makers, and platforms need to understand resolution risk before it becomes a crisis. Our deterministic scoring engine, explainable driver attribution, and dispute-adjusted pricing tools bring transparency to a problem that has historically been opaque and subjective.

Our mission is simple: make prediction markets safer and more efficient by making resolution risk measurable.

The Problem We Solve

Prediction markets are growing rapidly. Polymarket processed over $9 billion in volume in 2024. Kalshi is expanding into new categories every quarter. But as the market grows, so do the stakes — and the disputes.

Consider what happens when a market asks whether Bitcoin will reach "approximately $100K" and the price hovers at $99,800. Or when a market on a CEO resignation says "by end of day Friday" without specifying a timezone. Or when the sole data source for a government statistic goes down during the resolution window.

These aren't hypotheticals — they are real disputes that have frozen millions of dollars in capital for weeks. Traders who didn't see these risks coming paid the price.

$12.4M+
Capital frozen in disputes across major platforms in 2024-2025
21 days
Average dispute resolution time for HIGH-tier markets
73%
Of disputes stem from ambiguous wording or missing definitions

Founder

PM

Phil Maher

Founder & CEO

Phil Maher founded SettleRisk after experiencing the resolution risk problem firsthand as an active participant in prediction markets. After watching millions of dollars get locked in preventable disputes — disputes caused not by bad actors, but by ambiguous contract language, single points of failure in data sources, and timezone inconsistencies — he realized that the prediction market ecosystem was missing a critical piece of infrastructure.

Phil brings a rare combination of deep technical expertise and market intuition to SettleRisk. His background spans systems engineering, quantitative analysis, and fintech product development. He has spent years studying how markets resolve, why they fail, and what patterns are detectable before disputes occur. This domain expertise is encoded directly into SettleRisk's 15-driver taxonomy and deterministic scoring methodology.

Before SettleRisk, Phil built high-throughput data systems, worked on risk modeling, and contributed to open-source infrastructure projects. He is a strong believer in the power of prediction markets to improve decision-making at every level of society — and equally passionate about solving the trust and transparency challenges that hold the ecosystem back.

Phil's vision for SettleRisk goes beyond a risk scoring API. He sees resolution risk analytics as a foundational layer for the prediction market stack — as essential as price feeds are to traditional finance. Under his leadership, SettleRisk is building the tools that will enable the next generation of prediction market infrastructure to scale with confidence.

What Sets Us Apart

Deterministic and Reproducible

Given the same inputs and version stamps, our scoring engine produces identical outputs every time. No black boxes, no random variation. Every score includes the exact version stamps (heuristics, extractor, taxonomy) that produced it.

Explainable by Design

Every risk score comes with up to 15 drivers, each pointing to the exact text span in the resolution rules that triggered it. You don't just get a number — you get a complete audit trail of why that number exists.

Pricing, Not Just Scoring

We don't stop at risk scores. Our pricing engine computes dispute-adjusted fair prices, risk premiums, capital lockup costs in basis points, and fair spreads — the actual numbers market makers need to set their quotes.

Built for Trading Infrastructure

Rust-powered backend, sub-5ms staleness windows on the Fund tier, gRPC streaming for real-time alerts, batch APIs processing 1,000 markets per call, and HMAC-SHA256 authentication with replay protection. This is infrastructure built for production trading.

Our Approach

SettleRisk's methodology is built on three pillars:

1

Structured Rule Extraction

We use LLMs in a controlled, validated pipeline to extract structured data from natural-language resolution rules — identifying ambiguities, dependencies, temporal constraints, and edge cases that humans might miss.

2

Deterministic Scoring

Extracted features flow into a closed-form scoring engine with published formulas. Platform base points, driver contributions, mitigations, and complexity adjustments combine into a reproducible aggregate score.

3

Financial Modeling

Scores feed into our pricing engine, which translates abstract risk into concrete financial metrics: dispute probability, expected settlement delays (lognormal distributions), capital lockup costs, and fair spreads. These are the numbers that actually inform trading decisions.

Our Values

Transparency First

Our scoring methodology is published openly. Our formulas are documented. Our driver taxonomy is public. We believe that trust infrastructure must itself be trustworthy — and that starts with transparency.

Rigor Over Speed

We don't ship features that compromise determinism or reproducibility. Every score must be explainable, every version stamp must be tracked, every output must be verifiable.

Infrastructure Mindset

We build tools that other businesses rely on for critical decisions. That means extreme reliability, clear contracts, additive-only versioning, and never breaking backwards compatibility without a migration path.

Market Integrity

We are deeply committed to the health of the prediction market ecosystem. Better resolution risk tools mean fewer disputes, less frozen capital, more efficient markets, and greater public trust in prediction markets as a whole.

Join Us

We're building the risk infrastructure layer for the prediction market ecosystem. Whether you want to use our API, partner with us, or join our team — we'd love to hear from you.