Alternative · AI agent
Predly vs
SimpleFunctions.
Both products identify gaps between market prices and probability estimates across Kalshi and Polymarket. Predly surfaces those gaps as AI-accuracy alerts. SimpleFunctions ships the agent layer above raw signals: a causal-tree thesis system that decomposes any claim into testable sub-claims, an autonomous trading agent with a 7-gate risk cascade, computed indicators across 48K+ contracts, and a 56-tool MCP server that drops into any AI development environment.
Verified 2026-04 · public sources only · live SF data from /calibration
Verdict
Pick the one that fits how
you actually work.
Choose SimpleFunctions if
You are building agents, autonomous trading systems, or research workflows that need more than mispricing alerts — calibrated probabilities with public Brier scores, causal-tree thesis modelling with auto-evaluation cycles, regime classification across the full 48K-contract universe, computed indicators (implied yield, cliff risk, liquidity availability score), an autonomous trading agent, and a 56-tool MCP server that drops into Claude Code or Cursor in one line.
Choose Predly if
You want a standalone consumer product that surfaces AI-detected mispricings as ready-to-act alerts without writing any code or integrating an API. Predly's stated focus is the gap between market prices and AI-calculated true probabilities, delivered as a curated alert feed — that specific surface is their product focus.
Same upstream venues (Kalshi + Polymarket). Predly surfaces mispricing alerts for consumers. SimpleFunctions ships the agent layer above it: world model, theses, indicators, autopilot, MCP.
At a glance
Three things that
actually differ.
Everything Predly gives you — AI-derived probability estimates compared against live market prices, mispricing detection across Kalshi and Polymarket — SimpleFunctions also gives you, plus cross-venue arbitrage pairs, orderbook depth, and trade ideas with conviction.
On top of that, SF ships a causal-tree thesis system, an autonomous trading agent (Portfolio Autopilot, 1M-context LLM, 7-gate risk cascade), computed indicators across 48K+ contracts, and 56 MCP tools that no current PM alert product exposes.
SF publishes live Brier scores at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days. Predly claims 89% alert accuracy; their calibration methodology and breakdown are not publicly documented.
Side by side
10 dimensions · verified 2026-04SimpleFunctionsKalshi + Polymarket normalised, 48K+ active contracts indexed at /api/public/cross-venue/pairs.
PredlyKalshi + Polymarket coverage used as the basis for AI probability vs market price comparison.
SimpleFunctionsCross-venue arbitrage pairs at /api/public/cross-venue/pairs?preset=arb and trade ideas at /api/public/ideas with conviction scores and catalysts.
PredlyAI-calculated true probability compared to live market price, surfaced as mispricing alerts; publicly claims 89% alert accuracy.
SimpleFunctionsImplied yield, cliff risk index, liquidity availability score, event overround, τ-days, and regime label pre-computed across 48K+ contracts and available at /screen.
PredlyNo publicly documented computed indicator layer beyond the AI probability vs price gap.
SimpleFunctionsLive Brier scores at /api/calibration — segmented by venue, category, and price bucket, updated continuously from production data.
Predly89% alert accuracy claimed publicly; underlying calibration segmentation and methodology are not published.
SimpleFunctionsGET /api/public/market/{ticker}?depth=true returns the bid/ask ladder, spread, and slippage estimates.
PredlyOrderbook access not documented publicly.
SimpleFunctionsPOST /api/thesis/create decomposes any sentence into a causal tree, propagates probabilities, scans for edges, and runs an evaluation heartbeat with news scan and LLM eval.
PredlyNot in scope; Predly surfaces alerts on individual contracts, not decomposable causal models.
SimpleFunctionsPortfolio Autopilot — 1M-context LLM, 13 data sources, 7-gate risk cascade including kill switch, drawdown gate, and regime check.
PredlyNot in scope; Predly is an alert product, not an execution agent.
SimpleFunctions56 tools via claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp — compatible with Claude Code, Cursor, and any MCP client.
PredlyNo MCP server published.
SimpleFunctionsFull public REST API — /api/public/markets, /api/public/scan, /api/agent/world, orderbook, cross-venue pairs, all readable without authentication.
PredlyNo public REST or programmatic API documented.
SimpleFunctionsPublic REST, MCP, and CLI reads require no authentication. Authenticated thesis/intent execution is free up to 15M tokens, then pay-per-token.
PredlyPricing not publicly documented in detail.
Methodology
Verified 2026-04 from public sources only — Predly's documentation, public website, and publicly observable behaviour. We never claim non-public information about Predly's internals. SimpleFunctions claims on this page are computed live from /api/calibration, /api/public/cross-venue/pairs, and /api/public/markets — you can re-verify them yourself with curl.
Use cases
Same data, different
best fit per scenario.
Scenario 01
Building an AI agent that needs to reason about prediction market signals as part of a larger decision pipeline.
SimpleFunctions · best fit
SF's 56-tool MCP server exposes markets, theses, indicators, and trade ideas directly to any MCP client. The /api/agent/world endpoint delivers an ~800-token world snapshot for LLM context injection. The causal thesis system lets the agent decompose its own hypotheses and track them autonomously.
Predly
Predly has no documented API or MCP integration. Feeding its alerts into an agent pipeline would require manual extraction or scraping.
Scenario 02
A retail trader who wants a simple, no-code alert feed that flags mispricings on Kalshi and Polymarket without building anything.
SimpleFunctions
SF is primarily a developer and agent platform. While trade ideas at /api/public/ideas surface edge opportunities, the product assumes API or CLI usage and does not deliver a consumer-facing alert UI.
Predly · best fit
Predly is purpose-built for exactly this scenario: its product surfaces AI-detected mispricing alerts with a claimed 89% accuracy, designed for a trader who wants actionable signals without writing code.
Scenario 03
Decomposing a complex macro thesis — such as 'US recession by Q4 2026' — into tradeable sub-claims across multiple prediction markets.
SimpleFunctions · best fit
POST /api/thesis/create takes the sentence, builds a causal tree of testable sub-claims, scans Kalshi and Polymarket for each, propagates probabilities, and runs a recurring evaluation heartbeat with news scan and LLM eval. No other PM product has this.
Predly
Predly compares AI probabilities to market prices for individual contracts but does not offer structured thesis decomposition or causal tree modelling.
Scenario 04
Auditing the accuracy of a data source before building a trading strategy on top of it.
SimpleFunctions · best fit
SF's /api/calibration endpoint returns live Brier scores by venue, category, and price bucket — Kalshi 0.20, Polymarket 0.12 on T-24h price over the past 90 days — verifiable with a single curl call before committing any capital.
Predly
Predly claims 89% alert accuracy but does not publish a calibration breakdown by category, price bucket, or time window that can be independently verified.
Live data
The SimpleFunctions claims on this page are not marketing copy. Brier scores, market counts, and cross-venue pair counts are computed live from /calibration, /screen, and /api/public/cross-venue/pairs. All public, all free, all CC-BY-4.0.
FAQ
How does SimpleFunctions' thesis system differ from Predly's mispricing alerts?+
Predly compares an AI-calculated probability to the current market price and surfaces the gap as an alert. SimpleFunctions' thesis system takes a written claim, decomposes it into a causal tree of testable sub-claims, scans Kalshi and Polymarket for contracts that proxy each node, propagates probabilities through the tree, and runs a recurring evaluation heartbeat — news scan, price refresh, milestone check, LLM evaluation, confidence update. You can also inject external signals via /api/thesis/{id}/signal. The result is a structured, auditable research object, not a single alert.
Does SimpleFunctions surface mispricing signals equivalent to Predly's alerts?+
Yes. SF exposes cross-venue arbitrage pairs at /api/public/cross-venue/pairs?preset=arb and trade ideas with conviction scores and catalysts at /api/public/ideas. These are computed programmatically from live orderbook and cross-venue price data across 48K+ contracts. They are accessible via the REST API, the CLI, or any of the 56 MCP tools — no consumer UI required.
What is Portfolio Autopilot and how does it work?+
Portfolio Autopilot is SF's autonomous trading agent. It uses a 1M-context LLM alongside 13 data sources and a 7-gate risk cascade — kill switch, position limits, drawdown gate, regime check, and others — before placing any execution. It is designed for users who want autonomous position management rather than a manual alert-to-trade workflow. The agent reads thesis state, indicator data, and calibration scores as part of its decision context.
How can I use SimpleFunctions with Claude Code or Cursor?+
Run claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp to register the 56-tool MCP server in Claude Code, Cursor, or any MCP-compatible client. Tools cover markets, theses, indicators, orderbook, cross-venue pairs, trade ideas, and world snapshot. No authentication is required for read tools.
How does SimpleFunctions verify its own accuracy, and how does that compare to Predly's claimed 89%?+
GET /api/calibration returns SF's own Brier scores segmented by venue, category, and price bucket, computed over a rolling 90-day window. Current figures: Kalshi 0.20, Polymarket 0.12 on T-24h price. These are live production numbers verifiable with a curl call. Predly claims 89% alert accuracy, but the methodology — how alerts are defined, which time windows, which contract types — is not publicly documented, making direct comparison difficult.
Does Predly have a public API I can integrate?+
Based on Predly's public website and documentation as of 2026-04, no public REST or programmatic API is documented. Their product appears to be a consumer-facing alert interface. If you need to integrate mispricing signals into a code-driven pipeline, agent, or CLI workflow, SimpleFunctions' fully documented public REST API — covering prices, orderbook, indicators, cross-venue pairs, and trade ideas — is the addressable surface.
What computed indicators does SimpleFunctions expose that Predly does not?+
SF pre-computes six indicators across 48K+ active contracts: IY (implied yield), CRI (cliff risk index measuring probability of sharp price discontinuity near settlement), LAS (liquidity availability score), EE (event overround measuring aggregate take rate), τ-days (time to settlement), and regime label (adverse-selection classification). These are available at /screen and via the MCP server. Predly's documented surface is limited to the AI probability vs market price comparison.
Start for free.
Public endpoints are free for normal usage and rate-limited for reliability. Authenticated endpoints are free up to 15M tokens, then pay per token. No credit card to start.