Alternative · Analytics aggregator
Probalytics vs
SimpleFunctions.
Same upstream venues (Kalshi + Polymarket). Probalytics surfaces the microstructure: 1ms orderbook resolution, ClickHouse SQL access, and Parquet bulk export — purpose-built for quantitative researchers who need raw tick fidelity. SimpleFunctions ships the agent layer above that data: causal-tree thesis system, autonomous trading with a 7-gate risk cascade, calibrated world model, computed indicators across 48K contracts, and a 56-tool MCP server that drops into Claude Code or Cursor in one line.
Verified 2026-04 · public sources only · live SimpleFunctions data from /calibration
Category
Analytics aggregator
Differences
11
Use cases
4
Verified
2026-04
Verdict
Pick the one that fits how
you actually work.
Choose SimpleFunctions if
You are building agents, autonomous trading systems, or research that needs more than raw tick data — 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 index, liquidity availability score), and a 56-tool MCP server that exposes all of it to any agent framework without additional configuration.
Choose Probalytics if
You specifically need 1ms orderbook resolution, direct ClickHouse SQL for tick-level analytical queries, or Parquet bulk export at scale — Probalytics is built for quants and HFT-style buyers who need the raw microstructure, not the agent abstraction layer on top of it.
Same upstream venues. Probalytics surfaces tick-level microstructure via SQL and bulk export. SimpleFunctions ships the agent layer above it: world model, theses, autopilot, MCP.
At a glance
Three things that
actually differ.
Everything Probalytics gives you — normalised prices on Kalshi and Polymarket, orderbook depth, and historical data — SimpleFunctions also gives you, indexed across the same 48K+ active contracts.
On top of that, SF ships a causal-tree thesis system, Portfolio Autopilot (1M-context LLM, 7-gate risk cascade), and 56 MCP tools that no current prediction market data product exposes.
SF publishes live Brier scores at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days — so you can audit the platform's accuracy before committing to a workflow.
Side by side
11 dimensions · verified 2026-04SimpleFunctionsKalshi + Polymarket normalised, 48K+ active contracts indexed and searchable via REST, MCP, and CLI.
ProbalyticsKalshi + Polymarket normalised at 1ms resolution; Probalytics cites 813K Polymarket markets and 40K Kalshi markets covered.
SimpleFunctionsGET /api/public/market/{ticker}?depth=true returns bid/ask ladder, spread, and slippage estimate.
Probalytics1ms orderbook resolution with 200-500M event updates per day — the product's headline speciality for microstructure research.
SimpleFunctionsPer-contract history accessible via /api/public/market/{ticker}/history; not optimised for sub-millisecond replay.
ProbalyticsFull tick-level history queryable via direct ClickHouse SQL and exportable as Parquet — purpose-built for batch quant workloads.
SimpleFunctionsImplied yield (IY), cliff risk index (CRI), liquidity availability score (LAS), event overround (EE), τ-days, and regime label pre-computed across 48K contracts at /screen.
ProbalyticsRaw price, volume, and spread data; derived signals are computed by the buyer.
SimpleFunctionsLive Brier scores at /api/calibration — segmented by venue, category, and price bucket; Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days.
ProbalyticsNot published.
SimpleFunctionsPOST /api/thesis/create decomposes any sentence into a causal tree, propagates probabilities, scans Kalshi + Polymarket for tradeable edges, and runs a continuous auto-evaluation heartbeat.
ProbalyticsNot in scope.
SimpleFunctionsPortfolio Autopilot — 1M-context LLM, 13 data sources, 7-gate risk cascade (kill switch, position limits, drawdown gate, regime check) before execution.
ProbalyticsNot in scope.
SimpleFunctions56 tools via `claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp`; works with Claude Code, Cursor, and any MCP-compatible client.
ProbalyticsNo MCP server published.
SimpleFunctionsREST polling; real-time streaming is not the product's primary focus.
ProbalyticsServer-Sent Events streaming is on Probalytics's public roadmap but not live as of 2026-05.
SimpleFunctionsPublic datasets on HuggingFace and Kaggle, daily refresh, CC-BY-4.0 — designed for downstream LLM ingest and research replication, not columnar OLAP queries.
ProbalyticsParquet bulk export is a core offering and a primary differentiator for analytical batch workloads.
SimpleFunctionsPublic REST, MCP, and CLI reads require no authentication; pay-per-token only on thesis and intent execution above 15M tokens.
ProbalyticsBearer token authentication (api_xxx:sk_xxx) required; tiered plans listed at probalytics.io/pricing.
Methodology
Verified 2026-04 from public sources only — Probalytics's documentation, public website, and publicly observable behaviour. We never claim non-public information about Probalytics'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 research agent that needs to query prediction markets, evaluate a thesis, and execute autonomously.
SimpleFunctions · best fit
SF's 56-tool MCP server exposes the full surface — thesis creation, indicator lookup, cross-venue pair scanning, world snapshot, trade ideas — to any agent framework in a single setup line. Portfolio Autopilot handles execution behind a 7-gate risk cascade. No equivalent agent layer exists in Probalytics.
Probalytics
Probalytics is a data product, not an agent framework. It exposes REST and ClickHouse SQL for raw market data; the agent layer, thesis logic, and execution pipeline would need to be built entirely by the buyer.
Scenario 02
Running a quantitative strategy that requires 1ms tick fidelity, direct SQL access, and Parquet exports of full event history.
SimpleFunctions
SF provides per-contract history and REST endpoints but is not optimised for 1ms tick fidelity, columnar SQL queries, or Parquet-based batch pipelines.
Probalytics · best fit
Probalytics is purpose-built for this workload: 1ms orderbook resolution, direct ClickHouse SQL, Parquet bulk export, and 200-500M event updates per day. If tick-level quant research is the requirement, Probalytics is the better fit.
Scenario 03
Decomposing a geopolitical thesis into testable sub-claims and tracking how market prices update each sub-claim over time.
SimpleFunctions · best fit
POST /api/thesis/create decomposes the thesis into a causal tree, maps each node to relevant Kalshi and Polymarket contracts, propagates probabilities, and runs continuous evaluation cycles (news scan, price refresh, LLM eval, confidence update). Signals can be injected via /api/thesis/{id}/signal.
Probalytics
Probalytics provides the underlying price data with high fidelity but has no thesis decomposition, causal modelling, or evaluation pipeline. A buyer would need to implement this reasoning layer from scratch.
Scenario 04
Detecting arbitrage between the same event priced on Kalshi and Polymarket.
SimpleFunctions · best fit
GET /api/public/cross-venue/pairs?preset=arb returns normalised matched pairs with spread and conviction metrics already computed. The /screen surface adds CRI, LAS, and IY per contract for further filtering.
Probalytics
Probalytics covers both venues at 1ms resolution, which is sufficient for arbitrage detection, but cross-venue pair matching and derived spread metrics would need to be implemented by the buyer on top of the raw feed.
Migrate
From https://api.probalytics.io/v1/markets/{id}/orderbook to SimpleFunctions.
Same shape, no auth, same venues. Python example.
import requests
API_KEY = "api_xxx:sk_xxx"
headers = {"Authorization": f"Bearer {API_KEY}"}
resp = requests.get(
"https://api.probalytics.io/v1/markets/POLY_0x1234/orderbook",
headers=headers,
)
book = resp.json()
bids = book["bids"]
asks = book["asks"]
spread = asks[0]["price"] - bids[0]["price"]
print(f"Spread: {spread}")import requests
resp = requests.get(
"https://simplefunctions.dev/api/public/market/POLY_0x1234",
params={"depth": "true"},
)
data = resp.json()
bids = data["orderbook"]["bids"]
asks = data["orderbook"]["asks"]
spread = asks[0]["price"] - bids[0]["price"]
print(f"Spread: {spread}")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
What is SimpleFunctions's thesis system and how does it differ from a prediction market price feed?+
The thesis system at POST /api/thesis/create accepts any natural-language claim, decomposes it into a causal tree of testable sub-claims, scans Kalshi and Polymarket for contracts expressing each sub-claim, propagates probabilities through the tree, and then runs a continuous evaluation heartbeat — news scan, price refresh, milestone check, LLM evaluation, confidence update. You can inject external signals via /api/thesis/{id}/signal and fork public theses. A price feed delivers numbers; the thesis system answers: given what the market currently prices, how confident should I be in this claim?
Does Probalytics have an MCP server?+
Probalytics does not publish an MCP server as of 2026-05. SimpleFunctions ships a 56-tool MCP server accessible in one line: `claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp`. It works with Claude Code, Cursor, and any MCP-compatible client, exposing thesis creation, indicator lookup, cross-venue pair scanning, world snapshot, and trade ideas without additional configuration.
What is Portfolio Autopilot and how does it work?+
Portfolio Autopilot is SF's autonomous trading agent. It uses a 1M-context LLM, 13 data sources, and a 7-gate risk cascade — kill switch, position limits, drawdown gate, regime check, and additional guards — before placing any order. It is designed for builders who want a fully reasoned execution layer above the raw market data, not just a price feed. Probalytics does not offer an equivalent execution layer; it is a data-only product targeting analysts and quants who build their own execution.
Can I do ClickHouse SQL queries or Parquet exports on SimpleFunctions data?+
SimpleFunctions does not offer direct ClickHouse SQL access or Parquet export. Its data surfaces are REST (48K+ contracts, orderbook, cross-venue pairs, indicators), MCP (56 tools), and public HuggingFace/Kaggle datasets refreshed daily under CC-BY-4.0. If your workflow requires columnar OLAP queries over raw tick data at millisecond resolution, Probalytics is purpose-built for that use case. SF is optimised for agent-layer workloads: reasoning, thesis evaluation, indicator computation, and autonomous execution.
What computed indicators does SimpleFunctions publish and where are they?+
SF computes six indicators per contract across 48K+ active positions: implied yield (IY), cliff risk index (CRI), liquidity availability score (LAS), event overround (EE), time to settlement in days (τ-days), and a regime label classifying adverse-selection risk. They are pre-computed and accessible at /screen, per-contract via the REST API, and through MCP tools. Probalytics surfaces raw price, volume, and spread data; building equivalent derived signals requires additional computation on the buyer's side.
How does SimpleFunctions audit its own prediction accuracy?+
GET /api/calibration returns live Brier scores computed by SF itself, segmented by venue, category, and price bucket. Current figures: Kalshi 0.20, Polymarket 0.12, measured on T-24h prices over the past 90 days. These numbers update as markets settle and are re-verifiable with a single curl command. Probalytics does not publish equivalent accuracy benchmarks. Calibration data is useful when sizing positions or weighting signals: it tells you where the platform's consensus probability is historically reliable versus systematically biased.
Does Probalytics plan to add more venues beyond Kalshi and Polymarket?+
Probalytics's public roadmap lists PredictIt, Metaculus, and Manifold as planned integrations, but those venues are not live as of 2026-05. SimpleFunctions currently focuses on Kalshi and Polymarket as the primary tradeable venues, with 48K+ active contracts indexed cross-venue. The meaningful difference between the two products is not venue coverage — both cover the same core venues — it is what each product does with the data above the raw feed.
What cross-venue arbitrage tools does SimpleFunctions provide?+
GET /api/public/cross-venue/pairs?preset=arb returns normalised matched pairs across Kalshi and Polymarket with pre-computed spread and conviction metrics. The /screen surface layers in CRI (cliff risk), LAS (liquidity availability), and IY (implied yield) so you can filter for positions where the arbitrage is liquid and the timing risk is low. Trade ideas with catalyst and conviction are available at /api/public/ideas. Probalytics provides the raw tick data on both venues, but cross-venue pair matching and indicator computation would need to be implemented by the buyer.
Same category
Other Analytics aggregator alternatives.
SimpleFunctions vs Adjacent News
Forward-looking news platform with a unified API across 40,000+ prediction markets.
SimpleFunctions vs Dome
Developer infrastructure for prediction market data — acquired by Polymarket 2026-02-19, standalone API service in wind-down.
SimpleFunctions vs EventWaves
Polymarket analytics for finding high-edge markets via trader skill and momentum tracking.
SimpleFunctions vs FinFeedAPI
Mature normalized REST API across Polymarket, Kalshi, Manifold, and Myriad, built on CoinAPI enterprise infrastructure.
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.