Depth capture
DepthFeed Alternative for Polymarket Order Book Data
Overview
Quick comparison
Side-by-side from public documentation.
| Feature | DepthFeed | PolyOrderbooks |
|---|---|---|
| Historical prices | Yes — prices in snapshot payloads | Yes |
| Historical L2 books | Yes — full bid/ask ladders (up to 100 levels on Kalshi per docs) | Yes — full bid/ask ladders |
| Capture model | Event-driven websocket capture (Polymarket/Limitless); interval downsampling 1s–1d | 250ms capture; 250ms query resolution on every plan, including free |
| REST API | Yes — Bearer token REST API | Yes |
| Python tooling | Not publicly documented | Official Python SDK on PyPI |
| Bulk export | Not publicly documented | Enterprise S3 delivery (Parquet, CSV, JSON) — not on self-serve plans |
| Free tier | Yes — Explorer $0, 7-day BTC sample, 1 req/sec | Yes — Starter: 250ms L2, prices, and metrics at $0 (3-day history) |
| Starting paid price | $29/month Quant (public pricing at time of verification) | $19/month Pro (public pricing at time of verification) |
| Prediction-market coverage | Polymarket, Kalshi, Limitless | Polymarket crypto markets (Up/Down and related) — depth history, not metadata breadth |
| Historical window | Rolling 7/30/90-day windows by plan; full archive on Desk ($249/mo) | Continuous 250ms archive from 3-day Starter windows to 30–120-day paid windows; resolved markets stay queryable |
Source (DepthFeed): https://depthfeed.com/docs
Event-driven vs sampled query resolution
DepthFeed stores observations when the book changes; you can pull raw rows or downsample with interval without interpolating missing buckets.
PolyOrderbooks snapshots every 250ms and exposes query grids from 60s to 250ms — predictable joins across prices, metrics, and L2.
Where DepthFeed fits best
These fit event-stream workflows; stored-grid REST workflows are the PolyOrderbooks section below.
- Event-driven capture that preserves every book change without interpolation
- Multi-venue sports and crypto with live WebSocket feeds
- Backtest Lab, wallet intelligence, and Explorer tier on paid plans
- Recorded observations for event-driven research workflows
Where PolyOrderbooks fits best
- Uniform 250ms capture — predictable joins across prices, metrics, and L2
- Official Python SDK and MCP server for the same REST endpoints
- Resolved markets stay queryable; free L2 sample to validate before paying
- Published monthly API pricing with a free Starter tier
Backtesting use cases
DepthFeed fits multi-venue strategies needing recorded observations and live stream validation. PolyOrderbooks fits Polymarket crypto REST backtests at second-level grids.
Inspect a free 1-second L2 sample to compare ladder shape against DepthFeed exports.
FAQ
What is DepthFeed?
DepthFeed records full bid/ask ladders for Polymarket, Kalshi, and Limitless and serves discovery, metadata, and historical snapshots over a metered REST API plus WebSocket live books.
Does DepthFeed have historical Polymarket order books?
Yes — according to DepthFeed documentation, historical snapshots can include the full stored ladder via include_orderbook=true, with plan-dependent history windows.
DepthFeed vs PolyOrderbooks for backtesting?
DepthFeed uses event-driven capture with interval downsampling — strong when you want recorded observations at fine granularity across multiple venues. PolyOrderbooks uses uniform 250ms capture with REST query resolution — strong for Polymarket crypto API-first workflows without managing rolling plan windows.
What are DepthFeed alternatives?
PolymarketData and PolyOrderbooks for REST historical L2, Telonex for Parquet tick pipelines, Predexon for multi-venue REST plus ticks, polyReplay for BTC/ETH Up/Down tick archives.