← Back to blog

ZenHodl vs PMXT: Compare the Historical Data You Need

updated 2026-10-08 comparison data polymarket kalshi pmxt

By ZenHodl. Dataset documentation, research and model evaluations are linked in the article. A separate filtered ledger of bot-attributed trades, including losses and its admission rules, is public at /results.

PMXT and ZenHodl both offer ways to obtain historical prediction-market data. The old distinction “PMXT for live data, ZenHodl for history” is incomplete. Choose by the actual contract and window you need, then compare sample files.

ZenHodl sells the products discussed here. This is a focused comparison, not an independent ranking or a complete audit of either provider.

What PMXT documents

PMXT's open-source SDK is a venue abstraction for data and trading. Its hosted pricing page also lists historical order-book snapshots and cross-exchange matching. The public archive offers free recorded data for Polymarket and Kalshi, among other venues. These primary sources were checked on October 8, 2026.

A free SDK, hosted API and downloadable archive are different access options. Verify each option's coverage, timestamp fields, schema, license and limits. A listed matching feature does not establish that every sports contract is matched correctly for your research task.

What ZenHodl supplies

ZenHodl sells fixed downloadable releases for separate tasks:

The sample catalog links the released schemas. Use the current product pages for prices and reviewed release metadata. A fixed archive does not become a continuing feed because its vendor also sells an API.

Compare the artifacts, not the labels

Your question Check before choosing
Can I study this sport and month? Exact contract inclusion, first/last book observations and per-market gaps.
Can I replay book changes? Message versus snapshot semantics, retained levels, update cadence and reconnect gaps.
Can I align trades to score events? Exchange event time, retrieval time and a separately verified score timeline. Capture-time context is not enough.
Can I compare venues causally? Outcome mapping, settlement rules, quote age and a backward-only matching rule. Nearest-time matches may use future information.
Can I publish or redistribute the result? The particular file's license and the service's access terms.

PMXT may fit a unified API or a no-cost recorded-data workflow. ZenHodl may fit a particular frozen sports artifact whose schema and limitations you have inspected. Historical data, file ownership and cross-venue matching are not categorical ZenHodl exclusives.

Try the relevant free files

Compare a PMXT archive file with a ZenHodl sample for the same research question. Count observations and check field meaning before comparing advertised coverage. The historical-data buyer's guide provides a checklist; the Kalshi walkthrough demonstrates one dated audit.

Updated October 8, 2026: corrected the former live-only PMXT description and unlisted-pricing claim. No full payload comparison or exhaustive coverage ranking was performed for this update.

Related reading

Get ZenHodl Weekly

Dataset releases, research notes, and public results, including corrections.

Research and dataset updates from ZenHodl.

Want the data behind this post?

Historical sports prediction-market datasets with measured coverage, documented schemas, and disclosed gaps.