> ## Documentation Index
> Fetch the complete documentation index at: https://docs.polynode.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Case studies

> Dated, reproducible studies of PolyNode performance, coverage, and production behavior.

PolyNode case studies test a concrete product claim against a defined denominator. Each study records its date, endpoints, subscription semantics, source versions, measurement clock, eligibility window, adverse observations, and limitations.

<CardGroup cols={2}>
  <Card title="PolyNode vs Struct and RTDS" icon="stopwatch" href="/case-studies/struct-rtds-websocket-benchmark-2026-08-01">
    WebSocket latency, pending access, and confirmed-fill completeness measured on August 1 and retested on August 4, 2026.
  </Card>

  <Card title="Polygon block time and Polymarket" icon="clock" href="/case-studies/polygon-block-time-polymarket-2026-08-04">
    How Polygon's move from 2.0-second to 1.5-second blocks compressed Polymarket's pre-confirmation window without moving PolyNode's position in the transaction lifecycle.
  </Card>
</CardGroup>

## Evidence standard

Every benchmark case study should:

* define equivalent signals before measuring them;
* timestamp all competitors from the same monotonic clock and process;
* use an independent completeness denominator when one exists;
* exclude partially observed boundary intervals;
* preserve per-sample data, missing identifiers, source hashes, and connection lifecycle events;
* report unfavorable windows and unavailable comparisons;
* separate measured results from broader product claims;
* publish code with credentials supplied only at runtime.

These are point-in-time measurements, not guarantees about every network location or future provider version.