Replay the book. Test the fill. Keep the work.
TickFoundry Lab is a research workbench on top of the same capture that ships in every bundle: tick-by-tick Polymarket and Hyperliquid order books, replayed event by event in the browser, with a visible-book execution simulator anchored to any timestamp and private workspaces that reopen exactly where you left them.
No account needed for the sample: BTC Up or Down 4h on 2026-08-19 UTC, every outcome token, 25 levels, the same day as the public download. A free account adds the market·days you claim; Explorer and Premium include the Lab at no extra charge and open any day on your plan.
Replay and inspect
Mid, bid and ask over the window, a 25-row ladder with cumulative quantity and notional, the trade tape, and an inspector that shows the exact nanosecond receive time, the venue time, the clock kind, the depth cap and every quality flag. Step every event, or play at 1x, 10x, 100x and max with skipped frames counted.
Visible-book execution analysis
Walk a hypothetical buy or sell through the recorded ladder at the cursor: fills per level, VWAP, notional, residual, slippage in price units and basis points, snapshot age and completeness. Independent scenarios on historical snapshots, never a guaranteed fill and never an order.
Workspaces that reopen exactly
Save a replay or a simulation with its instrument, interval, cursor and view into a private workspace with markdown notes and bookmarks. Export the JSON definition and a Python snippet that reproduces the same numbers with the SDK.
Your data rights, unchanged
The Lab is included with every account and opens what your plan already covers: Explorer every day on your plan at 10 levels (Hyperliquid 25), Premium at 25 levels with a priority queue. Free accounts open the public sample day and the market·days they claimed; visitors get the sample day. Locked days are shown with the reason, never hidden.
Curated research recipes (spread and depth into expiry, slippage curves by order size, Hyperliquid liquidity by distance from mid), a Python SDK for Lab runs, and more venues in the workbench. The Lab stays included with your account.
- Everything shown is historical and labelled so; UTC everywhere.
- A downsampled chart says its resolution; stepping every event is exact.
- Missing data is never zero: an empty bin under missing coverage is unknown.
- Fees are excluded unless you choose a model, and the result says so.
- Private research (notes, parameters, results) never reaches analytics; event names and timings only.