prediction_calibration

shallow

io.github.pipeworx-io/aviationstack · Verify this server

REALIZED win rate by shown price and minutes elapsed for short-horizon crypto up/down markets, net of taker fee — read from our own settled-outcomes archive, not from live prices. Answers the one question no venue will: "a side shown at 82 with five minutes left, how often did it ACTUALLY win?" Returns a grid of (price bucket x minute elapsed) -> n, realized_win_rate_pp, mean_shown_price_pp, calibration_gap_pp (realized minus shown — zero means the book is calibrated), breakeven_win_rate_pp (the shown price plus the taker fee that was actually in force for that market), and net_edge_pp (realized minus breakeven, which is what a taker would have kept). WHY AN ARCHIVE EXISTS AT ALL: a settled 15-minute window is exactly what these venues stop serving. Measured 2026-10-22, a polymarket.com window drops out of the default market listing within minutes of closing (its trade tape and resolution stay retrievable by a closed-market lookup, which the collector now uses), and the other venues serve settled windows only briefly or thinly. So this tool reads a local archive collected window-by-window as they settle. THE THREE VENUES ARE NEVER POOLED. all three compare an END 60-second average with a START 60-second average, but on different feeds: polymarket.com on a Chainlink TWAP, polymarket-us and Kalshi on a CF Benchmarks BRTI simple average. The feeds disagree near the threshold (measured 2026-10-24 on btc: polymarket.com and Kalshi settled differently on 1.6% of windows both hold, polymarket-us and Kalshi on none), so venue:"all" returns one grid PER VENUE and `venues_pooled` is always false. EVERY RATE CARRIES ITS 95% INTERVAL, AND EVERY EDGE SAYS WHETHER IT SURVIVES IT. Each cell gives realized_win_rate_ci95_lo/hi_pp (Wilson, so it never reports an impossible win rate above 100), the same interval carried onto net_edge_ci95_lo/hi_pp, and net_edge_distinguishable_from_zero. READ THAT BOOLEAN BEFORE net_edge_pp: at these sample sizes most cells are false, meaning the interval straddles zero and a net_edge_pp of +4pp is NOT evidence of an edge. For scale, at a 90% win rate even n=300 gives a 95% half-width of ~3.4pp while net edges here run +2 to +4pp — so a comfortable-looking sample still cannot call a small edge real. `verdict` says it per cell in plain words, and the first note says how many cells survived. A THIN CELL WITHHOLDS ITS RATE ENTIRELY. Any bucket holding fewer than 30 settled windows reports status "thin" with realized_win_rate_pp and every interval null rather than printing a rate from a sample too small to mean anything — a 2-sample "100% win rate" is the most misleading thing this tool could say. ALWAYS READ `coverage` BEFORE THE GRID: it gives the first and last window held, windows_total vs windows_settled, every UTC day inside the span with no windows at all, and the longest run of consecutive missing windows. A grid over a store with a three-day hole looks entirely normal and is measuring a different population. mean_fill_price_pp and fill_vs_shown_pp are populated for polymarket.com only — it is the one venue of the three that publishes a trade tape for these markets, so it is the only place realized FILL can be compared with the shown price instead of assumed equal to it.

100.0/100

1 trials · measured 10 days ago

prediction_calibration scores 100.0/100 on Vouch's measured behaviour index, from 1 real invocation trials against io.github.pipeworx-io/aviationstack, measured 12 Nov 2026 under methodology v0.2.0. Every measured component scored 100.

Component breakdown

ComponentWeightValue
Reliability35%not applicable
Schema integrity25%100.0
Failure behaviour15%not applicable
Latency15%not applicable
Concurrency10%not applicable

Tool details

Transport
remote + stdio
Credential class
self-provisionable
Input schema
not declared
Output schema
not declared
Side-effect classification
unclassified

Score history

DayScoreTierMethodology
2026-11-12100.0shallowv0.2.0

Probe evidence

ProbeOutcomes
schema_integritypass: 1

Raw request/response logs are not archived yet — the outcome counts above are drawn directly from every recorded trial.

Embed this score

Available for every tool, scored or not — not a verification perk. Always links back to this page.

Vouch score: prediction_calibration
[![Vouch score](https://vouch.tools/api/tools/bc6ce8af-fb64-424e-9479-db0837ecc7c9/badge.svg)](https://vouch.tools/tools/bc6ce8af-fb64-424e-9479-db0837ecc7c9)
prediction_calibration — Vouch