For every resolved Polymarket market, OddsBridge captures the last tick before resolution — the "closing line" — and benchmarks it against the actual outcome. Use this to gauge how sharp PM is by sport, how well-calibrated its prices are, and where your own bets stand vs the close. The dataset rebuilds nightly from a tick-level history no third-party aggregator publishes.
Bucket the closing implied probability into 10% bands. A perfectly calibrated market would have the actual YES rate track the bucket midpoint (50–60% bucket → ~55% YES rate).
| Closing price bucket | Markets | Resolved YES | Drift vs midpoint |
|---|---|---|---|
| 0–10% | 0 | — | — |
| 10–20% | 0 | — | — |
| 20–30% | 0 | — | — |
| 30–40% | 0 | — | — |
| 40–50% | 0 | — | — |
| 50–60% | 0 | — | — |
| 60–70% | 0 | — | — |
| 70–80% | 0 | — | — |
| 80–90% | 0 | — | — |
| 90–100% | 0 | — | — |
| Resolved | Market | Outcome | Closing price | Result |
|---|
For each Polymarket market that resolves, we identify the last tick (price quote) we captured for the YES side before the resolution timestamp. That price is the closing line — the market's belief about the outcome the instant before resolution. Calibration buckets group markets by their closing price into 10 equal bands. Favorite accuracy looks at all markets where the closing price was > 50% and asks how often they actually resolved YES. We exclude markets with placeholder end dates (UMA marks some markets with 2028-01-01) and markets where we have no tick within 6 hours of the resolution timestamp.
Want CLV on your own bets? Connect your PM wallet and OddsBridge surfaces a CLV column per bet plus aggregate avg-CLV and beat-close-rate.