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

# Calibration

> How often markets at each price actually resolved Yes.

A market priced at 30¢ is saying Yes happens about 30% of the time. **Calibration** checks that against settled markets: of every market priced in a band some days before it closed, how many resolved Yes?

```bash theme={null}
curl "https://api.routeur.app/v1/insights/calibration?venue=kalshi&category=Sports&horizon_days=7" \
  -H "X-API-Key: $ROUTEUR_API_KEY"
```

| Parameter      | Values                                                     |
| -------------- | ---------------------------------------------------------- |
| `venue`        | `kalshi`, `polymarket`, or `all` for both together         |
| `category`     | A category, or `All`                                       |
| `horizon_days` | `1`, `7` or `30`: how long before close the price is taken |

Each 10-cent band returns `markets`, `avg_price_usd` and `yes_rate`. The gap between them is the band's edge:

* **`yes_rate` above `avg_price_usd`** — Yes resolved more often than priced; Yes was cheap.
* **`yes_rate` below `avg_price_usd`** — Yes was rich, so No was cheap.

## Where the prices come from

| Venue      | Settled markets                                                                        | Price used                                                              |
| ---------- | -------------------------------------------------------------------------------------- | ----------------------------------------------------------------------- |
| Kalshi     | Every market settled since 2021, from Kalshi's archive                                 | Midpoint of the best bid and ask in the daily candle, or the last trade |
| Polymarket | Markets in imported history that have settled, rechecked every few hours as more close | Last recorded probability of the day                                    |

<Warning>
  Kalshi history covers tens of thousands of settled markets; Polymarket's is still small, so many of its bands have too few markets to mean much. Always read `markets` before reading `yes_rate`, and remember this is history, not a forecast.
</Warning>

To turn a band into a rule you can act on, with a confidence range and the markets that fit it today, see [Strategies](/concepts/strategies).