# bias zones

Detection of systematic biases in prediction market pricing, based on Becker (2025) research.

## Two Types of Bias

### Longshot Bias

Markets with low YES probability (typically <15%) tend to be **overpriced**. People overestimate the chance of unlikely events.

* The crowd buys lottery tickets
* A market at 8% might have a true probability closer to 4%
* NO positions in longshot markets have a structural edge

### Favorite Bias

Markets with high YES probability (typically >85%) tend to be **underpriced**. People underweight near-certain events.

* The crowd doesn't pay up enough for high-probability outcomes
* A market at 92% might have a true probability closer to 96%
* YES positions in favorite markets have a structural edge

## Detection

The terminal flags bias zones based on:

| Zone                | Price Range | Bias Direction       | Signal             |
| ------------------- | ----------- | -------------------- | ------------------ |
| **Strong Longshot** | <10%        | Overpriced           | Strong NO edge     |
| **Longshot**        | 10-20%      | Slightly overpriced  | Moderate NO edge   |
| **Neutral**         | 20-80%      | No systematic bias   | Evaluate on merits |
| **Favorite**        | 80-90%      | Slightly underpriced | Moderate YES edge  |
| **Strong Favorite** | >90%        | Underpriced          | Strong YES edge    |

## How the Terminal Uses It

{% stepper %}
{% step %}

### Model adjustment

The composite probability applies a bias correction when markets are in extreme zones.
{% endstep %}

{% step %}

### Visual warning

Markets in bias zones show a badge in the Models tab.
{% endstep %}

{% step %}

### Kelly interaction

Empirical Kelly's uncertainty penalty increases in bias zones (edges look larger but are harder to estimate).
{% endstep %}
{% endstepper %}

{% hint style="info" %}
Why It Matters

Bias zones don't tell you the outcome. They tell you that the price is likely systematically wrong in a particular direction. This is a statistical edge that exists across many markets, not a guarantee for any single market.

The strongest application is portfolio-level: consistently betting against longshot bias and with favorite bias has historically been profitable across prediction markets.
{% endhint %}

<details>

<summary>Caveats</summary>

* Bias detection uses fixed price thresholds. The actual bias boundary varies by market type and liquidity.
* Low-volume markets in bias zones are less reliable (thin markets can sit at extreme prices due to lack of activity, not mispricing).
* Near-resolution markets naturally move to extremes, which is convergence, not bias.

</details>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://yesno-1.gitbook.io/yesno-docs/bias-zones.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.