# analysis

Middle-right panel. AI-powered market analysis with three tiers of depth.

## AI Verdict

When you select a market, the Analysis panel generates:

* **Verdict**: YES, NO, or LEAN YES/NO
* **Confidence**: Percentage score
* **Summary**: One-line assessment

The verdict is generated by an AI model using real-time news, market data, and quantitative signals.

## Intelligence Brief

Below the verdict, a structured breakdown covers six dimensions:

| Section         | What It Covers                                              |
| --------------- | ----------------------------------------------------------- |
| **Edge**        | Where the model sees mispricing or opportunity              |
| **Timeline**    | Resolution date, urgency level, time-sensitive factors      |
| **Momentum**    | Price direction, acceleration, trend strength               |
| **Smart Money** | Whale activity, holder concentration, institutional signals |
| **Data**        | Volume trends, market activity, liquidity depth             |
| **Contrarian**  | Cases where the crowd may be wrong                          |

## Analysis Tiers

Three tiers of analysis depth, each using a different AI model:

| Tier           | Model            | Tokens | Use Case                                    |
| -------------- | ---------------- | ------ | ------------------------------------------- |
| **Scout**      | Gemini 2.0 Flash | 300    | Bulk screening, quick reads                 |
| **Analyst**    | Claude Haiku 4.5 | 1,500  | Standard analysis with full quant framework |
| **Strategist** | Claude Sonnet 4  | 1,500  | Deep dive with full intelligence brief      |

The default tier for market selection is Analyst. Click **Deep Analysis** to trigger a Strategist-tier analysis.

## How It Works

{% stepper %}
{% step %}

### 1. News fetch

Brave Search API finds 5 recent articles relevant to the market. Search queries are sector-aware (e.g., sports markets search for "odds prediction injury report").
{% endstep %}

{% step %}

### 2. Signal integration

If enriched signals are available (9 categories), they're included as market intelligence context.
{% endstep %}

{% step %}

### 3. AI generation

The AI model receives market data, news, signals, and a sector-specific system prompt. It produces a structured JSON response with verdict, confidence, and analysis sections.
{% endstep %}

{% step %}

### 4. Fallback

If the primary AI provider fails, a fallback provider (Groq) is used. Up to 2 retries with backoff.
{% endstep %}
{% endstepper %}

## Methodology

The AI is instructed to follow a strict analytical framework:

{% stepper %}
{% step %}

### 1. Fact Extraction

What is verifiably true right now?
{% endstep %}

{% step %}

### 2. Base Rate

Start from the current market price as the crowd's estimate
{% endstep %}

{% step %}

### 3. News Delta

What has changed since the market last moved?
{% endstep %}

{% step %}

### 4. Edge Calculation

Is there a gap between fair value and market price?
{% endstep %}

{% step %}

### 5. Confidence Assignment

How certain is the assessment?
{% endstep %}
{% endstepper %}

The system prompts enforce zero hallucination: the AI must cite sources and flag uncertainty rather than speculate.


---

# 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/terminal/analysis.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.