# composite probability

The terminal's core model. Synthesizes 11 independent signals into a single fair value estimate for any prediction market.

## How It Works

The composite probability starts from the current market price (the crowd's estimate) and applies adjustments from 11 signal categories. Each signal can shift the estimate up or down within defined bounds.

```
Fair Value = Market Price + Sum(Signal Adjustments)
```

Each signal adjustment is:

```
Adjustment = Signal Weight x Signal Strength x Max Shift
```

Where:

* **Signal Weight**: How much influence this signal has (configurable via calibration)
* **Signal Strength**: Normalized value from -1 to +1 based on the signal's reading
* **Max Shift**: Maximum percentage points this signal can move the estimate

## The 11 Signals

| #  | Signal                      | Max Shift | What It Measures                                    |
| -- | --------------------------- | --------- | --------------------------------------------------- |
| 1  | **Momentum (7d)**           | +/- 8%    | Short-term price direction and acceleration         |
| 2  | **Momentum (30d)**          | +/- 6%    | Medium-term trend                                   |
| 3  | **Volume Trend**            | +/- 5%    | Whether trading activity is increasing or declining |
| 4  | **Whale Flow**              | +/- 7%    | Net direction of large trades                       |
| 5  | **Holder Conviction**       | +/- 6%    | Concentration and balance of YES vs NO holders      |
| 6  | **Buy Pressure**            | +/- 5%    | Recent vs historical buy ratio                      |
| 7  | **Smart Money**             | +/- 8%    | Combined holder + whale + momentum signal           |
| 8  | **Volume-Price Divergence** | +/- 5%    | When price and volume disagree                      |
| 9  | **Lifecycle**               | +/- 3%    | Market maturity and urgency adjustments             |
| 10 | **News Sentiment**          | +/- 6%    | Direction from AI-analyzed news                     |
| 11 | **Order Book Imbalance**    | +/- 4%    | Bid vs ask depth ratio                              |

{% hint style="info" %}
Bounds

The fair value is clamped to \[1%, 99%] to prevent impossible probabilities. Individual signal adjustments are capped at their max shift values.
{% endhint %}

## Interpreting the Output

| Scenario                          | Meaning                                                          |
| --------------------------------- | ---------------------------------------------------------------- |
| Fair value **above** market price | Model sees the market as underpriced (potential YES opportunity) |
| Fair value **below** market price | Model sees the market as overpriced (potential NO opportunity)   |
| Fair value **near** market price  | Signals are mixed or the market is efficiently priced            |
| **Large gap** (>10%)              | Strong signal alignment. Check which signals are driving it.     |
| **Small gap** (<3%)               | Marginal edge. Taker cost may eat the difference.                |

{% hint style="info" %}
Calibration

All signal weights are adjustable via the [Calibration](file:///6948179/calibration/parameters.md) panel. The three presets (Conservative, Balanced, Aggressive) apply different weight profiles. You can also set weights manually.
{% endhint %}

{% hint style="warning" %}
Limitations

* The model assumes signals are partially independent. In practice, some signals correlate (e.g., whale flow and smart money).
* News sentiment depends on AI accuracy. The model can be wrong when news is ambiguous or fast-breaking.
* Low-volume markets produce noisier signals with wider confidence intervals.
  {% endhint %}


---

# 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/composite-probability.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.