Metrics

The metrics include provides AI-driven, machine-learned insights—such as predictions and smart money indicators—categorized by domain-specific groups to help identify advanced match trends.

Availability: This data is only included in the response when explicitly requested using the query parameter include=metrics.

Object Schema

Field	Type	Description
team_id	integer	The ID of the team the metric relates to (uses 0 for neutral or match-level metrics).
type_id	integer	The unique ID associated with the metric type.
developer_name	string	The unique constant identifier (e.g., PREDICTION_GL_CONFIDENCE).
value	number	The calculated value or probability of the metric.
meta	object	Optional. Additional context such as reasons, line, or labels. Omitted if null.

Dynamic Metadata
The meta object is dynamic. Its properties (such as line, reasons, or note) vary based on the developer_name. Developers should implement flexible parsing for this object as fields are specific to the individual metric type.

Filtering Metrics

You can filter metrics by specific IDs or entire domain groups using the following patterns:

By Type IDs: filter[metrics]=types:1,2,3
By Developer Types: filter[metrics]=developer_types:predictions,smartmoney

Example Response

"metrics": [
  {
    "team_id": 0,
    "type_id": 105,
    "developer_name": "PREDICTION_GL_CONFIDENCE",
    "value": 70,
    "meta": {
      "note": "Base Confidence=70 (Clear xMetric advantage), Bonus=0, Final Confidence=70"
    }
  },
  {
    "team_id": 0,
    "type_id": 106,
    "developer_name": "PREDICTION_GL_UNDER",
    "value": 2.93,
    "meta": {
      "line": 3,
      "reasons": ["Both teams create few dangerous attacks"],
      "market_id": 3
    }
  }
]