# Event Profiling

<figure><img src="/files/arSilDTacGFIOTIL2sec" alt=""><figcaption></figcaption></figure>

The Event Profiling API OpenAPI specification is available for [download here.](https://sr-gitbook-prod-openapi.s3.eu-central-1.amazonaws.com/trading-and-risk-management/insight-tech-services/Event-API-production-oas30.json)\
Explore the API [endpoints here.](/itservices/models/event-profiling-api/endpoints.md)

The **Event Profiling API** is a dynamic AI-powered service within Sportradar's Insight Tech Services portfolio. It delivers real-time recommendations for live time delay (LTD) and event ratings, enabling precise, data-driven adjustments to betting validation processes. By analysing match dynamics, player profiles, market conditions, and historical performance, the API helps operators optimise bet acceptance strategies and manage liability exposure more effectively.

This solution builds upon Sportradar’s long-established Event Data Latency (EDL) model, used by MTS clients for years, and now makes these insights available via a robust API interface.

### What it does

The API provides:

* **Live Time Delay (LTD)** based on Sportradar’s Unified Odds Feed coverage type (e.g. official data feed, venue, TV).
* **Suggested LTD (sLtd)** generated from the AI-powered Event Data Latency (EDL) model. This is an optimised LTD value based on historical performance under different delay conditions.
* **Suggested Event Rating (sEr)** used to dynamically assign event risk levels on a scale from 1 (high risk, low limits) to 10 (low risk, high limits).

These values are generated and updated in real time once the first live bet is placed on an event.

### When it’s triggered

Use the Event Profiling API:

* During live betting operations to adapt LTD settings
* Before or during a match to assess event-level risk
* As part of automated decisioning for bet validation or rejection
* In conjunction with player-level profiling (e.g. LBS, sCCF) for more granular control

### Input requirements

API requests require an array of fully qualified event IDs in the format:

```json
sr:sport:{id}/sr:category:{id}/sr:tournament:{id}/sr:match:{id}
```

These identifiers represent the full event hierarchy needed to fetch valid results.

Example payload:

```json
{
  "ids": [
    "sr:sport:20/sr:category:1017/sr:tournament:36349/sr:match:45315478",
    "sr:sport:20/sr:category:88/sr:tournament:36377/sr:match:45733470"
  ]
}
```

Authentication is required. Refer to the [Endpoints](/itservices/models/event-profiling-api/endpoints.md) section for full endpoint setup.

### Output schema

Each returned item includes:

* `id`: Full event ID
* `ltd`: Live time delay based on feed type (e.g. 6s, 7s, 8s)
* `sLtd`: Suggested live time delay based on EDL model
* `sEr`: Suggested event rating (scale 1–10)
* `lastUpdate`: UNIX timestamp of latest model update

If no data is available yet, null values are returned.

### Interpretation guidance

* **ltd** reflects static coverage delays (e.g. TV or official feed) from Unified Odds
* **sLtd** is adaptive and accounts for latency-influenced betting profitability
* **sEr** defines the risk posture for an event. Use this for adjusting validation settings or applying specific business rules

Scores and recommendations are calculated as soon as live activity is detected. The API is especially powerful when used together with player-level scoring models such as **Late Bet Score (LBS)** or **Bot Score** for layered decision logic.

For detailed integration instructions, schema samples, and rate limits, consult the [Endpoints](/itservices/models/event-profiling-api/endpoints.md) and other pages.


---

# 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://docs.sportradar.com/itservices/models/event-profiling-api.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.