Late Bet Score (LBS)

The Late Bet Score API OpenAPI specification is available for download here. Explore the API endpoints here.

The Late Bet Score (LBS) API is part of Sportradar’s Insight Tech Services and is designed to help operators detect accounts that may be exploiting latency windows during live betting. By assigning a probability score to each profiled account, it enables precise control over bet validation delays and supports efforts to mitigate financial risk, ensure fairness, and identify suspicious behaviour.

This model plays a central role in helping operators address late-bet abuse, a form of exploitative activity where customers place bets just before bet stop, often leveraging a momentary advantage in market timing.

What it does

The Late Bet Score represents the probability that an account is engaging in late betting abuse. Late bets are defined as bets that show a sudden increase in ticket value within a few seconds of being placed, usually in response to an odds change.

The model evaluates a customer’s historical betting data – particularly live singles – and analyses a range of behavioural signals such as bet timing, market volatility, rejection rates, and profit/loss volatility. Based on these features, the model returns a score between 0.00 and 1.00, where higher values indicate a stronger likelihood of late betting.

Operators can use this score to apply targeted live time delays, escalate accounts for review, or prevent delay skipping during high-risk events.

When it’s triggered

The model is run once daily, profiling accounts that placed at least one live single bet the previous day. LBS is not computed for inactive accounts or accounts without relevant betting activity. Once calculated, the scores are made accessible via the API and can be queried at any time.

Input requirements

Each request must specify the following identifiers:

• bookmaker – the assigned Insight Tech or MTS bookmaker ID

• sub-bookmaker – the sub-entity of the bookmaker (use * to retrieve all sub-entities)

• customer - customer's unique ID

Some endpoints also support range filters (lower, higher) or pagination using startKey. A full breakdown of all accepted parameters is available in the Endpoints section.

Output schema

A typical response includes the list of items, and each object under responseItem contains:

• srcId: compound source identifier for internal tracking

• lbs: A decimal value between 0.00 and 1.00 representing the Late Bet Score

• id: entity identifier

Interpretation guidance

Late Bet Scores provide directional insights rather than deterministic classifications. Here's how to interpret them:

• High scores (e.g., ≥ 0.90) suggest a strong likelihood of late-bet abuse. These accounts are typically candidates for enforced validation delay.

• Moderate scores (e.g., 0.70 to 0.89) indicate some level of concerning behaviour and may warrant monitoring or conditional delay.

• Low scores (e.g., < 0.70) reflect normal, non-suspicious betting patterns.

• Missing scores generally mean the account is new or inactive.

Operators can set their own thresholds for what constitutes a risky LBS value, and these thresholds can also be used in automation logic for delay enforcement or Live Time Delay Offset integration.