We now display a rating representing the confidence that an IP address is abusive, based on its reports. This rating is available for all recent and newly reported addresses. This attribute will not be included in the current check API endpoint, but rather the next revision.
The formula is based on various factors, such as the number of reports from registered users, from anonymous users, the recency of those reports, and if applicable, the trustworthiness of those registered users. In the coming weeks we will be continually tuning this formula. Let us know if any granted ratings are strikingly inaccurate.