Insurance Scoring Model from Cambridge Mobile Telematics Receives Regulatory Approval from 29 States [PRWeb]
Cambridge Mobile Telematics (CMT) today announced that its telematics-based insurance scoring model has been approved by insurance regulators in 29 states across the US. To accurately estimate driving risk, the scoring model uses contextual driving behaviors such as phone distraction and at-risk speeding, in addition to traditional telematics factors like hard braking. Auto insurers can work with CMT to use this scoring model to price policies in telematics programs, including those that use CMT’s DriveWell platform.
Implementing scoring models based on telematics programs has benefits to drivers, insurers, and society at large. According to the National Association of Insurance Commissioners, linking insurance premiums more closely to actual individual vehicle or fleet performance allows insurers to more accurately price premiums, increasing affordability for lower-risk drivers. Further, a performance-based insurance policy gives these drivers the the ability to control their premium costs by incenting them to reduce miles driven and adopt safer driving habits. While insurers know these programs work, they lack the data and infrastructure to create their own.
“Many insurers don’t have a sufficient volume of telematics data to develop their own model,” said Fred Khoury, Director of Client Partnerships at CMT. Prior to CMT, Fred spent 18 years managing the auto product of top insurers. “Using CMT’s pre-approved model as a starting point streamlines the process for insurers and provides a fast implementation path. We are delighted with the fast approval rates and believe it is a testament to the quality of analysis and the predictive capabilities of the models.”
The scoring models were developed by CMT’s data scientists and actuaries using telematics data from hundreds of thousands of drivers covering billions of miles. Working with insurance regulators to obtain approvals, Iowa became the first state to adopt the insurance scoring model in May of 2018. Since that time, 28 more states have approved the model to date, including Alabama, Arizona, Arkansas, Colorado, Delaware, Illinois, Indiana, Iowa, Louisiana, Maine, Maryland, Michigan, Minnesota, Missouri, Montana, Nebraska, New Jersey, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, South Dakota, Tennessee, West Virginia, Wisconsin, and Wyoming.
“We have carefully analyzed the interaction between driver behavior and real-world crashes,” said Bill Bradley, Principal Data Scientist at CMT. “CMT’s scoring model provides a lift of 22x between the best and worst drivers, predicting which ones based on how they operate a vehicle.”
The “lift” refers to ratio of the likelihood of a crash for the worst 10% of drivers according to the model compared to the best 10% of drivers. The lift chart for the scoring model is shown in the picture below.
With scoring model adaption, CMT has already seen distraction decline across DriveWell users. In addition to the 29 states with regulatory approval, its insurance scoring model is currently with regulators across 21 more states pending approval.
“Our analysis shows that distracted driving is one of the most significant factors that predict crash and claims rates,” added Hari Balakrishnan, CMT’s co-founder and Chief Technology Officer. “With the behavioral incentives available in telematics programs using CMT’s DriveWell platform, phone distraction reduces on average by 35% in 30 days, making roads safer for everyone.”