A Few Words About Data
New car shoppers looking for quick verdict on a particular model have a tough go of it. Most professional car critics don’t quantify their praise (or criticisms), and those who do tend to employ a bewildering variety of grading systems. Even among the published critics we cite on MotorMouths, this grab-bag of methodologies makes it rough for shoppers to easily grasp what the experts think of a particular car.
So, how to transform this mishmash of opinion into something more easily understood? Glad you asked.
Here’s what we do. Teams of editors and analysts carefully read thousands of published reviews from leading automotive critics, and assign every opinion expressed by the critic to six specific editorial categories. All statements about the vehicle, positive or negative, are weighted and tallied, and that value is then run through one or more of 95 distinct proprietary normalizing algorithms, which yields an appropriately-weighted numerical score for the specific vehicle, ranging from 0 to 100. Basically, we’re converting a critic’s review into an actual grade, and one which accurately and precisely reflects the critic’s opinion about the car. These individual grades are then averaged together utilizing class-appropriate weighted averages, to yield an Overall Score for a vehicle. At last count, we’d analyzed 70,000 published car reviews by more than 600 professional automotive critics, and our methodology and algorithms have grown stronger and more precise with each review.
By the way, if a critic has already assigned their own numerical score to a vehicle, we adjust that score using our criteria, so that all scores can be fairly ranked against each other, no matter their source. Of course, if a critic disagrees with the score we’ve derived from their review, they can tweak it by contacting us. (Since MotorMouths.com founding, many tens of thousands of reviews ago, we’ve only had to adjust a single score. So we think we have a pretty good sense of how individual critics judge vehicles.)