J.D. Power withholds repair rate data from car buyers, but does provide ratings in the form of dots. In addition to the normal problems with dot ratings–they’re quite vague–J.D.’s have an additional, serious weakness: they only go down to two.

These days, most cars are at least fairly reliable. There’s not a huge difference between the most reliable cars and those around the average, maybe a third of a repair trip per year. So while J.D. Power likes to focus on the “winners,” those models that had the fewest problems reported in its surveys, it doesn’t really make sense to buy one of these just because of reliability. Plenty of competitors are nearly as reliable.

Instead, car buyers should be most concerned about not buying those models that are least reliable. There aren’t many of these, but they are out there. Unfortunately, with J.D.’s dots, you cannot identify them. Everything that is below average gets a two-star rating, marked “the rest.” This lumps the least reliable models together with others that scored near the average. And you cannot tell them apart.

Why does J.D. Power do this? Probably to avoid upsetting a manufacturer that pays them millions in fees each year. But in the process they avoid telling car buyers what they most need to know. It’s far more important to know which models are least reliable than which models are most reliable.