Least reliable used cars: 1997-2012
 

looks like a contentious survey judging by some of the comments. As they say, cars costing the most to repair is a different thing to cars that are inherently unreliable and breaking down a lot. Looks like this is more about the cost of parts to repair rather than just reliability issues on their own. I also think that the number of cars out there should be factored in and the cost spread out over them all - if you have a complex limited / low production vehicle it will always skew the figures vs a much higher production run :rolleyes:

Interesting that they've lumped both C5 and C6 RS6 into one group - surely an error as they're completely different cars?!

Any analysis of statistics is only as good as the data and its statistical significance (eg sample size) and the quality of people doing the number crunching, to use some common sense and not go beyond the level of detail the data supports.....

I've done similar data analysis some years ago using lease car data (much bigger sample size) and you do generally find the usual suspects in the list, but I would've ruled out anything like the M5/RS6 as being too small a sample compared to the multiple thousands of rep-mobile vehicles to compare them to