I'm glad you posted this. The thread is mostly full of anecdotal claims, which makes sense since few people clock enough time in enough cars in a lifetime to claim statistical significance. However if we're talking about maintenance costs, it's really about average value and standard deviation.