By on August 29, 2011

Thanks in part to the help of people from TTAC, TrueDelta received a record number of responses to July’s Car Reliability Survey—over 22,300. Updated car reliability stats have been posted to the site for 570 model / model year / powertrain (where warranted) combinations. With partial results for another 464 cars, the total is now over 1,000. These stats include car owner experiences through the end of June 2011, making them over a year ahead of some other sources.

Among 2011s for which we received enough responses, the redesigned Jeep Grand Cherokee is the only one that’s clearly worse than average thanks to common problems with the optional air suspension (also a common problem area in Mercedes SUVs), sunroof rattles, and a transmission shudder. Get one without the air suspension or the sunroof, and the risk of problems goes way down. The new Buick Regal might also be worse than average, but we have only limited data for this model so far.

The 2011 Fiesta improved to “about average” this time around, so there don’t seem to be many new problems with the car once the initial glitches are taken care of. We’ll have initial results for the 2011 Explorer and 2012 Focus the next time around, in November, with a preview for participants in October. Most new or revised models for which we received enough responses are also near the average, including the BMW 5-Series, BMW X5 / X6, Chevrolet Cruze, Ford Edge, Honda Odyssey, Hyundai Sonata, Infiniti M, Kia Sorento, Nissan JUKE, and Volkswagen Jetta.

Three new 2011s clearly had clean starts: the Honda CR-Z, Hyundai Elantra, and Toyota Sienna.

We’ve also updated statistics for the percentage of cars that required no repairs or 3+ repair trips in the past year. These statistics can be more useful than the averages.

We’ll update these stats again in November. The more people participate, the more models we can cover and the more precise these results will be.

To view the updated results:

Car Reliability Survey results

Repair odds stats

Come across something interesting? Have a question? Post it in the comments.

Michael Karesh operates TrueDelta, an online source of automotive reliability and pricing data.

Get the latest TTAC e-Newsletter!

27 Comments on “TrueDelta Updates August Reliability Stats...”


  • avatar
    CJinSD

    The 2011 Buick Regal might be a bit worse than average? 85 trips per 100 cars is considerably worse than average, a reflection that the Regal is more eurotrash Opel than traveling-salesmen Buick.

    • 0 avatar
      mike978

      It is worse than average, considerable is a subjective term like bit. Shows a predilection rather than fact.

      I am surprised by the Regal, as I am for the Fiesta when it first came out. Both of these cars have/had been on sale in Europe for a couple of years before introduction here and as such the niggles should have been worked out. Certainly compared to an all new car like the Sienna which is US specific.

      Also Euro-trash? It may be below average in reliability (depending what goes wrong – could be minor) but still has a good rep for interior quality, driving dynamics etc.

      • 0 avatar
        CJinSD

        I think anything over 54/100 would constitute the start of worse than average for cars no more than a year old, based on looking around at a few model lines. 85/100 is closer to the worst performance than it is to average. Compare it to the 2011 Toyota Camry’s score: 0/100 failures with a sample size of 32. I wasn’t surprised by the Regal’s or the Fiesta’s performance. I’d be more suprised if the 2011 Sienna hadn’t produced an excellent 22/100 result.

      • 0 avatar
        mike978

        Agreed 85 is not great, no argument from me. You were the one quizzing Mk to change bit to considerable. Both subjective terms.

      • 0 avatar
        Pch101

        I am surprised by the Regal, as I am for the Fiesta when it first came out. Both of these cars have/had been on sale in Europe for a couple of years before introduction here and as such the niggles should have been worked out

        The Insignia gets a low ranking in the UK JD Power survey. http://www.whatcar.com/car-news/jd-power-survey-2011/other-family-cars—part-2/257096 As the Whatcar? review describes it:

        Vauxhall describes the Insignia as ‘sculptural artistry meets German precision’ – a fancy way of saying that it’s stylish and well put together using quality materials. Vauxhall’s reliability record is no better than average, though, and owners rated the Insignia’s quality and reliability as below average in the 2011 JD Power survey.

        So I’m not surprised. Opel and Vauxhall aren’t exactly the most respected brands in Europe. Then again, the US drivetrain options aren’t quite the same (and it wouldn’t surprise me if the trim levels varied as well), so the American versions don’t compare directly to most of the European versions.

    • 0 avatar
      marjanmm

      “Eurotrash” because statistically you are likely to have one repair trip a year? Do you really hate going to your dealer so much that all the other aspects of the car are insignificant compared to less than one repair trip a year?

      • 0 avatar
        mike978

        Exactly – we are “fighting” over 0.5 visits a year (average) vs. 0.8 visits a year with the first year Regal. 0.3 visits per year extra, so that means 1 out of 3 Regal owners will have that extra visit and the other two will have the average number. Or is my math out? Either way if the fault is not a major problem (like my Subaru foam issue) and was fixed at a routine service then I would consider the other aspects of the car and not be hung up about 0.3 (or even 0.6) visits a year. As MK said earlier the 3+ and zero fault data is probably more useful in the real world.

      • 0 avatar

        This is why I like to provide the actual numbers, which no one else does in their publicly-available reliability stats. Some people want to avoid a repair at all costs. Others are willing to deal with an extra repair every couple of years in return for advantages in other areas. I think the numbers make it much easier to make such tradeoffs.

        With a small sample size like we have for the Regal the difference between 0.5 and 0.8 could be due to sampling error. I personally put more stake in a result based on a small sample size if the result is either under 25 or over 90.

      • 0 avatar
        CJinSD

        mike and marjanmm,

        I hope you realize that you’re arguing that this entire endeavor is a waste of time. The Regal’s score was one of the worst of the entire study. If it isn’t worth identifying the worst performers and avoiding them, what use is it charting flaws at all? Might as well buy whatever you want and then blame fate when your choice has predictable problems that your crummy dealer can’t fix. 85/100 does not mean that the average was less than one visit to a service bay, it only means that there was an average of less than one initial visit to a service bay for repair. Frustrating follow up repair attempts for the same problem are freebies. Which reminds me, I need to go return my bicycle to the Trek dealer today for their third attempt at fixing issues that began in the spring and caused my bike to be in the shop for close to 3 months so far. I’m glad I bought a high quality car.

      • 0 avatar
        geeber

        The problem with the Regal is that people who want this type of car generally do not set foot inside a Buick dealer, and people who do visit Buick dealers do not want this type of car. The number of problems uncovered by TrueDelta’s survey is not the car’s main challenge.

      • 0 avatar

        CJ,

        My own feeling is that:

        1. The differences between cars are often much less meaningful than most people expect. I expected this to be a key finding before I posted my first results five years ago. My own sister once wondered if she was reading the results wrong, because all of the cars she was considering were about the same. My response: you’re reading them correctly, they are about the same.

        2. There are still some troublesome cars, especially early in a model’s run.

        3. Some people will care about differences that others don’t find meaningful.

        I put the numbers out there, and each person can figure out how meaningful the differences are based on their own priorities. What I don’t do: summarize everything into opaque ratings or have “cars to avoid” or “top picks” lists (as much as I’ve been told I should).

        If you like a car so much that you’re willing to deal with a few repairs, then by all means buy it. You’ll at least know what you’re getting into.

        If, on the other hand, you want to avoid repairs at all costs, at least realize that many alternatives aren’t much less reliable, and that we’re generally dealing with the odds of a single repair in a given year, not the number of repairs per year.

      • 0 avatar
        Demetri

        Depends on what the repair is. If there’s a new rattle behind the dash every year, who cares, I won’t even have something like that fixed. If it strands you on the road once a year, it’s a piece of garbage.

      • 0 avatar
        marjanmm

        @Michael,

        The numbers you publish are very important as they dispel a popular conviction that buying a “below average” is an absolute disaster. They clearly show that “below average” may mean just one repair trip a year in average. Of course, odds of a lemon will still be much greater for a “bellow average” compared to say Toyota however in absolute terms those odds are very small.

        When JD power publishes their relative lists everyone seems to assume all models of “below average” makes will be in the shops three times a month.

    • 0 avatar

      The 2012 could be much better, for two reasons:

      1. Another year to work bugs out.

      2. Regals will now come from Oshawa, historically GM’s highest quality assembly plant, instead of Germany. If a car is well-engineered then it doesn’t much matter where the pieces are put together. But there are exceptions.

  • avatar

    The sample size for the Regal wasn’t up to the minimum yet, so the 85 could be inaccurate. On the site this stat is asterisked and visible only to members. I’ll remove the “bit,” though, as the “might” covers this possibility.

    Note: I added this comment fairly early, but it didn’t immediately follow cj’s like I intended (computer lost connection when I first submitted it).

    • 0 avatar
      CJinSD

      I agree. The small sample makes the result questionable, but there is nothing to suggest that the Regal might only be a bit worse than average. If the results prove to be indicative, it is another turkey from GM.

      • 0 avatar
        mike978

        A turkey? No inbuilt prediliction there then.
        85 is not great and it should be better, but lets consider the following – small sample size, possibly very minor items that go “wrong” and can easily be fixed at a routine service and first year blues. The Fiesta is a good case that the second year can (not always) bring good improvements.

        I notice you ignore the Cruze, which first year in the US has a reasonable score. But that doesn`t fit into your narrative of GM building only turkeys. The 2011 Toyota Rav4 was worse than the Curze (by 1 – 52 vs 51, which is trivial to me) and the Rav4 has been around for years so they should know how to make it. I bet that doesn`t fit your “narrative”!

      • 0 avatar

        For a GM car in its first model year that is doing very well in these results, check out the 2010 Camaro. They had a minor problem with loose spoiler bolts early on, but otherwise there have been very few reported repairs. GM did something right on this one.

  • avatar
    threeer

    Michael, any initial feedback on the 500? The car appeals to me in a great way (especially the upcoming Abarth) but then I think about it being built at a Chrysler plant…in Mexico…and then I start backpeddling a little…

    • 0 avatar

      I’d love to know how the 500 is doing myself, but not yet. There are a dozen signed up so far, so we might have a rough indication in November. More will be needed for reasonably precise results. If anyone knows a 500 owner, please encourage them to join.

    • 0 avatar
      windswords

      The Toluca Mexico plant is one of the most modern plants Chrysler has, capable of fully flexible manufacturing. It builds not only the 500 but the Journey CUV. Before that it built the PT Cruiser, which despite its shortcomings was rated as reliable by CR. The work force has been nothing but cooperative with management and have been well treated by the company AFAIK.

  • avatar
    mike978

    I have a question around what is covered in the stats (and I am a contributor to truedelta data). Is the trips per year number when someone has a fault (of any severity) and takes in into the dealer to be fixed? I ask because I did record a while back one fault for my Subaru Legacy which had the foam coming out from a rear door molding. Obviously not very serious and could wait until a periodic service appointment. In this case it would raise the number of trips per year stat for the Legacy and yet be of little importance to any future buyer since most people think of reliability in terms of “will I be stranded?”, does the transmission stop working etc.

    My (long) query above may help people use the trudelta data in an appropriate manner since it seems that cars nowadays are reliable in the old sense of the word (i.e. not being stranded).

  • avatar

    The current statistics equally count repair trips in which a problem of any severity was successfully fixed. This is a very robust metric, which is very helpful when working with small sample sizes.

    Nearly a year ago I added a more detailed severity question to the survey. Once we have a year of data with this question (i.e. next time) I’m going to attempt to look at severe problems alone, as measured by how essential were they to the operation of the car. Or maybe I’ll still include all problems, but weight them. The danger here is that the more rare something is–and breakdowns that leave people stranded are rare for fairly new cars–the larger the sample size you need to precisely measure it. If you have only one or two breakdowns reported, these could be flukes. So I won’t be able to provide these stats for nearly as many models.

    The components that most often force a tow are the alternator, starter, and battery. We don’t even count the last as a battery’s lifespan depends very much on how well it is treated. They’re also fairly cheap to replace. Other parts, like radiators, water pumps, fuel pumps, and wheel bearings, tend to provide enough warning that people get to the shop before the car stops running (unless the driver ignores the signals).

    And outright engine or transmission failures? They’re exceedingly rare. Even when these are replaced on a car with under 120k miles on it it’s usually because of high oil consumption or difficulty shifting, not because of a total failure that requires a tow.

    • 0 avatar
      mike978

      Michael, thanks for the explanation. I was not suggesting that the stats be changed since they seem robust. My point was for those who use the data (for good i.e deciding what car to buy and for bad i.e. to criticise makes they don`t like) it would be good to know if severity was included at all.

      My Subaru example is a great one, it meant the Legacy didn`t get a 0/100 like the 2011 Camry CJ loves but if all that stands between a 0/100 and the Subaru’s low score is some foam then it is insignificant.

      • 0 avatar

        The descriptions for all reported repairs are posted to the site, so it’s possible to see what’s behind the numbers. For the new Subaru the most common problem has been steering wheel vibrations. Might seem minor, but it was hard to fix in some cases and Subaru bought more than a few cars back.

        The 2011 Camry is in the last year of its cycle. “Last year” cars often do very well. Two likely reasons: they’ve have years to work the bugs out, and with the new one coming sales tend to be soft, so they run the line slower.

  • avatar
    geeber

    Thanks for posting these results. We participate in the survey with our two cars – a 2003 Accord EX and 2005 Focus SE – and both of our particular cars appear to be more reliable than the survey results would indicate. Our mileage is considerably higher than the average mileage for both cars, too.

    • 0 avatar

      Take a look at the “odds” stats: even with the least reliable models there are quite a few owners who’ve reported no repairs. And the two you own are far from the worst, even if 2003 is the least reliable year for a fairly recent Accord.


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

  • Re: Toyota May Kill V6 Camry

    Sky_Render - “Fans of the Toyota Camry”? I assume the demographic you’re referring to has a large overlap with AARP membership.
  • Re: Toyota May Kill V6 Camry

    NormSV650 - Superchargers has 100,000 mile bearing coupler and oil replacement, turbos do not. When properly maintained with synthetic motor oil turbo charters can last around...
  • Re: Toyota May Kill V6 Camry

    28-Cars-Later - 1. $10K+ premium for a Lex V6? No thanks. 2. Gives Toyota dealers a compelling loaded product at a lower price point than Avalon and allows them to compete with...
  • Re: EPA Mandates Real-World Testing For All Automakers

    SCE to AUX - “EPA Mandates Real-World Testing For All Automakers” Good. Then the Optima Hybrid could be rated higher than today’s...
  • Re: Don’t Drink the Google Kool-aid: Autonomy’s Limits in Three Sips

    Vulpine - Mr. Smitka offers some very valid points, but his article is launched off of a poor premise. I, for one, don’t...
  • Re: Piston Slap: The Express’ New Mission?

    rpol35 - Ouch! That’s a heck of a very pertinent story, you even got the Panther reference included – what a bad break.
  • Re: 2015 Opel Corsa Revealed

    geozinger - “No, what I see here is brand hate over product news.” Nothing new here folks. I’ve been following this site for years and if you can’t say...
  • Re: Junkyard Find: 1986 Buick Somerset

    sirwired - Well, clearly I know nothing about design, because I think for an ’86 car, that “pod” radio control looks pretty neat. Could use some large...
  • Re: Cadillac ATS-V Coupe Spotted At The Ring

    SomeGuy - You mean LT4 right? Either way, whatever engine they put in there will be vastly more reliable than a BMW motor (which is their competitor).
  • Re: Toyota May Kill V6 Camry

    Lie2me - Were they new or just a cumulative total, they’ve been around a few years. I live 4 blocks from a Ford/Lincoln dealer and I can’t remember the last time I saw...

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

  • Brendan McAleer, Canada
  • Marcelo De Vasconcellos, Brazil
  • Matthias Gasnier, Australia
  • Tycho de Feyter, China
  • W. Christian 'Mental' Ward, Abu Dhabi
  • Mark Stevenson, Canada
  • Faisal Ali Khan, India