By on August 16, 2010

Thanks in part to help from TTAC readers, TrueDelta received a record number of responses to last month’s Car Reliability Survey—nearly 18,000. Updated car reliability stats have been posted to the site for 458 cars, up from 404 three month ago. There are partial results for another 351.

These stats cover through the end of June. Other sources of car reliability information will not cover the most recent months until the summer or even fall of next year.

Among the 2010s we have full results for, few are “worse than average.” Exceptions include the Ford Taurus, Hyundai Genesis Coupe, Jaguar XF, and VW diesels.

The Taurus has had a common problem with chrome trim peeling off the tail lights. The VW diesels had a problem with O2 sensors that seems to have been resolved–the updated result is a big step in the right direction. So both of these are expected to improve to about average in future updates.

The Genesis Coupe continues to worsen, and is not expected to improve before the second model year. The Genesis sedan has improved in its second model year.

The Jaguar XF is already in its second model year, without evident improvement after some early issues with the 2009 were resolved.

We don’t yet have enough data on the 2011 Hyundai Sonata and Kia Sorento. The data we do have suggest that both are about average, with no obvious common problems. Full results for these with the next update, in November.

Despite the recent media circus, Toyotas continue to be most consistently better than average.

Additional participants are always helpful. The more owners participate, the better the information we can provide to everyone.

To view the updated results:

Car Reliability Survey results

Get the latest TTAC e-Newsletter!

25 Comments on “TrueDelta Updates Reliability Survey...”


  • avatar
    JMII

    Bummer about the Genesis Coupe, what has been going wrong with them?

  • avatar
    don1967

    Nice work, Michael, in particular the level of detail sought on problem areas and fuel consumption. I am glad to participate in your surveys, and hope others will help you build your database.

  • avatar
    ajla

    Not surprised to read about the problems on the Genesis Coupe.

    That car has so many growing pains it makes the very first run of the Zeta Camaro look like a Corolla by comparison.

    Hyundai definitely needs to do something about the car’s quality, but with sales so low, I wonder if they’ll just wait for a refresh.

  • avatar
    zbnutcase

    Im sorry, but chrome peeling off tailights is hardly a reliability issue!

    • 0 avatar
      SecretAznMan

      True. That’s a quality issue. For an example of a reliability issue, see the VW forums regarding TDI fuel pump failures costing $10,000 not covered by the warranty. I know forums have a higher % of problems than in real life, but there were enough reported issues to see it was not just a collection of freak incidences. I was so close to buying a TDI Sportwagen…

    • 0 avatar

      “Reliability” is actually more appropriate than “quality.” It’s not the perfect word, but a better word does not exist, at least not in English.

      With “quality” you get into how nice the interior materials are, NVH, and so forth.

      For issues that keep the car from running I’d use “dependability.”

    • 0 avatar
      NulloModo

      Minor trim issues seem to be the only real sticking point regarding the D3/D4 platform vehicles. To Ford’s credit, these are usually resolved quickly after they are reported, and they are easily fixed with a single dealership visit free of cost to the customer.

    • 0 avatar
      gimmeamanual

      Nullo, I don’t mean to pick nits, but unless you’re compensating me for my time and gas money getting there, that return trip is free only in that I’m not giving you money.

  • avatar
    obbop

    It’s when the rust begins peeling off that the cabin becomes unbearably cold and drafty in winter-time.

  • avatar
    SomeDude

    “Despite the recent media circus, Toyotas continue to be most consistently better than average…”

    Toyotas continue to be **reported** as being better than average. That’s the power of perception for you.

    • 0 avatar

      We ask whether or not the car has been in the shop for a repair. It’s a yes or no question specifically designed to eliminate the role of perception.

      Perception can play a big role in Consumer Reports’ survey because they ask people to only report problems “you considered serious.” This opens the door wide for under-reporting in a way our survey does not.

    • 0 avatar
      NulloModo

      Michael –

      Digging into your site a bit, there are a fair bit of reports that mention nothing more than routine maintenance. Is your software set up such that routine maintenance visits are segregated from repair visits, or is it fair to say that the data is corrupted to some degree from those who can’t tell the difference from routine maintenance and actual repairs?

      Adding to the point above, what about people who buy cars second hand, and then have to handle things that might technically be ‘repairs’ but would have been covered by routine maintenance by previous owners? If someone buys a vehicle with 80K on the odometer that has never had the timing belt replaced, and then has a timing belt failure and reports it on your site as such, should that be considered a repair and register as a hit against that vehicles reliability? I’d say no, as it should fall under routine maintenance that was simply no accomplished when it should have been, but how do you account for and correct such data?

    • 0 avatar
      th009

      It’s definitely set up to separate maintenance from repairs (I have both our cars registered on the site). It appears some people can’t tell the difference, though!

      That should be fairly even across makes/models, unless some cars generally attract buyers with a lower IQ. : )

    • 0 avatar

      The displayed repair histories actually combine responses from two surveys. On the main reliability survey any maintenance items that people report are coded out. But anything done to the car can be reported using the second survey, if people want to post it.

      The stats only use responses provided through the main survey.

      The point of the second survey is so people can post excluded items without having them affect the results.

      Both timing belts and timing chains and their tensioners–and even water pumps–are excluded after 70,000 miles, to avoid giving cars with timing belts a misleading advantage.

    • 0 avatar
      NulloModo

      Michael –

      Thank you for the info, it looks like you are actually very forward thinking in your process, another detail that puts you head and shoulders above CR.

      If you are accepting suggestions, I’d also recommend some sort of consideration for other items that should be replaced under regular maintenance, such as spark plugs, as well as weighting things like transmission failures on higher mileage vehicles based on any kind of data you can come up with based on how often people actually follow the owner’s manual on when to change the fluid. I’d also prefer to see repair data completely segregated from routine maintenance. While it may make some sens to have some sort of rating showing vehicles that require more intensive, and expensive, routine maintenance than others, allowing responders to choose to include that as potential repair data just lets those pissed off about the normal cost of operating their vehicle skew the results.

      It would also be nice to see some sort of visual difference in responses on the site for whether a particular owner reported an issue as regular maintenance or as a repair. Weighting repairs for major issues, such as engine/transmission/major electrical vs minor issues such as trim/lights/minor electrical would also help really put your system ahead of the game.

      On a final note, your login system leaves a bit to be desired. I always have to go through my old e-mails to look up my member number to log in to your site, I just can’t seem to commit a random numerical string to memory the same way I can the traditional username/password that other sites require. Converting your login system to allow for ease of use would probably help with your participation as well.

    • 0 avatar

      I’m always open to suggestions.

      I was told today that a new login system is ready for testing.

      If it passes, it will be on the site before the end of the month. I wanted it done a year ago, but have had trouble getting things done that I don’t do myself. This has been a huge source of frustration.

      Spark plugs are among the items always excluded as wear items.

      The repair data is totally segregated on the database, just not in the presentation on the site. As you note, some people want to vent. The current system lets people vent without distorting the stats. This is how we solve the problem you mention.

      Weighting repairs is tricky. It sounds great in theory, but in practice any weights would be arbitrary, and different people would want different weights.

      On top of this, you’d need much larger sample sizes. I do internally have special codes for engine and transmission replacements or rebuilds, but haven’t been doing anything with these yet. They’re so rare that you’d need a huge sample size to have meaningfully accurate stats for such repairs.

      Perhaps in the future, when we have much larger sample sizes, I could give people the option to weight such repairs, and provide their own multiplyer.

    • 0 avatar
      golden2husky

      I concur with the login comment of NM. I, too have to dig through old emails to log onto the site…still must say that I am impressed with the value of the site. I have had two coworkers – one who is a CR dot worshiping slut – sign on…

  • avatar
    srclontz

    I think TrueDelta is a great resource and I encourage everyone I know to join. Repair trips per year per 100 cars helps greatly to put reliability in perspective, far more than stars or circles.

  • avatar
    gimmeamanual

    Michael, instead of coming up with your own weighting system, why not just use the weighting points out of the automotive FMEA manual?

    • 0 avatar

      Must say I’ve never heard of the automotive FMEA manual. What is it, how finely to they classify different repairs, and how do they assign weights? Why do they assign weights?

      A final issue with weights is that some owners don’t know what was done to repair the car. They only know the car had a problem when they took it in, and no longer had it when they picked it up. Down the road, when we have larger sample sizes, we might simply be able to exclude such cases. Right now we don’t have that luxury.

  • avatar
    Lokki

    I’m a member of TrueDelta, and I find it both very interesting and useful, particularly as the database continues to grow. I’ll be in the market for a car next summer, and I’ve started to sift through the potential entrants for the prize (sic) of my business.

    For a while, I’d been sort of flirting with the idea of a Hyundai Genesis but TrueDelta shows that -while improving- they’re just haven’t arrived yet.

    I’ve also been surprised that some of the Hondas really aren’t doing as well as I’d expected.

    Finally, looking at the MINI’s has helped me answer the age old question – what do you get when you cross an English car with a German car? Answer – you get an Italian car. That is, a car with plenty of quirks and problems that is so much fun to drive, you forgive it.


Back to TopLeave a Reply

You must be logged in to post a comment.

Subscribe without commenting

Recent Comments

New Car Research

Get a Free Dealer Quote

Staff

  • Authors

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