NHTSA Data Dive: 95 Cars Ranked In Rate Of Unintended Acceleration Complaints

Posted in Uncategorized by wolferadio11 on February 28, 2010

February 27, 2010

[Update ans Disclaimer: As noted below, this spreadsheet will be updated as we accumulate more detailed sales stats for the years (’05-’10) covered. That will very likely change the rankings somewhat. The Lincoln may actually be #2. But this is not about which car is #1 or #2; it’s about finding patterns in certain makes, and within makes. It’s an attempt to see if these statistics can shed light on a complex and opaque issue. As an example, why the Toyota Yaris is so low in reported incidents. It’s more about these patterns and discrepancies, than about singling out the car with the highest rate; so please don’t take the current exact rankings as the final word. It’s a work in progress. The fact that the complaints are not tabulated by individual MY also limits this substantially, as running changes in a given car during the five year period will change things significantly. So this data dive is fundamentally flawed; take it as such. But nevertheless, it’s still a huge step over the raw data that Edmunds put out, which doesn’t begin to account for the number of any given cars sold.]

Numbers and statistics are largely useless without context. took a first good step in going through NHTSA’s data base and reporting the number of UA events reported per make, brand and vehicle. But what was obviously missing was the correlation to the number of cars on the road in relation to those numbers. We’ve taken the next (tedious) step, and the results are most interesting indeed. They’re certainly not completely conclusive, but we’re not finished yet. The full list of 95 cars follows, as well as our methodology, a stab at some analysis, and more questions to still be answered.

First, our methodology. Edmund’s NHTSA data was for model year 2005 – 2010 cars, to date. Lacking easy complete sales data, I’ve taken the 2008 MY sales for these vehicles, and multiplied by five to arrive at a working number. If detailed sales numbers are to be found, I will update them. But I doubt it would change the numbers significantly.

Obviously, the cars that had lower numbers of events reported are going to be statistically less reliable. And unless we go back to the NHTSA and mine the original data, if that level is available, we don’t know what type of UA event was reported. Was it a likely wrong pedal application, which typically happens at low speeds and often in parking lots (hopefully not near a cliff)? Did it come on after merging into a freeway? Having that level of detail would allow us to make further assumptions, especially if certain cars had higher incidents of a particular type of UA event.

The other thing is consider whether the given vehicle had brake override or not. On the one hand, it’s interesting to note that no European brands were reported, and they pretty much all have brake override. But then so does Chrysler, and several of their cars, especially certain Jeep models, had fairly high rates of reported UA.

There are other factors to consider, like the demographics and use of the specific vehicles. I suspect strongly that the Lincoln TC, the Grand Marquis and the Crown Vic all have the same mechanical and electronic systems. Yet the TC came in #1, the Marquis #9, but the Crown Vic at #32. Is the high proportion of CV use in taxi and police service an issue here? And what are the TC and Marquis owners’ median age?

So many questions and so few answers: Ford, which does not have brake override, had some models fairly high on the list, but the popular Fusion was very low with a rate of .005, and its sibling Milan with a .006. The fact that these virtually identical cars came in so closely gives support to these statistics being reasonably accurate. Other examples are GMC Sierra and Chevrolet Silverado at # 95 and #96. And the Malibu and similar G6 are at #84 and #88.  But then when it gets down to such few reported events, randomness increasingly becomes a factor.

GM’s low UA rate is undeniable: their highest car is the Cadillac DTS, at #40. It also has a notoriously high median age of its owners. The mechanically related Lucerne and Impala are not far apart, with the Lucerne being somewhat higher and undoubtedly having an older median age of its owners. I’m conjecturing here, but the numbers tend to bear it out.

Clearly, Toyota vehicles are skewing to the upper end of the range, although the Corolla/Matrix is down at #38. And the Yaris is very low on the list at #79. What exactly puts the Lexus ES 350 at such a high rate is certainly worth exploring, especially since it was involved in the two must publicized UA events. Its general mechanical similarity to the Camry is well known; without breaking out the types of events the ES 350/330 has been involved with, its difficult to say. But additional information would be indicative, since the Camry is not all that high on the list at #11. Are Lexus floor mats thicker and deeper than the Camry’s? Are the electronics substantially different? Are Camrys imported from Japan involved at higher rates? ( all Lexus ES models are also imported).

The questions go on and on. I will continue the quest of turning up statistics that shed further light on this issue. And your comments , analysis and questions will be most appreciated and helpful. One thing: keep in mind that these are only those cars that had UA incidents reported to the NHTSA (actually, I removed a few out-of production cars from the list). Missing of course are all those cars with no reported incidents, including whole brands, like Hyundai, Kia, Mazda, Subaru, Volvo, and all the Europeans (except Saab). Did I miss someone significant? Full list follows:

 Reposted from

3 Responses

Subscribe to comments with RSS.

  1. Justin said, on March 4, 2010 at 11:32 AM

    wow. So it took a few people to die before this gets corrected? I’m sure those people’s families are pissed beyong belief seeing this. Poor poor Toyota and the hundreds of millions this is costing them. How do the guys that make these decisions sleep at night?

  2. P.E. Gould said, on June 1, 2011 at 6:44 PM

    Add another auto to the list – 2010 Infiniti G37 Sedan – I experienced two instances of unintended acceleration – the Sales Manager told me that with despite all the brouhaha that in actuality there had only been 4 actual cases of unintended acceleration – and – in those instances it had been determined that the driver was at fault.

    The car is now at the dealers – when it’s returned, via tow truck – no one in their right mind who values life would be willing to drive it. Future of the car? I plan to have the car crushed – life is too valuable to take a chance that the car will “get” me the next time it decides to take off and fly.

    NHTSA was notified on 5/24/11. I have asked my insurance company (GEICO) and the organization that helped finance the car (Pentagon Fededal C.U.) to work with me – to try to recoup something from this mess – any other suggestions as to who else I should go to?

  3. FRE said, on April 27, 2012 at 10:12 PM

    This report is woefully inadequate because it fails to indicate the type of transmission that cars have. It’s my guess than in most case, and perhaps in every single case, the car had an automatic transmission. It’s difficult to see how this could happen with a manual transmission.

    There should be follow-up reports which indicate which type of transmission cars involved in unintended acceleration had.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: