Welcome! Log In Create A New Profile

Thanks for the update.thumbs up *NM*


Expect the best, and accept no substitute.

Products for your Boxster, Cayman and Carrera.
Good to see that someone has picked up where ppbb left off.

A few years ago I started conducting a car reliability survey, and a thread on ppbb got many Boxster and Cayman owners involved. The advantages of this survey are that we post the repair frequencies as numbers, and not just dots, and update these results promptly every three months. Also, we post the descriptions of all reported repairs, so it's possible to see what's behind the numbers.

We recently updated reliability stats for the Boxster and Cayman that include owner experiences through September 30, 2010. Other sources of car reliability information won't cover the months since April until the summer or even fall of next year.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 59, about average

2007: 61, about average, small sample size

2006: 53, better than average

2005: 45, better than average

2004: 71, about average, small sample size

2003: 55, better than average

2002: 92, about average

2001: 98, better than average

2000: 69, better than average

1999: 156, worse than average, small sample size

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

I'd like to report specifically on the IMS failure rate, but we will need more participants to do this. A few more failures have been reported recently, but still about two to three per 100 cars in the past year.

Thank you, once again, to everyone who has been helping. We'll have further updates, possibly including additional model years depending on the number of participants, in February and May.

To see how competitors compare, and to sign up to help:

[www.truedelta.com]



Edited 1 time(s). Last edit at 12/11/2010 12:08PM by mkaresh. (view changes)
Updated results around the end of this month.

We have some new pages that display all reported repairs by problem area:

[www.truedelta.com]

From this page it's possible to select a different model year or specify a problem area.
Since it is a self-selecting survey on a subset of internet boards, how do you adjust your statistics to reflect a very evident selection bias?

1998 986 Turbo-Look Cab 172,000 Miles Dilithium Crystal Supercharger
Sorry, just saw this.

This is largely handled through the methodology. The survey is continuous, not a one-time event, and collects data going forward rather than asking about repairs that occurred before someone joined.
We have updated reliability stats for the Boxster and Cayman that include owner experiences through December 31, 2010. Other sources of car reliability information won't cover the months since last April until the summer or even fall of this year.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 39, better than average

2007: 59, about average

2006: 49, better than average

2005: 54, better than average

2004: 88, about average, small sample size

2003: 58, better than average

2002: 124, about average

2001: 104, about average

2000: 90, better than average

1999: 106, about average, small sample size

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986). Some of these failures might have occurred before the car was enrolled, in which case they would not be included in the above stats. So this count might err on the high side. The total sample size for 1999-2005 is 220 cars. No IMS failures have been reported for cars 2006 and newer.

Thank you, once again, to everyone who has been helping. We'll have further updates, possibly including additional model years depending on the number of participants, in May and August.

To see how competitors compare, and to sign up to help:

Porsche Boxster reliability ratings and comparisons
Thanks for posting Michael. I appreciate your efforts at TrueDelta and getting more of us involved. cool smiley
We have updated our car reliability stats for the Boxster and Cayman that include owner experiences through March 31, 2011.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 41, better than average

2007: 65, about average

2006: 54, better than average

2005: 73, about average

2004: 62, better than average, small sample size

2003: 65, better than average

2002: 112, about average

2001: 83, better than average

2000: 84, better than average

1999: 97, better than average, small sample size

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986).

No IMS failures have been reported for cars 2006 and newer. None have been reported for any year since the beginning of 2011, with over 220 owners of 1999-2005 cars participating.

Rear main seal and air/oil separator repairs are far more common.

Thank you, once again, to everyone who has been helping. We'll have further updates in August and November. With more participants we could provide more precise information and cover all model years.

To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
Surely I misunderstand this part: "Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars ... with over 220 owners of 1999-2005 cars participating." This sounds like you are saying there were 6 IMS failures out of 220 cars reporting. I'm no statistician, but doesn't 6/220 equate to a 2.7% failure rate for the 1999-2005 cars?
Over the course of the 2010 calendar year, yes. The total percentage will be higher, because more failed before 2010.

On the other hand, last time I checked none had been reported since the beginning of 2011. I think there's a chance that the great majority of those than are going to fail have already failed at this point.
If that was typical ( and it likely was not) there would be a 27% probability over 10 years.

The distribution is likely a rising curve, with nearly none the first year, rising to "lots" after 25 years.

I'd like to see the data, year by year, by model year - if the corpus is large enough.

Grant

Grant gee-lenahan-at-vee-eff-email-dot-net
You deserve credit for trying to determine reliability, but what good are the numbers you're producing at this point? If you go over to Planet 9, you'll see more than 30 reports of Boxster/Cayman engine failures highlighted in just two posts in this thread: [www.planet-9.com]. Plus, how do you reconcile your "about average" rating for 2005 cars against Consumer Reports' "much worse than average" rating for major and minor engine issues? And given that you've published data on 771 Boxsters and Caymans when their total production figures crossed the 300,000 mark this past June tells me you have a long way to go. When you've only surveyed 3/10s of 1 percent of all owners, you're not publishing "reliablity stats." They're just "stats." The effort is appreciated, but there's no science here and such a small sampling really doesn't help owners or those wanting to purchase cars.
The size of the population has no bearing on the necessary sample size. This fact is the foundation of all research based on samples.

With the planet-9 threads, that's thirty cars out of how many? And how long ago did those failures occur? Were they all within the past year? Or over the entire lifetimes of the cars?

With CR, what failure rate does a "much worse than average" rating for engine issues represent? And what time period does that rating cover? And how prone are CR's ratings to error, even with large sample sizes?

These other sources of information actually tell you a lot less than they seem to at first glance.
It's refreshing to find an unbiased expert on science, facts and statistics.
As an auditor, the calculation of sample size to get a meaningful statistic is a well known equation. Based on population, sample size and sample method being discussed here, the numbers will have a 95% accuracy rate with a plus or minus range of about 6 points. Not an unreasonable variance and certainly not significant enough to invalidate the trend data.

You should use an Internet search engine to query "statistical sample size". Or take an audit or statistics class to learn about survey confidence factors.

Dave - 06 987 S coupe SG/NL; gone (but still my first love): 03 986 AS/GG/BK;
Plenty of stats classes in the past here. I'm not sure which formula you're using, but the ones I've used are based on the variance (in the statistical sense, you've used the word differently) and do not include the population at all. I use a metric with a relatively low variance to make the most of small sample sizes.

I did find a widget like the one you seem to have used here:

[www.surveysystem.com]

But lower on the same page we have:

"The mathematics of probability proves the size of the population is irrelevant unless the size of the sample exceeds a few percent of the total population you are examining."

Which is essentially what I stated earlier. So it seems the widget is for use with small populations.
correct. My comment was actually to the comment about 30 failures on another forum. There is nothing random in asking who has had an IMS failure on a public online forum nor is a 2 post comparison relevant to the sample subject population.

I'm not a statistics guy, but in our auditing the sample size requirement is based on accounting principles and ceases to increase above a certain level (around 40) and with a large population such as 300,000, it makes the population irrelevant. Based on our audit methodology against this population, you can assume about a 95% accuracy rate within a range of + or - of 4 to 6 points with a relatively small sample as long as the sample is random. The sample methodology you are using seems as close to random as possible given the inability to pick randomly against the entire population. Being that it is online, possibly dilutes randomness a small amount but correlation with the online population against owner population is impossible but most likely not linked in a substantial way. However, with participation against volunteers and with no data from prior participant activity, my sense was it was good enough to use a base sample size. I used the top of the range as an example. Thus, the failure rates you are compiling are likely to be within the error range of other statistical measurements being used for comparison since it's unlikely that any such activity has a full population to draw from.

But I digress. eye rolling smiley There are of course other factors that influence mechanical longevity that are not represented in the data such as oil change interval, driving habits, etc. One of the posts noted an IMS failure rate of 2.7%. Given the sample error range, guesses of 50% are not in any way supported. It's not zero, but it's not very high either.

Dave - 06 987 S coupe SG/NL; gone (but still my first love): 03 986 AS/GG/BK;
Sorry, after I posted I realized that might be who you were responding to.

One unique way we compensate for the lack of a random sample: the survey collects data going forward, from the month the person signs up. So any past experiences that might have influenced them to participate won't influence the results.

Other reliability surveys and pretty much all forum polls are retrospective, and so much more open to non-response bias.

I do think it's possible that the cumulative failure rate is 15-20%, with most of these having already happened. If you have a forum with thousands of owners, it's easy to have 30 reported failures.
As I'm reading Dave & mkaresh's last two posts, it seems like the stats really do nothing but confirm that the IMS failure rate could be anywhere from near zero up to 20%. No offense intended, and if I'm misreading your posts and data, please feel free to correct me ... but it really seems like that wide a variance isn't much different than the various WAGs that have been expressed frequently here and in other forums for several years. I therefore suspect people will draw conclusions from your stats according to whatever they want to believe. I know you tried hard and I appreciate that, but it just seems like reliability stats are not going to provide any real information to resolve the controversy or help people make decisions.
I think you're confusing the failure rate for the past year with the combined failure rate for all years since the cars left the factory, and then reading the variance as extending from the low end of the former to the high end of the latter.
mkaresh, if I read this correctly (and for sake of argument assume your numbers are accurate w/o error), are you saying the annual IMS failure rate is about 2%, but the cumulative life time failure rate is about 20%? That might make sense: 2% per year over 10 years would be 20% failures. If this is correct, a 20% life time failure rate over 10 years is very troubling. I owned my 1999 for ten years and I expect to own the 2008 for maybe that long too. A 1 in 5 chance of total engine failure is not a warm and fuzzy feeling. Please tell me that I am somehow miss-reading your numbers.

1999 Arctic Sivler/black/black (sold) 2008s Silver/black/black - so predictable 2011 Outback 8/24/2011 first Grandson
% is not cumulative unless 2% of every 1999 sampled (assuming same samples too) failed each year over a 10 year survey. Then you could say over 10 years, 20% of 1999s sampled failed.

Whereas, if over a 10 year review of each MY, you had 2% failure rate per MY (not 2% per MY per survey year and assuming same samples and sample sizes are used for each MY) you'd only have a 2% total failure rate.

Not to be confused w/ a 10 year review of all MYs over the last 10 years... then, in order to have a 20% failure rate you'd need to have the following (assume samples and sample sizes are same for each MY):

MY 2010 - 20% of all 2010s fail in 2010
MY 2009 - 10% failure of all 2009s fail in each year: 2009, 2010
MY 2008 - 6.66% failure of all 2008s fail in each year: 2008, 2009, 2010
MY 2007 - 5% failure of all 2007s fail in each year: 2007, 2008, 2009, 2010
MY 2006 - 4% failure of all 2006s fail in each year: 2006, 2007, 2008, 2009, 2010
MY 2005 - 3.33% failure of all 2005s fail in each year: 2005, 2006, 2007, 2008, 2009, 2010
MY 2004 - 2.85% failure of all 2004s fail in each year: 2004, 2005, 2006, 2007, 2008, 2009, 2010
MY 2003 - 2.5% failure of all 2003s fail in each year: 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
MY 2002 - 2.22% failure of all 2002s fail in each year: 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
MY 2001 - 2% failure of all 2001s fail in each year: 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010

Clearer?

Remember the old saying: "Statistics don't lie... but liars use statistics" winking smiley

More nuances of course but generally speaking this covers the bases. Warm & fuzzier now? smiling smiley



Edited 1 time(s). Last edit at 08/18/2011 09:08AM by Burg Boxster. (view changes)
I am not taking any shots at MKARESH and his efforts to compile data w/ my 'liars use statistics' comment. Quite the contrary. I applaud his efforts and truly hope he captures relevant, accurate and statistically defensible data so all Boxster owners can breath easier. Thus far, his surveys seem to be the most objective out there. Someone posting links to other forums requesting you jump on the band-wagon and list if you've had a failure is too biased. Why? B/c you'll mostly get only failures reported which will yield a 90% failure rate. Likewise, someone spouting off 4 of 5 callers to their shop last week all had failures doesn't mean an 80% failure rate eye rolling smiley How do you account for the tens of thousands (and/or more) who didn't and never will call?

Again, M96/97 motors are not infallible. However, in my personal experience and based on all the data I've read and seen, failure rates are not anywhere near the 20% hysteria inducing levels widely promulgated here.

smiling smiley
I don't have enough data to nail down the cumulative failure rate, partly because this isn't the primary focus of the survey. One thing that has crossed my mind: for serious problems that seem to be common I could conduct special surveys of those who've signed up to participate. Many other things on my plate right now, but I'm going to seriously consider doing this.

Based on what I have seen I do not doubt that it could be as high as 20%, and is likely at least 10%.

Given the way some people talk as if the failure rate is virtually 100 percent, I thought this was actually good news.

Also consider that it's possible that most of the IMS bearings that are going to fail have already failed at this point.They do seem to have been more common in the past than over the last year or two. A special survey could get a better idea of the mileage at which IMS bearings tend to fail, and perhaps a point past which they are highly unlikely to fail if they haven't already. There could be a bimodal distribution, with another point at which failures again increase owing to "normal" wear. Cars 2007 and newer don't seem to have the problem at all.

Even if failures don't become less likely over time, might you get a fuzzier feeling if you thought of it as an 80-90 percent chance of not having a catastrophic engine failure? I don't tend to take this view myself, but know people who do, and they're usually happier.
I've participated in the survey. I've been one of the trouble free...heck almost expense free ... statistics.

But when you post an invitation to participate, you get a disproportionate number of people with an agenda to participate. And when the sample is as small as it is, that selection bias influences any meaning the statistics you derive from the information might have.

IMHO no one has good figures...not Porsche...not Jake...not Mike. Because no one receives all the information.

Porsche IMHO fails the test because so many out of warranty owners never go near a dealer. Or, if they do, once diagnosed, they go elsewhere for their solution or do it themselves so Porsche don't get a representation of the total population especially of the older cars. Not every failure goes to a Porsche dealer for a Porsche remanufactured engine.

Jake...IMHO he only gets the calls from those worried or already with a problem...the many who just drive their car trundle on and he never hears from them. So, if he gets 5 calls a day times 220 working days a year, he gets to talk to what percentage of the 986, 987 and 996, 997 population...darn small. Biased sample. If you talked to mostly people whose product had failed, you'd think the problem was pervasive too after a while.

Mike ... IMHO he gets the self selecting...those with an agenda, or just those willing to participate...but not necessarily representative of the total population of M96 engines. If the statistics were derived only from those who had been reporting regularly before the IMS failure report I'd be more inclined to think they weren't skewed. I applaud Mike's efforts and over time they may tell us more as they get more representative with more participation.

I don't ascribe dark motives to Jake...I've talked with him several times (always on his nickle and sometimes for over an hour) and my take is he really really cares about our engines far more than profits (though no one can continue in business without profits nor fund new products without profits). I've routed to him many queries from online posters where I thought only someone like he could help and he has always gotten into the discussion both promptly and with no possible revenue motive. He could have blown off my "can you help this guy" emails with "too busy" if he was only about revenue but he hasn't and I'm perfectly willing to say those actions to me overcome many faults he might seem to some to have (don't we all have some?). Those who have dealt with him and what he has done to/for their cars seem to have a very high opinion.
I don't think agendas are much of a factor given the way the survey and survey process are designed. Willingness to participate is more of a factor, but any correlation this might have with the owner's experience with the car is minimized by the research design. Not eliminated, but not a source of substantial distortion.

This comes through in the reported repair frequencies--they'd be much higher if people with problems were much more likely to participate. In this case, the percentage of people reporting IMS failures would likely be much higher. We have quite a few models in the survey for which hardly any problems have been reported.

The primary limitation, by far, is the number of participants. The results clearly become more precise as the number of participants increases.
But the invitation to participate is via online forum postings. Isn't there a probability that more of the folks reading online forums are there because they had a problem? And isn't there a greater probability of participation if you have something to report (a problem) than if you don't? I know in my case there were sometimes years in between anything worth reporting. And I'm sure I delayed my participation until I had something.
Luckily the great majority of participants respond following the end of each quarter as the emails request, whether or not they've had a repair. If there's no repair only an approximate odometer reading is needed--takes me about 10 seconds for my own cars.

With those who respond more sporadically we follow-up with special emails to fill the gaps.

When we do have an isolated response for a repair, it's often excluded from the analysis.

But what's the point of participating if doing so might distort the results?
I concur about the self-selecting sample, and I have an admission to make that kind of helps prove Mike's point. I ... um ... didn't participate in the sample because I hadn't had any troubles with my 08 Cayman and had barely owned it a year or so when the survey was started. I mean, really, I can't be the only one who figured, "What is the point in participating if you don't have anything to report?" Plus, I can make up a lot more excuses if anybody wants to hear them. [I'm bad, I know.]
I should clarify my earlier comment. About forty-percent of the people who sign up never respond, no matter what. Many intend to, but it's just one of those things that will always be put off until tomorrow. (I have quite a few such things myself.) The non-participation of this group doesn't so much directly distort the results as reduce the sample size, which reduces the precision of the results.

Another 20 percent or so respond once or twice, but not after that. Most of these responses are automatically discarded from the analysis, whether or not they report a repair. (We do have some venting in this group.)

What I meant in my previous post is that the great majority of the data included in the analysis comes from people who participate quarter after quarter regardless of whether or not they have a repair to report. For this reason the data resemble what would be obtained from a random sample even though the initial sample is self-selecting and not random.

One thing people seem to realize and yet not realize is that a report of no repairs is just as useful as a report of a repair--both are equally essential to the calculation of accurate results. It has been very hard to communicate this because no one has enough time to carefully read all of the email they receive.
probably less than or equal to the range Mike estimates (if in fact more afflicted people tend to sign up).

Grant (who does participate and is another nearly cost free statistic)
Reporting a 2001 Porsche Boxster S in Tuscon Arizona IMS bearing failure at 102,600 miles. (not my Boxster, found on craigslist, emailed owner for questioning) Owner report there was no warning or strange noise, engine shutdown, dealership confirms no compression, and metal pieces in oil pan.
So?
Boxsterra - 2 months ago
We have updated reliability stats for the Boxster and Cayman that include owner experiences through September 30, 2011.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 28, better than average

2007: 75, about average, might be high because final year of standard warranty

2006: 39, better than average

2005: 72, about average

2004: 44, better than average, small sample size

2003: 53, better than average

2002: 135, about average

2001: 97, better than average

2000: 120, about average

1999: 71, better than average, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds," to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop.

For the years where we have enough data for these:

2008: 67, < 1

2006: 65, < 1

2001: 51, 11

Porsche Boxster Lemon-odds and Nada-odds

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986).

No IMS failures have been reported for cars 2006 and newer.

No IMS failures have been reported for any model year since the beginning of 2011, with over 200 owners of 1999-2005 cars participating.

As before, rear main seal and air/oil separator repairs are far more common.

Thank you, once again, to everyone who has been helping. We'll have further updates in February and May. With more participants we could provide more precise information and cover all model years.

To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We have updated reliability stats for the Boxster and Cayman that include owner experiences through December 31, 2011.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 31, better than average

2007: 72, about average, might be high because final year of standard warranty

2006: 49, better than average

2005: 56, better than average

2004: 16, better than average, small sample size

2003: 42, better than average

2002: 97, better than average

2001: 95, better than average

2000: 100, about average

1999: 60, better than average

We have two additional statistics, "Nada-odds" and "Lemon-odds," to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop.

For the years where we have enough data for these:

2008: 72, < 1

2006: 71, 4

2005: 49, 4

2001: 42, 4

Porsche Boxster Lemon-odds and Nada-odds

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986).

No IMS failures have been reported for cars 2006 and newer.

No IMS failures have been reported for any model year since the beginning of 2011, with over 200 owners of 1999-2005 cars participating.

Rear main seal and air/oil separator repairs remain far more common.

Thank you, once again, to everyone who has been helping. We'll have further updates in May and August. With more participants we could provide more precise information and cover all model years.

To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons

Porsche Boxster repair odds
Thanks for the update Michael. Appreciate your site (I participate in it with 3 of my cars, all different OEMs).
I was wondering why there is no info listed for 1997 and 1998 models.
Quote
p-wagen
I was wondering why there is no info listed for 1997 and 1998 models.

Not enough owners signed up. The 1998 is only three short, though.
We have updated reliability stats for the Boxster and Cayman that include owner experiences through June 30, 2012.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 42, about average, stricter recalibrated scale

2007: 34, better than average

2006: 45, better than average

2005: 33, better than average

2004: 29, better than average, small sample size

2003: 38, better than average

2002: 65, better than average

2001: 78, better than average

2000: 70, better than average

1999: 110, about average

We have two additional statistics, "Nada-odds" and "Lemon-odds," to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop.

For the years where we have enough data for these:

2008: 72, 6

2006: 65, 4

2005: 73, 3

2002: 59, 8

2001: 58, 3

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986).

No IMS failures have been reported for cars 2006 and newer.

No IMS failures have been reported for any model year since the beginning of 2011, with over 200 owners of 1999-2005 cars participating.

Rear main seal and air/oil separator repairs remain far more common.

Thank you, once again, to everyone who has been helping. We'll have further updates in November and February. With more owners involved we could provide more precise information and cover all model years.

To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
I appreciate all of the work that Mkarash has done to accumulate and speak on the data. Whether IMS failure rate is 2% or 20%, its 100% if it happens to you. I was wondering what % of cars if driven to say 200K, would experience an IMS related failure. As I posted before, I had an IMS failure at 39K and 4 years old. Now I have 151K (chassis), and I am wondering what my chances are of having another failure? It would be interesting to know how many cars out there have say, 200K and have experienced IMS failures.

BarryL
2002 2.7 base tip
Dublin, CA
151K
We have updated reliability stats for the Boxster and Cayman that include owner experiences through September 30, 2012.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008: 42, about average, stricter recalibrated scale

2007: 29, better than average

2006: 38, better than average

2005: 34, better than average

2004: 50, better than average, small sample size

2003: 46, better than average

2002: 67, better than average

2001: 80, better than average

2000: 44, better than average

1999: 117, worse than average

We have two additional statistics, "Nada-odds" and "Lemon-odds," to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop.

For the years where we have enough data for these:

2008: 68, < 1

2006: 74, 4

2005: 78, 3

2001: 44, < 1

Important note: these cars are only driven an average of 5,500 miles a year, about half as much as the average car. No adjustment has been made for this.

Six IMS failures were reported for the 2010 calendar year, most of them for 2004-2005 model year cars (with the 2005 being a 987 rather than a 986).

No IMS failures have been reported for cars 2006 and newer.

No IMS failures have been reported for any model year since the beginning of 2011, with over 200 owners of 1999-2005 cars participating.

(Update: We had one report of a failing but not yet failed IMS in the following, fourth quarter.)

Rear main seal and air/oil separator repairs remain far more common.

We'll have further updates in February and May. We'd love to provide more precise stats and cover more model years--just a matter of getting more owners involved.

To see how competitors compare, and to sign up to help:

Porsche Boxster reliability ratings and comparisons



Edited 1 time(s). Last edit at 02/01/2013 12:03PM by mkaresh. (view changes)
thanks for the update
Interesting info, especially on the 987's with no IMS failures. If the ave. mileage is 5.5K/yr, that would put the early 987's at about 45k--what our old 06 had, with no IMS issues.

Having said that, IIRC, most of the IMS failures I've seen posted on 986 models have occurred after 50K miles,so the 987 record of no IMS failures on the 06 and newer cars might be just a function of low miles rather than design improvements.

Adding to that is Grant's experience on what was thought to be an IMS failure--that was not. Can't help but wonder how many other failures there have been incorrectly attributed to the IMS.
We've updated our reliability stats for the Boxster and Cayman to include owner experiences through March 31, 2013.

Repair frequencies, in terms of repair trips per 100 cars per year:
2008 Boxster: 38, low
2007 Boxster: 43, low
2006 Boxster: 21, low
2005 Boxster: 21, low
2004 Boxster: 52, low, small sample size
2003 Boxster: 48, low
2002 Boxster: 64, low
2001 Boxster: 46, low
2000 Boxster: 23, low
1999 Boxster: 105, moderate

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:
2008 Boxster: 69, < 1
2006 Boxster: 83, < 1
2005 Boxster: 85, < 1
2001 Boxster: 60, < 1

We'll have further updates in August and in November. To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster and Cayman reliability ratings and comparisons
We've updated our reliability stats for the Boxster and Cayman to include owner experiences through September 30, 2013.

Repair frequencies, in terms of repair trips per 100 cars per year:

2008 Boxster: 29, low
2007 Boxster: 46, low
2006 Boxster: 26, low
2005 Boxster: 22, low
2004 Boxster: 42, low, small sample size
2003 Boxster: 60, low
2002 Boxster: 62, low
2001 Boxster: 56, low
2000 Boxster: 30, low
1999 Boxster: 42, low

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2008 Boxster: 71, < 1
2007 Boxster: 75, < 1
2006 Boxster: 77, < 1
2005 Boxster: 75, < 1
2001 Boxster: 60, < 1

We'll have further updates in February and in May. The more owners participate, the more comprehensive and precise these will be.

To see how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster to include owner experiences through December 31, 2013.

Repair frequencies, in terms of repair trips per 100 cars per year:

2013 Boxster: 0, low, small sample size
2008 Boxster: 33, low
2007 Boxster: 47, low
2006 Boxster: 39, low
2005 Boxster: 40, low
2004 Boxster: 47, low, small sample size
2003 Boxster: 52, low
2002 Boxster: 85, moderate, small sample size
2001 Boxster: 57, low
2000 Boxster: 39, low
1999 Boxster: 24, low, small sample size

We have had one report in the past two years of an IMS bearing failure, to a car that had the bearing upgraded when its clutch was changed.

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2008 Boxster: 74, < 1
2006 Boxster: 67, < 1
2001 Boxster: 65, 3

We'll have further updates in May and in August. The more owners participate, the more comprehensive and precise these will be.

To see the repairs behind these stats, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
Our reliability stats for the Boxster now include owner experiences through March 31, 2014, making them nearly a year ahead of other sources.

Repair frequencies, in terms of repair trips per 100 cars per year:

2014 Boxster: 11, low
2013 Boxster: 0, low, small sample size
2008 Boxster: 31, low
2007 Boxster: 53, moderate
2006 Boxster: 47, low
2005 Boxster: 43, low
2004 Boxster: 35, low, small sample size
2003 Boxster: 60, low
2002 Boxster: 96, moderate, small sample size
2001 Boxster: 54, low
2000 Boxster: 48, low
1999 Boxster: 11, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2008 Boxster: 80, < 1
2007 Boxster: 68, < 1
2006 Boxster: 58, < 1
2005 Boxster: 62, < 1
2001 Boxster: 63, 3

Thank you, once again, everyone who has been helping. Next update in August.

For a deeper dive, including the stats for other cars and repair descriptions, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster to include owner experiences through June 30, 2014 (making them about 14 months ahead of some others).

Repair frequencies, in terms of repair trips per 100 cars per year:

2014 Boxster: 18, low
2013 Boxster: 16, low
2008 Boxster: 30, low
2007 Boxster: 33, low
2006 Boxster: 44, low
2005 Boxster: 45, low
2004 Boxster: 61, low, small sample size
2003 Boxster: 73, low
2002 Boxster: 87, moderate, small sample size
2001 Boxster: 64, low
2000 Boxster: 41, low
1999 Boxster: 24, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2008 Boxster: 82, < 1
2007 Boxster: 71, < 1
2006 Boxster: 66, 3
2005 Boxster: 61, < 1
2001 Boxster: 59, 3

Next updates in November and in February. The more owners participate, the more comprehensive and precise these will be.

For repair descriptions, the stats of other cars, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster to include owner experiences through September 30, 2014.

Repair frequencies, in terms of repair trips per 100 cars per year:

2014 Boxster: 20, low
2013 Boxster: 13, low
2008 Boxster: 23, low
2007 Boxster: 36, low
2006 Boxster: 40, low
2005 Boxster: 45, low
2004 Boxster: 42, low, small sample size
2003 Boxster: 53, low
2002 Boxster: 68, low, small sample size
2001 Boxster: 61, low
2000 Boxster: 45, low
1999 Boxster: 32, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2008 Boxster: 89, < 1
2007 Boxster: 64, < 1
2006 Boxster: 71, < 1
2005 Boxster: 66, < 1
2001 Boxster: 63, 6

To view the repairs behind these numbers, check the stats for other cars, and sign up to help improve this information (next update in February):

Porsche Boxster reliability ratings and comparisons
I was a manufacturing Quality Engineer, predictive and field reliability was part of the job. Blanket statement, take it for what it is worth in your opinion... with the best of circumstances accurate field data is a lofty expectation.
Within our company we targeted top 50 customers and flagged failures for failure analysis... from that we were able to verify failures and work to root cause. Logic being... best opportunity to collect failure case data, verify failure and root cause to lowest level component and drive next level component level failure analysis based upon failure rate. Monitor total install base and flag all failures for FA based upon trends.
Getting back product and collecting data as to what was occurring leading to or at time of failure required an extensive program. Failure analysis yielded real data. Our methodology was based upon total install base, calculated run time against total number of failures reported to determine a estimated Mean Time Between Failures. We compared that to the predictive and demonstrated MTBF to monitor trends. This required high touch, efficient processes and extensive buy-in across the company. There was so much more to this program but you get the idea.
Questioning the methodology provided by the poster based upon formulas, sample size... I believe it is better served to accept there is a problem, which we are aware of. The data provided indicates a trend(s) of the sample size made available to work with. Predictive or comprehensive... not feasible.
I applaude the effort... with what data you have available, nice work.
We've updated our reliability stats for the Boxster to include owner experiences through the end of 2014.

Repair frequencies, in terms of repair trips per 100 cars per year:

2014 Boxster: 21, low
2013 Boxster: 11, low
2008 Boxster: 19, low
2007 Boxster: 49, low
2006 Boxster: 30, low
2005 Boxster: 40, low
2004 Boxster: 44, low, small sample size
2003 Boxster: 54, low
2002 Boxster: 36, low
2001 Boxster: 65, low
2000 Boxster: 52, low
1999 Boxster: 33, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster: 84, < 1
2008 Boxster: 86, < 1
2007 Boxster: 64, < 1
2006 Boxster: 76, < 1
2005 Boxster: 71, < 1
2001 Boxster: 59, 8

We'll have further updates in May and in August. The more owners participate, the more comprehensive and precise these will be.

For the details, including how competitors compare, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
So, how have these cars been doing lately?

Our reliability stats for the Boxster and Cayman now include owner experiences through March 31, 2015. (Others are nearly a year behind.)

In terms of repair trips per 100 cars per year ("low" is best):

2014 Boxster and Cayman: 17, low
2013 Boxster: 13, low
2008 Boxster and Cayman: 23, low
2007 Boxster and Cayman: 53, low
2006 Boxster and Cayman: 29, low
2005 Boxster: 38, low
2004 Boxster: 40, low, small sample size
2003 Boxster: 65, low
2002 Boxster: 31, low
2001 Boxster: 70, low
2000 Boxster: 64, low
1999 Boxster: 63, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster and Cayman: 79, < 1
2013 Boxster: 92, < 1
2008 Boxster and Cayman: 82, < 1
2007 Boxster and Cayman: 58, < 1
2006 Boxster and Cayman: 73, < 1
2005 Boxster: 73, < 1
2001 Boxster: 50, 8

No stat for a model year? Then we very much need more participants for it. Same for those with small sample size notes. Next updates in August and November.

For repair descriptions, the stats for other cars, and to sign up to help improve this information:

Porsche Boxster reliability
Our reliability stats for the Boxster and Cayman now cover the year ending June 30, 2015. (Others are over a year behind, and so report how these cars were doing when a year younger.)

Repair trips per 100 cars:

2015 Boxster: 0, low, very small sample size
2014 Boxster: 14, low
2013 Boxster: 10, low
2008 Boxster: 29, low
2007 Boxster: 53, low
2006 Boxster: 29, low
2005 Boxster: 42, low
2004 Boxster: 33, low, small sample size
2003 Boxster: 36, low
2002 Boxster: 30, low
2001 Boxster: 65, low
2000 Boxster: 61, low

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster: 85, < 1
2013 Boxster: 89, < 1
2008 Boxster: 83, 3
2007 Boxster: 57, < 1
2006 Boxster: 71, < 1
2005 Boxster: 61, < 1
2001 Boxster: 57, < 1

If a model year isn't listed, or its sample size is small, then we especially need more owners involved.

For the descriptions of all reported repairs, results for other models, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster to include owner experiences through September 30, 2015.

Repair frequencies, in terms of repair trips per 100 cars per year:

2015 Boxster: 11, low, very small sample size
2014 Boxster: 11, low
2013 Boxster: 17, low
2009 Boxster: 40, low, very small sample size
2008 Boxster: 27, low
2007 Boxster: 61, moderate
2006 Boxster: 32, low
2005 Boxster: 49, low
2004 Boxster: 33, low, small sample size
2003 Boxster: 45, low
2002 Boxster: 47, low
2001 Boxster: 57, low
2000 Boxster: 65, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster: 88, < 1
2013 Boxster: 88, < 1
2008 Boxster: 80, < 1
2007 Boxster: 51, 4
2006 Boxster: 76, < 1
2005 Boxster: 58, < 1
2001 Boxster: 63, < 1

We'll have further updates in February and in May. The more owners participate, the more comprehensive and precise these will be. If a model year is missing or has a small sample size, we especially need more participants for it.

For the details, including repair descriptions, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster and Cayman to include owner experiences through the end of 2015, about eight months ahead of other sources.

In terms of repair trips per 100 cars per year--lower is better:

2015 Boxster and Cayman: 9, low, small sample size
2014 Boxster and Cayman: 12, low
2013 Boxster and Cayman: 22, low
2009 Boxster and Cayman: 32, low, small sample size
2008 Boxster and Cayman: 38, low
2007 Boxster and Cayman: 42, low
2006 Boxster and Cayman: 32, low
2005 Boxster: 45, low
2004 Boxster: 41, low, small sample size
2003 Boxster: 56, low
2002 Boxster: 48, low
2001 Boxster: 54, low
2000 Boxster: 64, low, small sample size

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster and Cayman: 90, < 1
2013 Boxster and Cayman: 79, < 1
2008 Boxster and Cayman: 76, 3
2007 Boxster and Cayman: 67, 4
2006 Boxster and Cayman: 76, < 1
2005 Boxster: 60, < 1
2001 Boxster: 57, < 1

No stat, or a small sample size? Then we especially need more owners involved for that model year.

For the repairs behind these stats, reliability information on other models, and to sign up to help improve this information:

Porsche Boxster and Cayman reliability ratings and comparisons
We've updated our reliability stats for the Boxster and Cayman to include owner experiences through June 30, 2016. Reliability information elsewhere is based on a survey conducted in April 2015--well over a year ago.

Repair frequencies, in terms of repair trips per 100 cars per year (lower is better):

2016 Boxster and Cayman: 30, moderate, small sample size
2015 Boxster and Cayman: 24, moderate, small sample size
2014 Boxster and Cayman: 27, moderate
2013 Boxster: 26, low
2009 Boxster and Cayman: 15, low, small sample size
2008 Boxster and Cayman: 49, low
2007 Boxster and Cayman: 37, low
2006 Boxster and Cayman: 31, low
2005 Boxster: 44, low
2004 Boxster: 32, low, small sample size
2003 Boxster: 58, low
2002 Boxster: 28, low, small sample size
2001 Boxster: 54, low

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster and Cayman: 76, < 1
2013 Boxster: 82, < 1
2008 Boxster and Cayman: 54, < 1
2007 Boxster and Cayman: 69, < 1
2006 Boxster and Cayman: 74, < 1
2005 Boxster: 61, < 1
2001 Boxster: 65, 3

We'll have further updates in November and next February. The more owners participate, the more comprehensive and precise these will be.

For the repairs behind these stats, reliability information on other models, and to sign up to help improve this information:

Porsche Boxster reliability ratings and comparisons
We've updated our reliability stats for the Boxster and Cayman to include owner experiences through September 30, 2016.

Repair frequencies, in terms of repair trips per 100 cars per year (lower is better):

2016 Boxster and Cayman: 21, low, small sample size
2015 Boxster and Cayman: 23, moderate, small sample size
2014 Boxster and Cayman: 26, moderate
2013 Boxster: 29, moderate
2009 Boxster and Cayman: 8, low, small sample size
2008 Boxster and Cayman: 40, low
2007 Boxster and Cayman: 37, low
2006 Boxster and Cayman: 29, low
2005 Boxster: 39, low
2004 Boxster: 47, low, small sample size
2003 Boxster: 51, low
2002 Boxster: 47, low, small sample size
2001 Boxster: 53, low

We have two additional statistics, "Nada-odds" and "Lemon-odds", to indicate the percentage of cars with no repairs in the past year and those that required 3+ trips to the repair shop:

2014 Boxster and Cayman: 75, < 1
2013 Boxster: 79, < 1
2008 Boxster and Cayman: 62, < 1
2007 Boxster and Cayman: 71, 4
2005 Boxster: 65, < 1
2001 Boxster: 66, < 1

Additional participants always helpful, especially for years without stats or with small sample sizes.

For the repairs behind these stats, reliability information on other models, and to sign up to help improve this information:

Porsche Boxster and Cayman reliability ratings and comparisons
Really appreciate all the time/effort this reporting took. For my overview it looked like the 06 and 08 were great years, then the 08 began to slip last few years. And it looked like the 09 was running pretty bad stats till the latest example. Very interesting charts though.
Thanks
Sorry, only registered users may post in this forum.

Click here to login