LVDell

I never said we need 100% of PCNA's data. We just need access to the population we intend to sample. Sure, access to 100% of the data would be awesome and we could just report basic descriptive results as a number of failures (cause undetermined) per "X" number of motors produced or even total number and percentage.

The problem is that you will NEVER get eyes on the PCNA data set.

Trying to collect data from Rennlist is a two-fold problem......self-report (full of errors) and zero random sampling even of this convenience sample.

I'm not (nor are my other stats colleagues here on RL) going to go over a full stats debate again as this has been talked to death over the last decade.

ps. If I had to hypothesize what the number truly is based on what we see here I would say that if the self report data on a website is 10% failure rate then the actual number is probably about 1/10 of 1% IRL.

BruceP

Totally concur with Dell's comments. I think what gets people wound up about this kind of thread is that bad information is more damaging than no information is, both in terms of the value of the cars and the quality of the ownership experience.

Here's the most interesting piece of data I know of: Nobody has ever been able to certify a class against Porsche for this. When you consider the Type-A-ness of the typical Porsche driver, that's remarkable and telling.

LVDell

Winner winner chicken dinner!

Marco8

Guys, I don't know why this thread went so askew and has been purposely diverted to a discussion of statistics.

Asking a group of 200-500 996 owner's "who has had an IMS issue", and of those who replied approx 10% said yes and 90% said no -Done. Ask the same question of 997 owners 5% say yes 95% say no. Makes sense, younger car, updated IMS and fewer miles.

This is not an end all statistical study on IMS failures but a simple question, take it for what it is.

Thank you for everyone who has participated (including those who voted and then went on to condemn and sidetrack it).

LVDell

You don't know why? You said you did a search right? Fail.

Some people just don't get it and either don't realize they don't get it or do and then attack those that try to explain the flaw of a NOT so simple question and how to ask it.

But if you feel better asking a question that can have zero empirical support here other than flawed anecdotal water cooler chit chat then you did well.

Don't get your panties in a bunch when you are explained why this question has been asked a million times. Why it can be asked no differently for the millionth time. Why your thread is going to do nothing to get any sort of different answer (look up the definition of insanity) than any previous one. Etc, etc, etc..

Drive your car, enjoy it, or sell it if you worry about the hype.

Geez, how I keep getting dragged into this crap.

Marco8

By your own choice

LVDell

Good point. I guess I'm a sucker for crap driven by ignorance for which I have a background in. Time to stop thinking I can educate the ignorant.

I'm going to go enjoy my car, have fun worrying about the IMS.

jasper

Remember this post LV?

For what it's worth I think we have now firmly established that 10% of the 996 associated with Rennlist have had IMS failures. Of course it's been established before, and will likely be established again, but this is what we now know.

That's something at least. It's not much, but it's something. I would think it's the high water mark anyway.

LVDell

Sure do. And trust me, I tried. However, some just never get it and will continue to ask these silly questions.

But no, we haven't established that 10% of those on RL have had IMS failures. Remember, this has NEVER been a sampling but rather a voluntary poll.

But like I said already, I agree that "if" 10% is the water cooler number for the 996 forum then reality is probably closer to .01%

medtech

85,000 miles and no problems.

Interesting that your survey results agree w/ Jake Raby's estimates of about 10%.

FYI, the IMS problem was not addressed with the 997 until recently, so that 5% may increase with time.

LVDell

Anybody have any idea how Jake arrived at his number?

Tippy

Last I heard, my old car (2000 3.4L) far north of 100k still has it's original IMS.

Now my '87, well, its wrapped up and preserved right now waiting to be put back in....

1999Porsche911

Yeah. It was determined that 10% risk would sell more bearings than 1/10%. It is about as scientific as this survey is.

Michael Gibney

For me this is an interesting thread -- it is obvious that stats are well beyond those of us who have degrees that did not include a calculator! The best part is seeing that the number is between 0.01% and 10% and for me that's ok because when factored into the cost of the car it makes for sensible buying.
And LVDell -- thanks for getting involved -- in reading back over the entire thread -- you did get your message through to me.

LVDell

Thanks Michael. Very welcome

You seem to have a good approach to this. In otherwords, you know what the "perceived" hype is and decided the car is fantastic regardless. No sense in worrying about something so remote and just go out and enjoy it! Happy motoring........