How rare is this car?
#31
Seems high despite the aerokit, think it might be a mistake in the data.
#32
Seems weird, it's a Carrera S Guards Red/Black Full Leather, $149,560.
The custom tailoring might be doing it or it might be that the VIN decoders don't capture something.
List of options for this one:
By the way, below I attach some of the original price lists:
The custom tailoring might be doing it or it might be that the VIN decoders don't capture something.
List of options for this one:
- Self Dimming Mirrors
- Heated Front Seats
- Bose High End Sound Package
- Black Full Leather
- Carrera S Wheel in Black
- Guards Red
- Black Mats - Porsche lettering
- Adaptive Sport Sts Lthr Int
- Aerokit
- Sport Shifter
- Sport Exhaust System
- Seat Belts - Silver
- Custom Tailoring
By the way, below I attach some of the original price lists:
Last edited by nikandriko; 06-25-2015 at 02:52 PM.
#33
A little combinatorics....very nice.
Sean, as you know, the Muscle Cars are built to a standard so the manufacturer can offer a 1 of X provenance. Porsche has also done this with certain models that are built as limited release cars but it is the model (e.g. recent Speedster) that is limited in production and so numbered - but you could still customize the order to some extent. So, you are spot on in the idea that there are a lot, and I mean by this A LOT, of one-off 997s that are floating around because there are so many options. Sadly, if you really wanted to know just how rare your proposed combination is you would have to parse through the build sheets.
To appreciate how truly one-off a 997 Porsche is look at the possible combination of options, which would be C(n,r) = n!/r!(n-r)! (where n is the set size and r is the subset size). Also, proceed on the assumption that each option is equally likely to be chosen (you could weight the options but this complicates the math). Also assume that one-off and rare are the same thing.
In your example: a 997 2006 C4S with aero kit and power kit and lets assume that the two options were available for the entire production run (2005 to 2012). I'm just guessing at the eight year production run sent to the U.S. but lets just say it was 9000 units - so the set size is 9000. The subset is model years (of which there are 8) and the two options - so the subset is 10. The result is 1.95607 x 10^32 possible combinations. If you limit it to just the units produced in 2006 (1,156) then the possible combinations are 667,590. Alternatively, if you expand this sample size to all 997s and make the subset 11 (adding the C4S variant) the possible combinations are approximately 2.2 x 10^49. Of course not all options have an an equal chance of being chosen but you get the idea that we all pretty much own one-off cars! And that sort diminishes the cachet which is why PAG can charge so much for it's special edition / limited production models!
So, maybe the real question is: what is the distinction between one-off and rare. Now I am going to find a pin and count/calculate the number of dancing angels on the tip!
To appreciate how truly one-off a 997 Porsche is look at the possible combination of options, which would be C(n,r) = n!/r!(n-r)! (where n is the set size and r is the subset size). Also, proceed on the assumption that each option is equally likely to be chosen (you could weight the options but this complicates the math). Also assume that one-off and rare are the same thing.
In your example: a 997 2006 C4S with aero kit and power kit and lets assume that the two options were available for the entire production run (2005 to 2012). I'm just guessing at the eight year production run sent to the U.S. but lets just say it was 9000 units - so the set size is 9000. The subset is model years (of which there are 8) and the two options - so the subset is 10. The result is 1.95607 x 10^32 possible combinations. If you limit it to just the units produced in 2006 (1,156) then the possible combinations are 667,590. Alternatively, if you expand this sample size to all 997s and make the subset 11 (adding the C4S variant) the possible combinations are approximately 2.2 x 10^49. Of course not all options have an an equal chance of being chosen but you get the idea that we all pretty much own one-off cars! And that sort diminishes the cachet which is why PAG can charge so much for it's special edition / limited production models!
So, maybe the real question is: what is the distinction between one-off and rare. Now I am going to find a pin and count/calculate the number of dancing angels on the tip!
#34
Sean, as you know, the Muscle Cars are built to a standard so the manufacturer can offer a 1 of X provenance. Porsche has also done this with certain models that are built as limited release cars but it is the model (e.g. recent Speedster) that is limited in production and so numbered - but you could still customize the order to some extent. So, you are spot on in the idea that there are a lot, and I mean by this A LOT, of one-off 997s that are floating around because there are so many options. Sadly, if you really wanted to know just how rare your proposed combination is you would have to parse through the build sheets.
To appreciate how truly one-off a 997 Porsche is look at the possible combination of options, which would be C(n,r) = n!/r!(n-r)! (where n is the set size and r is the subset size). Also, proceed on the assumption that each option is equally likely to be chosen (you could weight the options but this complicates the math). Also assume that one-off and rare are the same thing.
In your example: a 997 2006 C4S with aero kit and power kit and lets assume that the two options were available for the entire production run (2005 to 2012). I'm just guessing at the eight year production run sent to the U.S. but lets just say it was 9000 units - so the set size is 9000. The subset is model years (of which there are 8) and the two options - so the subset is 10. The result is 1.95607 x 10^32 possible combinations. If you limit it to just the units produced in 2006 (1,156) then the possible combinations are 667,590. Alternatively, if you expand this sample size to all 997s and make the subset 11 (adding the C4S variant) the possible combinations are approximately 2.2 x 10^49. Of course not all options have an an equal chance of being chosen but you get the idea that we all pretty much own one-off cars! And that sort diminishes the cachet which is why PAG can charge so much for it's special edition / limited production models!
So, maybe the real question is: what is the distinction between one-off and rare. Now I am going to find a pin and count/calculate the number of dancing angels on the tip!
To appreciate how truly one-off a 997 Porsche is look at the possible combination of options, which would be C(n,r) = n!/r!(n-r)! (where n is the set size and r is the subset size). Also, proceed on the assumption that each option is equally likely to be chosen (you could weight the options but this complicates the math). Also assume that one-off and rare are the same thing.
In your example: a 997 2006 C4S with aero kit and power kit and lets assume that the two options were available for the entire production run (2005 to 2012). I'm just guessing at the eight year production run sent to the U.S. but lets just say it was 9000 units - so the set size is 9000. The subset is model years (of which there are 8) and the two options - so the subset is 10. The result is 1.95607 x 10^32 possible combinations. If you limit it to just the units produced in 2006 (1,156) then the possible combinations are 667,590. Alternatively, if you expand this sample size to all 997s and make the subset 11 (adding the C4S variant) the possible combinations are approximately 2.2 x 10^49. Of course not all options have an an equal chance of being chosen but you get the idea that we all pretty much own one-off cars! And that sort diminishes the cachet which is why PAG can charge so much for it's special edition / limited production models!
So, maybe the real question is: what is the distinction between one-off and rare. Now I am going to find a pin and count/calculate the number of dancing angels on the tip!
#37
Well, to answer seriously your question above, this wasn't a hack on the Porsche dealer admin page.
I recreated all the VIN numbers for the 2006 S models and used a free, public VIN decoder.
The data is still incomplete, the VIN decoder allows me max. 15 queries/IP/day so it will take me some time to fill in the gaps (although I'm using 6 different servers and 6-7 VPNs)
So far, I've detected 63 4S 2006 launch cars.
I'm curious as to whether there are more than the 64 that have been mentioned in the past and if so, how many more? 200 as theory suggests.. ?
The data for your pleasure and analysis is attached below:
I recreated all the VIN numbers for the 2006 S models and used a free, public VIN decoder.
The data is still incomplete, the VIN decoder allows me max. 15 queries/IP/day so it will take me some time to fill in the gaps (although I'm using 6 different servers and 6-7 VPNs)
So far, I've detected 63 4S 2006 launch cars.
I'm curious as to whether there are more than the 64 that have been mentioned in the past and if so, how many more? 200 as theory suggests.. ?
The data for your pleasure and analysis is attached below:
Last edited by nikandriko; 06-25-2015 at 02:51 PM.
#39
With all the tricks I'm doing (multiple IPs), I basically get ~200 queries/day and I've analyzed close to 500 cars by now. This is a $300/month worth of a subscription to their API.
To do 2 years (2007, 2008), say in 2 weeks, you need 4,000 queries per week, i.e. 600 queries/day. This is a $400+ monthly subscription.
Last edited by nikandriko; 06-19-2015 at 11:25 AM.
#40
I do think though there's something interesting about launch cars in that they're usually bought by early adopters who can't wait and grab one of the first cars available.
My car was originally bought in Tampa, Florida likely by an early adopter who later sold it at 25k miles on eBay.
The car is in extremely good condition and I think part of it is that there has been some self-selection bias in this process.
Since this 'PAG code - Launch code' has ignited my interest, I wrote a couple of scripts and I'm analyzing all the VIN codes of the 2006 S+4S cars.
Over a random sample of 362 analyzed cars so far (serial numbers from 500 to 2,000), I've found 49 launch cars, 47 of which are the $102,080 and there are also 2 at $142,445.
After roughly serial number 1,200, the launch cars stop. Extrapolating the sample stats, you get ~160 cars if the distribution is uniform which I expect it not to be as there should have been more launch cars produced over at the beginning, thus the end number when I analyze the full dataset should be close to 200 as previously calculated. The production of the launch cars seems to have started in July 2005 and lasted till September 2015.
Some interesting plots so far:
My car was originally bought in Tampa, Florida likely by an early adopter who later sold it at 25k miles on eBay.
The car is in extremely good condition and I think part of it is that there has been some self-selection bias in this process.
Since this 'PAG code - Launch code' has ignited my interest, I wrote a couple of scripts and I'm analyzing all the VIN codes of the 2006 S+4S cars.
Over a random sample of 362 analyzed cars so far (serial numbers from 500 to 2,000), I've found 49 launch cars, 47 of which are the $102,080 and there are also 2 at $142,445.
After roughly serial number 1,200, the launch cars stop. Extrapolating the sample stats, you get ~160 cars if the distribution is uniform which I expect it not to be as there should have been more launch cars produced over at the beginning, thus the end number when I analyze the full dataset should be close to 200 as previously calculated. The production of the launch cars seems to have started in July 2005 and lasted till September 2015.
Some interesting plots so far:
However, for the original post, there is no comparison between Detroit muscle cars and the Porsche line up.
Detroit routinely assembles a few key options to make a select number of that model more exclusive or special. Those are the ones collectors lust after for bragging rights decades later.
However, with Porsche, the game is entirely different. When I ordered my 2006 S Cabriolet with this and that, I had over 165 pages of options to choose from. This is a big difference from Detroit.
The bottom line is this: Every car out of Zuffenhausen is unique. There are no two alike. You can debate which special options are better than others, but with such a range of options the Detroit arguments simply pale in comparison.
Back in the mid 1980s, Porsche bragged about how you would rarely see the same car twice out of Zuffenhausen. Their options list was vastly smaller in those days. Every 60-90 days during production you would see the exact same optioned car (e.g., Guards Red Turbo with Black full Leather...). Today, we simply never see the same car twice. The Porsche 911 series in particular is a bespoke line of cars. Each is custom ordered either by a very marketing savy sales guy at the dealership, or by the ultimate customer utilizing the "Special Wishes" department of Porsche (which is what we used to call it).
If you really want bragging rights over an exclusive car, start with the exterior color! I can tell you that there are perhaps 4 now in my color in North America. (A GTS coupe in San Jose, a 4S cab in New Jersey, a S Cab in New York (with periodic visits to Florida in the off season) and mine. And that's just the color!
As suggested earlier, if you want true exclusivity, get yourself a GT2 with a Paint to Sample exterior and you will most likely be the only guy in North America with that!
#41
#42
I second this motion, and would love to see all the production numbers like this awesome chart you have shared for the full run of the 997 from 2005 to 2012 by model year, interior/exterior color., with S and 4s cars separated. I am trying to find how many cars were produced in GT Silver each year with Bi-color stone/black interior compared to other color combinations. As the 997 becomes more coveted, a chart like this will become invaluable for years to come..
#43
My analysis is coming close to its end. So far I've analyzed ~1,400 VINs/cars for 997 S+4S 2006 cars.
From those, I've detected 202 4S launch cars, of which 173 are black/beige at $102,080, same as mine. This is close to the initial estimate of 200, which I think it will come close to once the full October production data is considered and also accounting for some VIN mistakes that seem to exist.
This is how production looks like by month. Although for October I still need more data, there is no evidence so far of launch car production after that point.
UPDATE: Analyzing a few more cars today, i.e. 1,600 I got just one more $102,080 car, i.e. 174 in total. I also observed that black/sand beige 4S that either don't have the $102,080 MSRP (16) or don't have the PAG code (5) in between but have most of the additional equipment should be launch cars as well, it's just that the VIN info is not exactly accurate. There are 21 more of those. So, total number 195 black/beige launch cars.
UPDATE #2: After ~2,000 cars analyzed, results remain the same, 195 black/beige 4S launch cars. Updated .csv attached below.
The complete Excel file is here for you to play with:
From those, I've detected 202 4S launch cars, of which 173 are black/beige at $102,080, same as mine. This is close to the initial estimate of 200, which I think it will come close to once the full October production data is considered and also accounting for some VIN mistakes that seem to exist.
This is how production looks like by month. Although for October I still need more data, there is no evidence so far of launch car production after that point.
UPDATE: Analyzing a few more cars today, i.e. 1,600 I got just one more $102,080 car, i.e. 174 in total. I also observed that black/sand beige 4S that either don't have the $102,080 MSRP (16) or don't have the PAG code (5) in between but have most of the additional equipment should be launch cars as well, it's just that the VIN info is not exactly accurate. There are 21 more of those. So, total number 195 black/beige launch cars.
UPDATE #2: After ~2,000 cars analyzed, results remain the same, 195 black/beige 4S launch cars. Updated .csv attached below.
The complete Excel file is here for you to play with:
Last edited by nikandriko; 06-28-2015 at 10:28 PM.
#45