When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Man oh man, that cognac/espresso interior color is HOT, HOT, HOT to me. Now if I could find a coupe with that exact full leather interior, 18 ways (or 4 way plus), and in a COLOR (i.e., not white, black, silver; prefer guards/carmine reds and night blue/graphite/sapphire blues) I'd be set.
squid, can you also pls remind me what all those funny numbers mean? (I think i read months ago but forgot, and the post is buried in this million post thread)
10,637 miles, $137,830 msrp, $44,340 loss, 67.83%
$4.17/mile, mile factor 0.694/10k
$1,263.99/month(w), age 2.9 years (35 months), factor 0.876/year, 3,641 miles/year (warranty)
$1,109.11/month(p), age 3.3 years (39 months), factor 0.890/year, 3,194 miles/year (production)
Right now the two-three "factors" are what you combine with an exponent that is the age in years (or in 10k miles) to arrive at the difference between msrp as built and current asking price.
So theoretically you can compare the losses for 2 and 3 year old cars, same yearly loss in percent should have the same factor.
I have transported all this stuff into a better programming language and will do a bit more on-target stuff. Specifically, the first year distorts this number because most cars will have some money off the msrp as built. I think I have enough sample data now to do some machine learning by throwing it into tensorflow (hey it's the fashion and I need it on my resume) give me an extra factor to manipulate the starting number for the progression so that the factor doesn't change as much over the 2-4 years.
So, in the above example:
$1,263.99/month(w), age 2.9 years (35 months), factor 0.876/year, 3,641 miles/year (warranty)
msrp * factor^years = current asking price
$137,830 * 0.876^2.9 = $93,490
pfactor and wfactor are the same computation done on build date and warranty start date. Warranty start date gives better data but is not available for all cars.
I also want to do this separately for MSRP for the car model and MSRP as built. Most options have more value loss than the car, so the number also gets distorted by making highly optioned cars appear as better deals (lower factor), but in reality they aren't any cheaper, they are just filled with options you might not care about.
Generally it works well enough already for same model generation and even one generation back (in this case 997.2 cars look much in line with 991.1). It doesn't work great for 997.1 turbos and the like, too old, flattens out. For most purposes anything below 0.900 is a good deal and anything above 0.900 is not. However, I did not integrate miles and age into a single thing now, so naturally low mileage cars will have less attractive looking factors for the age items, but if you then look at the mile factor it will probably drop to 0.5 or so.
ETA: instead of using a 10 kilomile factor I should probably use 7.5. Or stop being lazy and actually use the median or something.
I also work on pretty charts to see whether prices actually went up.
ETA2: I want to preemptively point out that I am not crawling/scraping the Porsche web site. I just hit control-s a lot.
In the very beginning stages of considering a 991 as a daily and I'm loving this thread. Hopefully this car I came across has not already been posted. Seems like a too good to be true deal unless I am missing something from my very short, but fairly deep, assessment of the market.
In the very beginning stages of considering a 991 as a daily and I'm loving this thread. Hopefully this car I came across has not already been posted. Seems like a too good to be true deal unless I am missing something from my very short, but fairly deep, assessment of the market.
pfactor and wfactor are the same computation done on build date and warranty start date. Warranty start date gives better data but is not available for all cars.
I see little to no value in listing two separate calculations based on both in-service (warranty) date and production date... List just the in-service date. If it doesn't exist, perhaps just take the production date plus 4 months as an estimate.
For that matter, anything more granular than model year is essentially meaningless for value calculations, only useful in terms of remaining warranty, if any.
eg: Two identically optioned and same mileage cars, both 2012.5 model years, have essentially identical value regardless if it was the first 2012.5 produced or the last one built before rolling to model year 2013...
Sorry if already posted, it is a manual, has SC and sport techno wheels. Not a huge hot deal but thought the manual may interest some. Clear side markets probably make it a Porsche fan.
^^ Above post from September 2016. This ones getting around.
Looks like up for sale again at CarMax in Garland, TX.
Ahhhhhh, I knew there had to be something. Front and rear damage, but no airbags would be pretty minor, and definitely not structural. Don't like the prospect of paint work, but studying the pics it looks to have been done well. Any potential buyer should dig into that and require records of the repair.
Is that car as far below the market as I think it is?
Edit: now that im looking more closely for the damage on that car, it does seem as though some of the panel alignment in the rear is not quite right...
06-30-2017, 07:45 PM
