Why yes - half a million people can be wrong!

"BLCKSTRM" (BLCKSTRM) 09/07/2016 at 01:13 • Filed to: None

3

29

Over half a million people have read this “advice” article from Gawker’s Two Cents - and it’s totally bogus.

!!! UNKNOWN CONTENT TYPE !!!

I keep coming back to this article as a prime of example of non-car people thinking they’ve made sense of car data based on what they think are reasonable assumptions.

Which turn out to be such poor assumptions they should have just picked numbers out of a hat.

I’ll quote it so you don’t have to click the link to read it and contribute to the reader total (yes, I didn’t use the block quote thing because there was already a block quote in the quote):

“Mobile car repair provider YourMechanic.com analyzed their own data to find out which car brands and models require the most maintenance over time. They explain:

At YourMechanic, we have a massive dataset of the make and model of the cars we have serviced and the type of maintenance done. We decided to use our data to understand which cars break down the most and have the highest maintenance costs...First, we looked at which major brands cost the most to maintain over the first 10 years of a car’s life. We grouped all years of all models by brand to compute their average cost by brand. In order to estimate annual maintenance costs, we found the amount spent on every two oil changes (as oil changes are generally done every six months).

“According to their data, BMWs were the most expensive to maintain by far, with a 10-year cost of $17,800. They found the luxury brands were the most expensive in general, but many budget vehicles ranked high, too. Saturns were 6th on their list with a 10-year maintenance cost of $12,400.”

And then here’s the image:

I mean, what’s so wrong with just letting the Jalopnik crew write the automotive articles? You know, so it can be right?

Am I the only one who can see the major HEADLINE glaring issue here?

CRider > BLCKSTRM 09/07/2016 at 01:22

1

Kias cost as much to maintain as land rovers, good to know that.

TheBimmerGuyWhoNowOwnsAChevy > BLCKSTRM 09/07/2016 at 01:47

0

I know BMWs can be finicky, but are the new ones even remotely that bad? Also Land Rover I feel should be higher in that list.

Dr. Zoidberg - RIP Oppo > BLCKSTRM 09/07/2016 at 01:53

1

Well, Jalopnik did compare a CVT Civic Coupe to a Prelude, so there’s that.

Out, but with a W - has found the answer > BLCKSTRM 09/07/2016 at 02:34

0

In order to estimate annual maintenance costs, we found the amount spent on every two oil changes (as oil changes are generally done every six months).

Seems legit, though I doubt the list would change much if they’d take into account all wearables and the exact oil change intervals. I do wonder how it’d look if repairs were included.

AMGtech - now with more recalls! > BLCKSTRM 09/07/2016 at 03:22

8

What about cars that only need an oil change once per year?

What about the fact that if they’re only using data from cars they’ve worked on it’s inherently flawed?

What about other maintenance items like timing belts, brake fluid, And transmission services?

How is the data affected when people skimp on those other services?

What if the cost of maintenance for a set period of time is built in to the cost of the vehicle, how is it calculated out for this “study”?

If this includes repairs, do they take warranty periods into account?

How do third party warranty and service contracts affect the data?

How does geography affect the data? For example, a shop in Iowa is probably going to have a considerably lower labor rate than a shop in LA.

Does the data consider DIY versus independent versus dealer costs?

Why did they include cars that haven’t even been produced for the better part of a decade?

And finally, why does a bullshit, clickbait “article” about a “study” that couldn’t have been less scientific had it been done by kindergarteners bother me so much?

The Lurktastic Opponaught > BLCKSTRM 09/07/2016 at 06:26

0

They’re assuming their sample is representative of the population as a whole. It’s a pretty common mistake.

Conan > AMGtech - now with more recalls! 09/07/2016 at 06:56

1

Stop it with your professionalism and knowledge!

Flyboy is FAA certified insane > BLCKSTRM 09/07/2016 at 07:17

0

Also “every two oil changes” is a big time gap for some vehicles. On some German cars it is 15000mi. Alot can happen in that time

Party-vi > Out, but with a W - has found the answer 09/07/2016 at 07:30

1

It’s not legit, as BMWs usually have a 10,000-15,000 mile service interval for oil, so they would actually cost half as much.

nerd_racing > Out, but with a W - has found the answer 09/07/2016 at 07:37

0

my bmw tells me every 5000 miles to change it out at 7 quarts.

KusabiSensei - Captain of the Toronto Maple Leafs > AMGtech - now with more recalls! 09/07/2016 at 07:56

1

It’s all about the “authors” confirming their own prejudices. Toyota/Honda good, everyone else bad.

Land_Yacht_225 > BLCKSTRM 09/07/2016 at 08:26

4

This is totally bogus! It does not cost $12,900 to maintain a Mercedes over 10 years!

It costs $12,900 to maintain a Mercedes over a single year...

Dru > BLCKSTRM 09/07/2016 at 08:40

0

The wife and I have owned her Kia since almost new, and we have spent MAYBE 2 grand in five years. This would include two 30K services at a dealer, and a set of tires from a dealer. Now I know I overpaid for those 30K services, but I can safely say only Kia mechanics have touched her car, which matters to me when she drives as much as she does.

GUYMANDUDE > BLCKSTRM 09/07/2016 at 09:32

1

Yes. Yes. Yes. BMW’s are expensive to maintain. You should sell yours to me. I will take care of her.

404 - User No Longer Available > TheBimmerGuyWhoNowOwnsAChevy 09/07/2016 at 09:35

0

I’m just praying the timing chain on my N20 doesn’t blow up, that seems to be the only major issue I heard about it, otherwise seems to be a solid engine, not even reports about it burning oil.

nermal > BLCKSTRM 09/07/2016 at 09:49

0

I think cost-per-mile is a better indicator than cost-per-year.

Driving a vehicle 30k miles per year will have a different cost structure than driving 5k miles per year.

Also, what if you take into account quality of parts, and who provides service? Using the cheapest possible mineral oil vs Mobile 1 or equivalent synthetic?

How about tires? Summer / winter tire set vs all seasons? Off-brand generic vs Michelins?

Out, but with a W - has found the answer > Party-vi 09/07/2016 at 12:21

1

Sure, but then again, that’s the case with most modern cars. Does BMW have short and long intervals depending on yearly mileage/type of driving, or is it 10-15k for everyone?

The article is a far cry from anything even representing a valid, scientific statistical analysis, but if every sample is treated in the same, possibly incorrect way, the trends should stay similar. The numbers will be off, but the order should remain about the same?

BLCKSTRM > nerd_racing 09/07/2016 at 12:29

0

You sure that’s not the window sticker? My oil change light lasts at least 9000 miles on 8 quarts, sometimes much longer.

Party-vi > Out, but with a W - has found the answer 09/07/2016 at 12:31

1

The study should have been titled “Average cost of oil changes over the lifetime of a vehicle”. In no way does one metric represent the average cost of maintenance on any of the marques in the study.

BLCKSTRM > AMGtech - now with more recalls! 09/07/2016 at 12:38

0

Your first one was the one that got me going and is the one that still bothers me the most.

A 15,000 mile interval is at least 12 months for 95% of the population. And as far as I know most BMWs use 7-8 quarts, which is double the amount for a “normal” car - but also double the change interval.

I guess what gets me so bad is that it seems like YourMechanic.com should know better.

I’d make some snide remark along the lines of “remind me not to go there” - but I don’t take my cars to anyone anyway.

BLCKSTRM > TheBimmerGuyWhoNowOwnsAChevy 09/07/2016 at 12:42

0

Check some of the other comments - most BMWs use twice as much oil (7-8 quarts) and last 10,000-15,000 miles. Which is at least a full year, maybe longer.

Shitty assumptions yield shitty results.

Out, but with a W - has found the answer > Party-vi 09/07/2016 at 12:45

2

It’s all really confusing. The original article actually has some interesting metrics, like ‘unusually common issues’ and ‘failure to start’ rates, but information about the underlying data is missing.

And then you’ve got the above graph, which seems to indicate that they took costs other than an oil change into account, since oil consumption doesn’t increase that spectacularly during a car’s lifetime.

And this table’s estimates “include expensive one-off costs, like a transmission rebuild, that skew the mean higher”. But the 328i is cheaper to maintain than BMW’s average, so ¯\_()_/¯.

Party-vi > Out, but with a W - has found the answer 09/07/2016 at 12:47

1

I find it amazing that the cost to maintain an A4 Quattro is less than a fucking Cobalt. Good lord.

DynamicWeight > BLCKSTRM 09/07/2016 at 12:57

0

The worst, for me, is going by brand instead of model. Some cars in a brands line up will cost a lot more than others.

TheDudeAbides_(version 2.0) > BLCKSTRM 09/07/2016 at 13:23

1

My BMW only takes 6 quarts, but easily goes 13,000 to 15,000 (once over 20,000!) between changes without a single quart burnt (she’s ticked off a total of 135,000 miles in her 10.5 years of service).

I’m pretty sure I said as much in the comments of that article when it came out originally. It seemed BS from the get-go.

ETA: Am I the only “greyer” left on Oppo?!? It’s lonely down here. Does it matter that I’ve been graced with ‘out-of-the-greys” status on Jalopnik (not sure why, but it sure is nice)?

AMGtech - now with more recalls! > BLCKSTRM 09/07/2016 at 13:34

0

Agreed.

Mercedes does flexible services depending on conditions for much of our stuff, usually equates to one year. Many of our newer cars only go by time, 365 days. Oh, and our engine/oiling issues are almost non-existent.

I feel like most people who drive high line cars less than ten years old either service at the dealer or independent specialists, rather than a shop that works on all makes. So I doubt yourmechanic.com really knows as well as they think they do. And their data on those cars is probably painfully incomplete and full of outliers that skew the data.

nerd_racing > BLCKSTRM 09/07/2016 at 13:51

0

2011 128i for my region is programmed for 5000 once the dealer included maintenance runs out. Apparently it has more to do with climate than anything. I routinely do 7500 miles on mine. 10k is too much but 5 is ridiculous. I just reset it for quite a few changes in the stalk menu and then do the changes by mileage old fashioned way.

nerd_racing > Out, but with a W - has found the answer 09/07/2016 at 13:55

0

so according to this it costs roughly $3000 more over 10 years to maintain the 328i than a Mazda 6. If you break that down to monthly payments you get $25 more a month. I would believe that if you factor in needing LL01 rated oil and probably 1 to 2 quarts more of it per oil change. Also the maintenance computer in the BMW tells you absolutely everything as it nears, things like wiper fluid low (yes it will tell you every time you start the car) and if you wait long enough it pops up the service icon telling you to take it in so the dealer can refill your washer fluid. Right now I have one up for brake fluid flush. I’ve got to get out the pump and swap that out eventually.

e36Jeff now drives a ZHP > nerd_racing 10/06/2016 at 22:04

1

5 qts of LL01-approved Mobile One is $22 on amazon normal price(but its out of stock right now), and you can get a rebate from mobile 1 for $12/5qt jug right now, making 10qts $20. I’m about to do a $27 oil change on my ZHP.

Also, I think the survey is failing to take into account the fact that for the first 3 years/36k miles all maintenance on a BMW is free from the dealer.