Slashdot Mirror
Consumer Reports Expects Tesla's Model 3 To Have 'Average Reliability' (cnbc.com)
There may be only a few hundred Tesla Model 3s on the street, but Consumer Reports already has an opinion on the new car's dependability. From a report: "We are predicting that the Model 3 should have about average reliability," said Jake Fisher, director of auto testing for Consumer Reports. Average may irritate Tesla fans and the nearly 500,000 people who have reserved a Model 3, but Fisher believes people should understand what Consumer Reports expects from the new car. "We don't go around recommending that people buy cars that are below average, so if it is average or better, that is not a bad thing at all," said Fisher. "But let's be very clear, we are not giving it super high marks. We are saying it is basically par for the course." Consumer Reports has yet to buy a Model 3 and put it through a battery of tests, as the magazine does for dozens of vehicles. In addition, so few Model 3 cars have been delivered that Fisher and his team have yet to get a sense of how owners feel about their new Tesla.
How dare they not rate it as a perfect example of motoring!
pay to play.
As a Model 3 fan, I'm actually hear to say that I find it weird that you can rate the reliability of a car you've never even touched and which nobody has had on the road for any length of time, and is based on an entirely new platform from a manufacturer's previous vehicles.
Nothing, more, nothing less. Just strikes me as odd.
I'll BUILD someone to replace you. Some kind of gamma-powered monster, with a heart as black as coal!
How can you possibly rate anything not produced ? I know that Tesla is inside the Reality Distortion Field. Jobs left it to Musk in his Will, but how can you rate a car in beta, er, pre production ? Do CR writers have some Tesla in the 401 (k) ?
Yes this is weird. Any journalists that "pre-report" what they expect their future articles to say, have some serious integrity issues.
sounds like GM sponsored article. At least it is made in USA!
As a Model 3 fan, I'm actually hear to say that I find it weird that you can rate the reliability of a car you've never even touched and which nobody has had on the road for any length of time, and is based on an entirely new platform from a manufacturer's previous vehicles.
Nothing, more, nothing less. Just strikes me as odd.
What's even weirder is they are using the term 'predicting' and not 'rating'. Why would they call it a prediction if it were a rating?
But, yes, you can use past performance information from companies and the fact that they have admitted production line issues to predict the reliability of the first x number off the line. How accurately you can do that is up for debate, but that's what CR does.
Consumer Reports stakes their reputation on their reviews being above reproach on an ethical basis. They don't accept freebies from manufacturers. They don't use affiliate links. They don't accept sponsorships. Instead, they buy all of their products from the same stock that any other consumer would (rather than the hand-picked ones that oftentimes get sent to reviewers) and they make their revenue by charging people a fee to have access to their content. Sadly, in the Internet era, that business model has pushed them towards clickbait headlines designed to increase their membership, as evidenced by their very public-yet-baseless jabs over the last few years at whichever companies are popular (e.g. Apple, Tesla, etc.).
This is yet another of those jabs designed to drum up revenue. They don't even have a Model 3 in their hands yet, so when they say, "let's be very clear, we are not giving it super high marks", what they're actually saying is, "we have nothing meaningful to say at this moment, and we expect that the actual review we post won't make headlines, so instead we'll say something outlandish about the popular product now in the hopes that some suckers will sign up to read our final review". They're certainly not faithfully performing their duty to review things in an impartial manner based solely on the facts. Rather, they're sacrificing their integrity for the sake of a quick buck, as has sadly become par for the course with them.
Whatever reputation they still had in the circles I move in died years ago.
It's basically an indication from CR that Tesla should go ahead and pay them some money for better reviews. You can see that in a number of reviews they do on anything from cars to computers, they are the ones "recommending" Microsoft products like the Surface, until widespread issues came up that any tester would've found and after enough publicity they pulled their recommendation.
Consumer Reports is to consumers what Gartner is to businesses, at least Gartner is up front about the $250k startup fee.
Custom electronics and digital signage for your business: www.evcircuits.com
It's basically an indication from CR that Tesla should go ahead and pay them some money for better reviews.
FALSE
Consumer Reports takes ZERO money from companies & product they review.
Sure, you can disagree with their opinion, but their opinions are entirely their own.
>"Consumer Reports has yet to buy a Model 3 and put it through a battery of tests, as the magazine does for dozens of vehicles"
And even if they did, how are they supposed to know how "reliable" it is based on that? A day? A few weeks? A month? Knowing RELIABILITY comes with at least many months, if not years, of use and observation. The ONLY thing they can do is SPECULATE based on their hunches about the technology and materials, and SPECULATE based on OTHER models. So hopefully people won't freak out over being called "average" without any real data to back up such a claim.
Hahahahaha
Holy shit you actually believe that...
I see what you did there...
"So how can Consumer Reports predict the cars reliability?
"What we have is a lot of data from the Tesla Model S," said Fisher. "That gives us a little more confidence in the Tesla Model 3."
In addition, Fisher expects the simplicity of the Model 3 compared with Tesla's other vehicles means the car should have fewer problems."
they are saying it should be more reliable than model S...
Consumer Reports gives each car it tests a score that falls within a scale of zero to 100.
When the 2013 Model S came out, it got a score of 99 -- which has never been exceeded before or since -- with one Consumer Reports representative gushing that "If it could recharge in any gas station in three minutes, this car would score about 110." Ridiculous to say, when the scale only goes up to 100. But such were the Tesla goggles voluntarily donned by the staff of CR back then.
That that is is that that that that is not is not.
To improve the accuracy of their prediction they will survey 1000 people, each of whom have never seen a Model 3, and average the results. Seriously, given that they have no data to work with, betting on "average" is probably they best prediction they could make.
on a "car you've never even touched and which nobody has had on the road for any length of time, and is based on an entirely new platform from a manufacturer's previous vehicles."
If I read the article correctly, Consumer Reports has not tested any Tesla Model 3 yet.
Why do they have the need to predict anything then? They have no data at alll.
Your bias is showing. The fact that you jump on the anti Elon bandwagon is so very telling. We're talking the Model 3 here, not Tesla as a whole, and furthermore not Elon.
I think Musk and Co. are a pack of dillholes, but I expect better from CR than speculation. If you haven't tested it and torn it down the don't give me predictions, assholes.
You make prediction, a projection of how satisfied you will be with a future purchase of a new Tesla model, based in part upon current owner's experience and reviews of prior Tesla models.
When Consumer Reports advises its readers on what to expect their satisfaction with a new Tesla model will be, influenced by what Consumer Reports values in cars and other products they report on, they are "making stuff up"?
As a grandma yahtzee, I act in fear to say that I find it weird that you can deflate the re-hire-ablility of a word you've never understood and witch nobody has had on the frog and toad for any breadth of time, and is based on all your belong to the bomb that somebody set us up.
Odd indeededoodee
Of my four cars, one is a Leaf. I have never had to change the oil, or a belt, flush various fluids, replace a spark plug, catalytic converter, or alternator. I have never had to replace the transmission or get a tune-up of my engine.
Why? Because the Leaf lacks all of those things. The basic components of the car are a motor, controller, and a battery. The motor is AC brushless, the controller is solid-state, and the battery is under a 10 year warranty. By the way, mine has dropped only one bar in 30,000 miles.
All that I have done to the car in five years is change tires, fill the windshield washer, and replace the wiper blades. At some point, I will have to spend $5000 for a new battery, but it will be well worth the savings. My SUV goes through $240 in gasoline per month which is over $5000 in 21 months.
So, Consumers Report, what parts will break on any Tesla that would give it average reliability?
Consumer Reports has done this type of "reporting" for decades. It appears to me that they develop a policy on a subject and rate products on how they perceive the products adherence to CR's policy instead of how the policy works. It also seems to me that Telsa is becoming a company to hate -- like IBM during the early PC-era and Microsoft during the mid years of the same era. I noticed this type of behavior with CR and realized their reviews were valueless to me and stop my subscriptions. I rarely refer to CR any more and rarely give any credence to stories that reach the regular-type news media.
I find it weird that you can rate the reliability of a car you've never even touched and which nobody has had on the road for any length of time,
I suppose a lot of people here have not had much experience in manufacturing products. At my companies we have often had to declare to prospective customers the MTBF (Mean Time Between Failure) of products we hadn't gotten to production yet.
Typically one of our circuit assemblies (with large chips) will have an MTBF of 50K to 100K hours these days. Certainly we can't run a bunch of them in the test lab for that length of time to determine this. The product would be too old to ship before the tests were done!
The answer is that there are a number of accepted models (with vendors of software for them) that are used to do this. Telcordia/Bellcore, MIL-217, and so on. You gather data from the each of the components (resistors, LEDs, chips, DC-DC converters, etc.) and put them into the model and it calculates the predicted MTBF result. OEM customers will accept that value as long as you worked the model correctly.
It is actually kind of interesting to play with the model to see what happens when you change the product. If you add a fan component or a rotating-media disk drive you see a big impact in the MTBF.
Can you do this for a whole car? Yes it is possible but some components won't have MTBF figures established. For that you can substitute similar parts and make a judgement call. I would be surprised if Tesla hadn't done this already but I would also be equally surprised if CR had. Not in any rigorous way at least. It is a lot of work.
So I am pretty sure that they are just doing SWAG on most of it which I wouldn't trust much. Sure Tesla models do have their problems because they are damn near prototype units they are shipping. Things go wrong with them that has to be fixed but Tesla takes that in stride and fixes them. But it doesn't make for good-looking reliability predictions.
One thing about CR that I had noted ever since I started reading them back in the '70s. When it comes to American-made cars they always seem to have a bias against them. If Dodge or Chevrolet does something that Mercedes or BMW or Honda do they will shave the score off the American car but not the foreign car. I think they are better about this than they used to be (decades ago when I last paid attention) but it was very pronounced then and they could be taking it out Tesla now.
It's marketed to people that Tesla's sociopathic CEO sees as "poor"; people who can *only* afford $35,000 for a car and so like other "poor people" cars it will be designed to fall apart after it goes out of warranty, trapping its buyers in the never-ending loop of transportation as a service. Because what are they gonna do, go buy a more expensive car?
Tesla will always have the credits for starting to make the first usable electric cars, but now that other manufacturers are making electric cars they are getting way overhyped. There are already cars out there that are put together better and that I would rather have than a Tesla. So let's please not worship the company.
-- Cheers!
That's extremely optimistic given the quality of existing Tesla products.
What? It's actually the rich people cars with the more complicated equipment that tend to have the expensive maintenance. My '98 Ford Escort still runs great and cheap.
And Consumer Reports said in the article that they expect the Model 3 to be more reliable than the Model S, for similar reasoning.
This space intentionally left blank
Only if they report it anything other than average. In the absence of data average is a pretty good estimate.
I think it's more of a reaction to:
A) The hype surrounding the M3, and the hype surrounding MS/MX failures
B) The fact that they gave the MS really high marks initially but then had to reduce them due to reliability problems
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
It should be very obvious to all but the most blinkered that the model 3 has and will continue to have a lot of issues with quality control until they sort their production out and discover how their car breaks in the field. That is going to take a few years and the chances are the vehicle will be reliable after that.
As the adage goes, never buy version one of anything. This is true for software and should be double underlined and highlighted for cars.
Even after you test a Model 3, how do you rate its reliability against any car with an internal combustion engine?
Take a look at the top 10 car repairs of 2015: (From http://blog.credit.com/2016/04...)
Replacing an oxygen sensor – $249
Replacing a catalytic converter – $1,153
Replacing ignition coil(s) and spark plug(s) – $390
Tightening or replacing a fuel cap – $15
Thermostat replacement – $210
Replacing ignition coil(s) – $236
Mass air flow sensor replacement – $382
Replacing spark plug wire(s) and spark plug(s) – $331
Replacing evaporative emissions (EVAP) purge control valve – $168
Replacing evaporate emissions (EVAP) purging solenoid – $184
The model 3 doesn't even have any of these parts.
This is unfortunately true I fear. I feel they have been struggling recently to stay ahead of the curve, as companies have moved beyond yearly product updates and product lines have expanded beyond any reasonable attempt to categorize them. The last few times I have looked at their reviews for appliances, for example, I've found that the models they reviewed a few months earlier have been supplanted by new ones. It's a shame because for some time they were an amazing resource.
As time goes on, "average" reliability continues to improve across all brands. CR explains this quite clearly in their yearly auto issues. ICE's, as complex as they are, now go 100K+ miles without requiring any major service. Contrast that to, say, the 70's where you might have valve jobs, carburetor rebuilds, spark plug wire replacement, and a bunch of other stuff I'm likely forgetting to make a car of "average" reliability. In terms of the Model 3, it's a new model, at production levels Tesla has never seen before, with new tech, new parts, etc. Even automakers who are venerated quality leaders like Toyota and Honda see a dip in first-year models of a significant redesign. The first Model 3s are likely to have teething problems as they spend some time in the real world, you can only test so much. This doesn't mean it's a crappy car, it happens with every new model from any manufacturer. It also happens with phones, software of all types (games, open source projects, commercial software, operating systems), new houses, new office buildings, etc.
A few problems here:
1. You mentioned spark plugs twice. If you have coil-on-plug design, you replace them with the coils, else you replace them with the spark plug wires. It's one or the other.
2. You mentioned coils twice. See above.
3. If you get your car repaired at the stealership, you pay the prices you mentioned. That stuff is a lot cheaper at an indy mechanic, cheaper than that as a DIY.
4. I drive a BMW 528, 9 years old with 120,000 miles. Of all the repairs you mentioned, the ONLY one I have EVER had done is coils + plugs, for about $350 on a 6-cylinder car.
5. Had a battery replaced in my car for about $150 (massive battery, complicated by German engineering). My guess is the Tesla battery is far more expensive
Given no data points whatsoever, "average" seems a reasonable data point for the quality of a manufactured item. It is, by definition, the mean of the field.
Given that an electric car has dramatically fewer moving parts than an IC car, and that the technology for charging and motor controls is very mature, I'd expect above average.
Given that the company is manufacturing their own battery cells, a key failure point, and they have not been in that business for very long, I'd expect below average.
Given that this is the first model year from a company that has never produced vehicles at this scale, I'd expect below average.
I want a Tesla 3, and will buy one before the end of the decade if I can. That said, there are a fair number of unknowns that make a prediction beyond "average" an open question.
That was ages ago. CR is being quietly getting paid for ratings sin around 2009.
Take a look at the ratings and you can actually see who and when somebody started paying for the rating. You can actually see models from company X jump from the bottom of a list to the top ten from one moth to another.
This is very unresponsible of them, they haven't seen or tested the car, so how on earth can they even remotely comment on the reliability of the car. It seems to me, they are provoking Tesla into giving them a car to test.
It is my understanding that CR gets reliability data by surveying actual owners. But there are virtually no Model 3 owners right now, so that's out of the the question.
They buy the cars they test on the retail market to make sure manufacturers don't game the system by giving them a special model, so testing one is also out of the question right now.
I don't always use unix-like operating systems; but when I do, I prefer FreeBSD.