Elon Musk's Data Doesn't Back Up His Claims of New York Times Fakery

This whole controversy has been a little depressing to read...not that the Tesla vs. NYT discussions here have been worse than on other forums, but just because it shows how technical minded people are as easily swayed by preconceptions and alliances as more ostensibly non-scientific minds.

How many words have been expended in the other HN thread to allege that Broder -- after most have already established that he is a charlatan -- is receiving oblique funding from his Big Oil paymasters? It may very well be that Broder got a swimming pool full of BP-money in his offshore hideaway...but isn't it possible that just maybe, that Elon Musk has a vested interest in advocating for Tesla? Like, just a little bit?

It doesn't have to be that Musk is trying to cover up the truth. It could just be that this is his big project and he is overly sensitive to (some of it admittedly unfair) criticism to the point where he'll see malice where there is none. It's possible: bias from sentimental influence is not unheard of in the scientific community.

One of the most disappointing things about Musk's response was how he closed it with an out-of-context anecdote:

In his own words in an article published last year, this is how Broder felt about electric cars before even seeing the Model S: "Yet the state of the electric car is dismal, the victim of hyped expectations, technological flops, high costs and a hostile political climate.”

If you read that article, Broder was clearly referring to the controversy behind the Chevy Volt, which he also compared unfavorably to a "lawnmower".

Oh wait, that was Elon Musk who said that: http://articles.businessinsider.com/2009-05-11/green_sheet/2...

So basically, if you think Musk knows what he's talking about, then Broder spoke the truth about the Volt. Yet Musk uses Broder's assessment as a closing statement of damning circumstantial proof that Broder is decidedly anti-electric car.

Oh I know, this kind of cheap rhetorical trick is what all politicians and businessmen do, and it's OK if someone we all really admire does it, as long as his heart's in the right place. Maybe so, but I don't think it hurts to be a little more objective towards our heroes and realize that they can be prone to misjudgment too.

The "Norwich Charge" data point is all you need to understand what happened here. Notice how vague Broder is in his original account:

"After making arrangements to recharge at the Norwich station, I located the proper adapter in the trunk, plugged in and walked to the only warm place nearby, Butch’s Luncheonette and Breakfast Club, an establishment (smoking allowed) where only members can buy a cup of coffee or a plate of eggs. But the owners let me wait there while the Model S drank its juice. Tesla’s experts said that pumping in a little energy would help restore the power lost overnight as a result of the cold weather, and after an hour they cleared me to resume the trip to Milford."

At every previous charge, he noted the exact mileage remaining when he headed out. Why not this time? Because the estimated range when he left that charging station was less than he needed to reach his destination and he knew damn well that that was the case.

http://www.teslamotors.com/sites/default/files/blog_images/r...

(How he convinced the Tesla people to "clear" him at that point is a mystery that hasn't been followed up on by either party.)

Later he hand-waves past this by saying he was testing the superchargers, not normal chargers. But the fact is he knew he would run out of charge midway to his destination if he didn't charge the car longer. Instead, he decided to intentionally set up a situation where the car would be stranded.

Elon might have been better off concentrating on this point more, rather than some of the smaller inaccuracies. But the data is clear: the reporter was in full control of the situation and manufactured the failure himself.

I guess it's time to muddy the waters. The truth is in the middle!

Of course, the entire debacle can be summed up with the following exchange:

Q: "How many miles did you have left to drive on the final leg?"

Broder: "61"

Q: "And how many miles did your car tell you that you could travel before recharging?"

Broder: "32"

Of course, Broder claims that Tesla told him it was OK to leave. I think this is bullshit, but will never know. The fact is he basically was stupid enough to essentially leave a gas station with half a tank when he had a full tanks worth of distance to travel. The only way you would do this is if you want to get stuck. I don't care how scientific you are trying to be, you wouldn't risk being stuck out in the cold with a dead car unless you really wanted that to happen. Why we should take anything this man writes about cars seriously is beyond me.

"Convincing? No. His other chart shows the mile range dipped below zero, which would indicate the car could not move. "

Wow, I wish I could be as sloppy and awful at my job as these journalists are. Had he actually bothered to truly read the response he is commenting on, he would see that Musk states that, when the car's range dips below zero, it is running on reserve power. It takes some miles after "zero" for the car to actually stop moving.

What a hack. There is truly nothing worse than reading something by someone who has no education in comprehending quantitative data try to critique it.

However, I do agree with his statement on the climate control. That was something I noticed when I read Musk's response. Clearly, the Broder turned the heat down, just not as early as he thought/claimed.

"Tesla Motors initially expected to sell at least 5,000 units in 2012 and set a sales target of 20,000." (1)

As of January 2013's run rate, it takes GM 72 hours to sell that many cars. (2)

How about a comparable luxury target?

MB did 305,072 vehicles in 2012, so they do Tesla's sales every 2-weeks. (3)

Perfect comps? No way, but still . . .

To call Tesla an "early adopters" brand where you're going to run into some bullshit, is an under-statement.

I was considering a Model S as my next vehicle, and even if the NYT article is accurate that dissuades me zero percent. I get that this vehicle is going to have some problems - BUT IT'S ELECTRIC AND BIG AND SEXY AND FAST AND NOBODY ELSE HAS ONE!

But watching Musk's response to going full-Nerd-Nuclear (without CONCLUSIVE data) to trash a journalist and the institution of the NYT actually does give me pause in wanting to support that brand.

In other words, in my personal case (of one person, one piece of data) Musk is damaging Tesla's goodwill more than he's helping it.

(1) http://en.wikipedia.org/wiki/Tesla_Model_S

(2) http://www.gm.com/content/gmcom/home/company/investors/sales...

(3) http://www.cleanmpg.com/forums/showthread.php?t=46221

> Musk accuses Broder of thinking "the facts shouldn’t get in the way of a salacious story" (which is an odd choice of adjective for a car review)

The story was "Electric cars suck". Broder is well known for having anti-electric car opinions, so it's not surprising that he would want to write a negative review.

Argument 1: The length of the detour is not as important as the type of driving during that detour.

Argument 2: "But Broder never claims he turned down his heat at the 182 mile mark." - well, he does claim he turned it down and doesn't mention he had turned it up. People can lie by omission.

Argument 3: "But the section before he broke down it does look like he chugged along at a pretty low speed.", perhaps. But not at the 45 mph claimed in the article.

Argument 5: "No. His other chart shows the mile range dipped below zero, which would indicate the car could not move." The car has a small reserve. 0 charge would mean "charge the car right now".

Argument 5: "Broder also explains that he did not charge fully because of the time it took to charge. He wanted to show the real world experience of a real driver, who might not want to endure the hour and a half it takes to charge up, when only needing a certain amount of energy to get to point B." I hope any real driver would know that a distance of 60 miles would need enough charge for a 60 mile journey, and that giving 30 miles of charge would be a stupid thing to do. Just as if you needed to drive a distance needing 20 gallons you'd be stupid if you only filled with 10 gallons.

This article is, frankly, idiotic.

All it takes is one person to drive the same journey, but with proper charging, and to release the logs to disprove Broder's points.

Article makes a few good points. There has been a lot of "Hurry for data! Boo journalists! Off with the oil shill Broder's head" talk here and elsewhere, but Tesla's data seems too far from an ironclad refutation of the article. The supposed circling would have lasted less than five minutes at 10-15 mph. It is a strictly subjective "he said, she said" situation about whether Tesla did tell Broder that he was good to go in Norwich. And unless the towing company is lying, the car did shut down, contrary to Tesla's claims.

There is no clear winner or loser here.

If you need to bring in graphs to show that someone was going 50 instead of 45 MPH (if the data even matches the speedometer), you've already lost the battle. This is all very aggressive on Musk's part when the overall review still holds up (Tesla cars are bit ahead of the curve for normals). If the log showed gross exaggeration I could see this reaction, but Musk is being a brat and wasting a lot of goodwill. Remember, car reviewers are car reviewers, not QA engineers. Things will go wrong, numbers won't match up exactly. If you want positive coverage, don't post your MPG (or the electric equivalent) with "best case" numbers like a normal gas car. Give yourself some wiggle room. Calling the flatbed tow truck is a pretty bad negative event for a normal car owner who is used to being able to ride on E for a few more miles.

I thought the most convincing part of Elon's post was his 3rd bullet point, which relies on no data outside of the article, and which this Atlantic Wire journalist ignores entirely:

"In his article, Broder claims that “the car fell short of its projected range on the final leg.” Then he bizarrely states that the screen showed “Est. remaining range: 32 miles” and the car traveled “51 miles," contradicting his own statement (see images below). The car actually did an admirable job exceeding its projected range. Had he not insisted on doing a nonstop 61-mile trip while staring at a screen that estimated half that range, all would have been well. He constructed a no-win scenario for any vehicle, electric or gasoline."

The car said it would go 32. It went 51. Broder claimed this "fell short". Clearly it didn't.

The most interesting point on the range graph isn't mentioned in Tesla's repsonse (or anyone else's), which is the huge drop in range overnight at mile 400, fron 80 miles of range to approximately 20. That is the Achilles' heel of the Tesla that caused the car to run out of juice in this (possibly contrived) example.

This article doesn't back itself up. Claims are labelled as 'False' and the evidence presented is 'Google Maps says so'.

Maybe Musk's data doesn't add up (I doubt it). If that's the case, present something other than 'I plotted a route in Google Maps and it said 500. QED.'

It has to be asked, given the tone and rapidity with which this article went up: what relationship does Rebecca Greenfield (or maybe the editors at The Atlantic) have to Broder? This really feels like a "got your back" kind of thing. Media figures don't, as a general rule, write stories with the sole purpose of knocking down criticism of other media figures.

Elon's complaints have been silly from the get-go.

When a company reviews your car, they don't use the standards which you want them to use. They don't do things precisely as you want them to. Reviewers make mistakes, they are new to the equipment, they don't know everything about your car. And you know what?

That's a great thing.

Electric cars, specifically, are unique products. They are novel, new (well, technically they've been around for a hundred years, but haven't been common for the past eighty or so). People don't know what to think about them. Average Joe Sixpack from the NYT treating this car as a normal driver would, and providing us his opinion, is a godsend.

This "defense" basically amounts to: See, Broder didn't lie, he exaggerated. That, and the things Broder did to try and get stranded were fair game.

In that, they are making valid points, supported by facts. That it's the best they can do is telling.

I look forward to replication attempts.

Elon's own arguments are absolutely killing the credibility of electric cars:

-Turn the heat down for more range? Who the heck wants to drive around cold in an extremely expensive car?

-Drive slowly? Again, what?

Elon is simply agreeing with the reviewer's most damning findings: The consumer must make considerable sacrifices - in addition to spending more money - in order to properly use their Tesla car.

> Where Google thinks it's located—we don't know the charging station's exact location

No, we do. The photo here is of the northbound charging spots: http://wheels.blogs.nytimes.com/2012/12/21/tesla-begins-east...

Assuming he took the car lane, even if he drove past every single parking spot all the way back out to the road to the highway, he would only cover 0.4 miles. http://goo.gl/maps/VzRjc

The headline should be updated to begin with "The Atlantic Wire claims..."

Do people really think that Broder lied about the car dying?

My bet is that John M. Broder made a call to James Bennet (the third person Broder ever followed on Twitter. They have published many things together. e.g. http://www.nytimes.com/1999/02/13/us/president-s-acquittal-w...), who is the Editor-in-Chief of The Atlantic.

James hasn't tweeted since February 8th, but retweeted John Broder today about 9 hours ago (screenshotted for posterity in case someone following up to blog about this wants it later) https://twitter.com/JBennet

This now calls in question the integrity of James Bennet and Rebecca Greenfield. I haven't been able to determine a link between Rebecca and James other than that of subordinate. She's been at the Atlandic since July 2010 and he's been at the Atlantic since 2006. http://www.linkedin.com/profile/view?id=43905376 https://www.facebook.com/pages/James-Bennet/112357645448306

I tried finding the Facebook profiles for these three people to confirm connections, but it doesn't look like their profiles are publicly discoverable.

Any journalists care to take on a story of examining the relationships within the third estate here? There's probably a story.

I have the feeling that Musk shot himself in the foot with his original blog post. He accused Broder of a host things, that now are being reduced by some to seemingly critical points to the story while others debate minor points like the point at which the AC was turned on and if Broder missed the charging station or was trying to empty the battery.

I think it boils down to this:

1) Does Broders trip represent a typical experience driving a Tesla Model S?

2) Could Broder have avoided running the car down to empty and having it toed?

I'm sure there are tons of small nuances and subtle details here, but at the end of the day I'm not sure they matter. The ultimate question is whether or not the NYTimes reporter presented an honest story, and that appears to be far from true. it certainly looks like he went out of his way to try to manufacture a story that isn't reflective of actual experience. But even if that's not the case there's more than enough evidence that he twisted the account and left out crucial details in the service of titilation and excitement rather than factual accuracy.

Musk's job is to sell his cars, we accept that he might present things in a manner that is most flattering to his products. That's ok as long as he doesn't outright lie or mislead, we wouldn't expect anything else from any other company. But the reporter's job is to present the truth to his readers, and if he not only fails at that job but goes out of the way to present falsehoods what faith can his readers have in him? At this point, what makes him better than a reporter for a tabloid? Why should anyone believe anything he writes? If a reporter doesn't have a reputation for factual accuracy what good are they?

P.S. I'm reminded of Dateline NBC's "investigation" of the GM C/K series pickup trucks. The trucks were fundamentally of an unsafe design, with side mounted external gas tanks. But Dateline's crash tests included adding model rocket engines to ensure that the rupture of the gas tank caused an explosion, without revealing that to the viewers. And that, I think, is the problem here where the reporter was not content to let the facts stand on their own and went to every effort to juice up the story (although not the car). Investigative journalism is a hard job, but when you make that leap and start adding embelishments you are no longer doing journalism you are entertaining.

This article has confirmed crucial aspects of Broder's version with third parties, such as the truck driver who picked up the inert car (it doesn't matter if the battery still has 28% power if the parking brake won't turn off, iirc Priuses also refuse to discharge their batteries below a critical threshold), and used Google Maps to verify other issues (finding the Milford supercharger).

I haven't seen anyone attempt to verify actual speeds based on computing the area under the speed graph, but I suspect that this won't help Musk's case. 19/22 (ratio of the wheel size of the car driven and the wheel size specification) is actually the perfect ratio to explain the variation between recorded speeds and reported speeds. Perhaps one is based on GPS and another on wheel rotation.

The correct response to this story was: we need to improve our customer service, software, reporting of range (especially w.r.t. temperature) and messaging; customers need to better understand the tradeoffs of gas vs. electric cars, and not claiming a senior NY Times reporter will lie to generate some extra hits on a pretty minor story.

Musk's rebuttal is too wide reaching. It makes some accusations that the evidence supports, but does not prove. This leaves his rebuttal open to a counter rebuttal, when the whole issue revolves around one facet of the data:

The reviewer left Norwich with the car telling him he could travel 32 miles, when he needed to go 62 miles. If he hadn't done that, he would not have had an issue.

The atlanticwire article doesn't address this point at all, and how could it? The car said it would go 32 miles. It went more than that, but less than 62. Any reasonable person, needing to go 62 miles, would want the number the car says to be greater than 62, because what kind of idiot tries to go 62 miles with a car with 62 miles of range left? Now, what kind of full on retard tries that with 32 miles left?

Am I the only one who feels like Musk is doing more to soil the reputation of electric cars than to advance it? It totally sucks that the media is being a bunch of meanies about this stuff, but in many ways, Teslas still aren't practical for unwashed masses.

Also, their target market here will probably research this enough that the negative PR won't matter that much. It'll ruin public perception...just like they've done with diesel over the past decades. Now where's my diesel/electric car? Diesel/electric hybrid technology has been around since the first few years of the 1900s: http://en.wikipedia.org/wiki/Diesel-electric_transmission#Sh...

I am not impressed with our "progress".

You know all this Tesla talk makes me actually want to own one now. Sure there is a lot of talk about mileage and charging stations but on the whole it is going to get better. And the more discourse we have the better chance the product will be given a good deal of attention.

The articles about this event clearly explain in my opinion not who is wrong or right here but how much even the most smart reader's opinion can be guided or misled by what a specific author writes, whether by omissions, incomplete data or other rethorical techniques

This argument has nothing to do with facts, it is just one journalist attempting to defend another, and you can tell by the language used. For eg. in Argument 3 about cruise control and speeds, she says:

But the section before he broke down it does look like he chugged along at a pretty low speed.

"It does look like"? This is a graph, it has data points, its not about "looking like" - these points represent facts. At no point in the facts is there proof of cruise control being set at 55mph.

She is taking a graph with real data and attempting to re-interpret it to fit the NYTimes story. This is an emotional argument, as with 'oh no, this was actually about testing supercharge stations'.

Why not simply doing a retest? Seems easy enough, it's still winterly cold …

I really want to buy an electric car, but range anxiety keeps me from doing so. Unfortunately, the NYT article doesn't help.

The particular facts may be "peculiar," even fabricated, but the overall experience rings true. There are going to be trips that you can do easily today on petrol that would be difficult to do with the state of the art electric car from Tesla.

It's time for Tesla and Better Place (betterplace.com) to join forces.

This is a huge PR win for Tesla, what started out as one NYT review has blossomed into 4 or 5 articles and a he said she said drama.

It's interesting that one of the results of Musk's reaction might be for the next journalist to bring a camera crew. It wouldn't have to be a Top Gear-style hit piece: just record all the phone conversations with Tesla personnel trying to figure out why things are not reading consistently. It's not going to leave a favorable impression.

I've got a stupid question. Why doesn't some impartial soul(s), try to recreate exact conditions as Broder describes them in the article? Tell Tesla what you will do, film your trip there. If their experience doesn't match Broder's they should just abuse the car until they recreate the exact condition Broder ended up.

Regardless of all this, Tesla should look hard at a mapping solution that ties with their onboard sensors - the car can intelligently tell you where to stop and charge and how long to charge for by knowing where you are trying to get and what charging stations are along the way.

Elon's argument should have probably been more focused. That doesn't change the fact, though, that there are numerous pieces of evidence showing that Broder expected unreasonable things out of the car, and then blamed the car when it failed.

Shit, this is disappointing. And I was hoping we could announce a villain in this story before the weekend -- and now you tell me that it is more complicated than that?

Now I have to go back to being upset about there being horse-meat in lasagna.

Is Tesla allowed log data without informing the users? Will these logs also be generated by cars that people purchase?

I'm impressed that car runs as good as it does. Maybe there's hope after oil! Go Elon!

All The Atlantic Wire "Convincing?" points are rather lame to be honest.

A debate with DATA. How wonderful!

The problem I have with "he made it up": unless you have a computer keeping track of miles, speed and temperature, you are going to be "lying" about it. For example, I could easily say that I was riding at 67 mph but the computer point out that I actually went as high as 72, as low as 52 a lot of the time. Was I lying? Who the hell is going to remember or write about every speed change on a 500 mile trip? Was it a scientific study or a review for the average user to understand? But let's look at the big picture: I, personally, would NOT want a car that is so specific about the speed, cabin temperature and so on, or call a tow truck because the battery died out on you. 45 mph or 52mph, it's the same crap to me.

Tesla, thanks to the tracking, may have every tiny detail, but their users generally agreed with the reviewer--at a Tesla site forum no less.

Restaurant reviewer: Service was great but the food took an hour to arrive!

Restaurant owner: He is a liar, food was on the table with 50 minutes and I have order slips to prove him wrong.

Me: underwhelmed at the "evidence," 50min and an hour is essentially the same to me when waiting for food to arrive and I didn't expect the reviewer to hold a chronowatch.