Hacker News Reader

Comment by jackedEngineer | Hacker News Reader

I did some digging.The title is misleading -

YTD Tesla is way ahead - https://imgur.com/56eG7CN Actual Numbers - BMW registration for the month of July 2024 were up by 300~ cars.

Original Source - https://www.jato.com/resources/media-and-press-releases/bmw-... Quoting this comment -

streptomycin- Yes https://i.imgur.com/ZNv1AyD.jpg Although BMW is making some nice EVs and they are selling more than they used to, the headline is misleading.

ultrablack22 - Tesla produce a lot of for the EU in China. They have a schedule where for each quarter, months 1, 2 and 3, in Month 1, they produce cars. In Month 2 they sail them over from China. And in Month 3, they arrive and are sold.

BMW managed to overtake Tesla in the first month of Q3, barely. When Tesla really had no cars for sale.

View on HN

"The title is misleading."

OK, let's assume the title misled me. What do I now believe as a result that is incorrect.

Titles are not summaries. It is expected that readers will read the body of the submission not merely the title. Is the JATO Dynamics report accurate. The title is simply repeating an alleged fact contained in the report. Is it false.

If all a reader ever does is read titles, I am afraid "misled" is an understatement for how disllusioned that reader may become.

People defending Tesla and Musk always come across as try hard.

Titles may not be summaries, but you would be surprised to know how many people form their opinion about the body of an article based solely on the title.

Your statement about people defending Tesla and Musk always coming across as "try hard" is subjective. I could say the same about you, who consistently criticize Tesla and Musk as try hard, but I won't, because I prefer to focus on the facts rather than making generalizations.

I would have posted a similar comment if the submission title was something like "Samsung beats Apple in July sales by 300 devices" while the article revealed that, year-to-date, Apple had actually outsold Samsung by a 3:1 margin and if it wasn’t true. In such a case, I would consider the title to be misleading.

I chose to comment on this specific post because I am familiar with the automotive industry. It appears that the source cited in the article may be misleading because The data from https://eu-evs.com/bestSellersCharts/ALL/Brands/Bar/Year-by-... does not seem to support the claim made in the title.

In the future, instead of attacking the poster as "try hard," it would be best for everyone to read beyond the title, verify the sources, and engage in constructive discussions.

Announcing BMW sold more EV’s in a single month (and they are using registrations which means actual sales) is interesting on its own. Sure it might just be random fluctuation at this point, but that just means they are neck and neck.

IMO waiting for a full quarter may be more reasonable, but waiting for a full year and it’s old news.

    > they are using registrations
    > which means actual sales

No, you register a car in order to drive it on public roads in a given country.

Some purchased cars are never registered (track use, private export etc.), or defer the registration (no reason to pay road tax if you're not driving it yet).

I didn’t say they were guaranteed to be 1:1, but every registration is an actual sale. The point is the number of sales is going to be grater than or equal to the number of registrations. So this isn’t some gameable metric like number of cars manufactured.

It’s possible there is some underlying bias here, but there’s no reason to assume it favors one company over the other enough to offset a 300 car difference in a single month especially as the total numbers just aren’t that high.

I'm not arguing the thrust of your point here, just noting that there isn't a 1:1 correspondence in case anyone's under that particular misconception.

I agree that registrations is probably the least bad metric, and I doubt BMW is gaming it in this case.

But saying it isn't gameable is a bridge too far. Manufacturers can trivially game it, it's just marginally more expensive to do so than to shift manufacturing numbers between quarters

E.g. BMW used to operate its own car rental, they could manufacture 1k vehicles each month in Q3, then sell them all to themselves in the last month of Q3, and register them at the same time.

They can also do this with any "self dealing" or corporate registrations by simply sitting on inventory and then lowering their prices, consumers would take care of the rest.

The problem with self dealing is they need to pay taxes at registration. So while sure you can inflate the numbers it isn’t just an on paper thing this is treated as a sale and there’s real consequences of that.

As to shifting the quarter something is sold, people aren’t getting a car before registration here. Selling of excess inventory at a discount is still actual sales.

Sure, but now we've shifted from "isn’t gameable" to "gameable for a price", perhaps to inflate the stock price as you make the news for outselling Tesla or something.

And yes, you'd need to deliver the car to consumers. I'm just saying that you can time shift when that happens, if you wanted to inflate the numbers in the short term for whatever reason.

By your metric sales are also “gameable for a price” as long as you are willing to take a big enough loss essentially an unlimited number of people will buy new cars at 1,000$ a pop.

What makes metrics like manufacturing actually “gameable” is you can still sell the cars manufactured with the buzz from manufacturing more X than anyone else. Pump the numbers for 6 months and then cut production and coast until product is sold off. Your out a little time value of money but it doesn’t take much free play to cover that.

Right, the useful number instead of a comparison between cars sold to the European public and cars sold at the location as a matter of the logistical accident that it is the head quarters of the manufacturer.

Registration is useful because it cuts out cars manufactured in Germany but shipped to South Africa. We don’t want to bias these numbers based on each company’s internal logistics. That stuff is more relevant from a jobs perspective than a market perspective.

> BMW registration for the month of July 2024

Do the manufactures consider sales to dealers in their sales numbers or are they actual sales to people that then registers the car as in use?

Registration is generally done when the consumer takes possession of the vehicle.

I understand when registration is done, but that does not answer the question.

Are the numbers what the makers sell to the dealers but have yet been sold to a buyer, or are they legit sells to buyers? I can see where the makers pad their numbers by using what was sold to the dealers to have on the lots as they've technically sold the car, but it's a misleading number. Same game as storage makers using powers of 10 vs powers of 2 in their capacity claims.

the quote says “registrations” therefore it is when possession is taken

Ya, it is kind of weird because the Dealer is actually the OEM's customer. But they measure in sales to the consumer, which is a more useful number because:

1. Dealers will order certain high demand vehicles all day long when they are allowed to do so ("allocations") even though they don't have specific consumers in mind to buy them.

2. When new models become available they will also order a bunch of that new model.

3. They will order some random vehicles to look nice in their showroom or on the lot. Fun example: I recently drove by a Porsche dealership that had a bunch of Macan's lined up out front -- one in each color!

So, sales to the consumer is more useful because it's the only way to measure actual consumer demand. This is also why you sometimes see a model that hasn't been manufactured for several years show up in these new car sales reports... they were sitting on a lot unsold for that long.

You can also find numbers for how many of each model they've manufactured each month, which is useful for different reasons of course, and sometimes they sit on the manufacturer's lot for weeks or months before they're delivered to dealers. We saw this coming out of the pandemic when there weren't enough trains to transport the cars (most cars leave the plant on a train).

Reply on news.ycombinator.com