Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[Fact Checking]

More telling will be what Bloomberg itself does with the recent high-ball estimate of 26.9K for Model S/X production in 2019Q1. That's well outside their 'source' figure, which is actually 21K for S/X in 2019Q1.

The source corrected the Bloomberg author, Tom Randall (@tsrandall) | Twitter, via a reply to his tweet, yet Randall ignored the correction to his estimate. So he is now on his own. He owns the 26.9K S/X production estimate.

Can you please link to the 26.9k figure? I tried to find it but couldn't.

 

[lklundin]

It amounts to just about a rounding error, but the January + February delivery numbers for Denmark are:

Month Jan Feb
Model S 1 7
Model 3 2 79
Model X 2 1
All models 5 87
Total: 92

Bilimp.dk - Statistik

(Numbers regarding this month seem to be available only after it ends).

For Denmark the 2019 Q1 delivery numbers are out:

Month      Jan   Feb   March     Q1
Model S      1      7     12     20
Model 3      2     79    306    387
Model X      2      1      3      6
All Models   5     87    321    413

Bilimp.dk - Vis nyhed

PS. Trying to use CODE mark-up...

PPS. Per the above link, for March the Model 3 came in second in its class:

364 VW Passat
306 Tesla Model 3
207 Mercedes C-class
203 BMW 3
200 Volvo V60
156 Audi A4
132 Skoda Suberb
124 Kia Optima
105 Hyundai Ioniq
100 Ford Mondeo

Ouch...

 

[Toumal]

As I mentioned upthread, I expect +30% with the first generation of 'Maxcells'. This will give Tesla's flagship Model S a 435 mile (700 km) range

I think it's way to soon to predict performance or technical specs of a future battery cell that benefited from Maxwell developments. I'm convinced that this is very much a long term investment that may even turn out to not yield much in terms of results in the end.

 

[Artful Dodger]

Can you please link to the 26.9k figure? I tried to find it but couldn't.

Yeah, I think you replied already the original comment by @SteveG3

vistacruiser7 on Twitter

"@tsrandall Confusion re your Bloomberg story today. Visible Alpha (VA) has consensus S/X production at 21K for Q1. Your report cited VA as data source but said avg S/X prod est 26.9k. Apparent this is due to your leaving out most of VA's data set. A 25+% impact, wondering why?"​

Steve would have a link to the original article on Bloomberg.

Maybe its been 'disappeared' from their website?

 

[SPadival]

No need to adjust your television sets. Gali has slowed down in this podcast, which is pretty good:

 

[Fact Checking]

Yeah, I think you replied already the original comment by @SteveG3

vistacruiser7 on Twitter

"@tsrandall Confusion re your Bloomberg story today. Visible Alpha (VA) has consensus S/X production at 21K for Q1. Your report cited VA as data source but said avg S/X prod est 26.9k. Apparent this is due to your leaving out most of VA's data set. A 25+% impact, wondering why?"​

Steve would have a link to the original article on Bloomberg.

Maybe its been 'disappeared' from their website?

Yeah, I was interested in the original link/source.

I remember some tweet or blog post from Tom Randall where he took only part of the Visible Alpha S/X production estimates. He might have deleted it, realizing that it's misleading or outright wrong. It wasn't an official Bloomberg article IIRC, but I cannot find it.

His latest official Bloomberg article is this one:

"Bloomberg Tesla Model 3 Tracker Suggests a Possible Surprise"​

Very positive towards Tesla and towards Q1 deliveries, with a bit of a disclaimer:

"The international expansion might be overheating Bloomberg’s production model this quarter."​

I really think you are barking up the wrong tree here: Bloomberg is a big organization with around 20,000 employees and Tom Randall is one of the good guys, not related to the FUDsters attacking Tesla and Elon.

 

[lklundin]

Yeah, I was interested in the original link/source.

I remember some tweet or blog post from Tom Randall where he took only part of the Visible Alpha S/X production estimates. He might have deleted it, realizing that it's misleading or outright wrong. It wasn't an official Bloomberg article IIRC, but I cannot find it.

His latest official Bloomberg article is this one:

"Bloomberg Tesla Model 3 Tracker Suggests a Possible Surprise"​

Very positive towards Tesla and towards Q1 deliveries, with a bit of a disclaimer:

"The international expansion might be overheating Bloomberg’s production model this quarter."​

I really think you are barking up the wrong tree here: Bloomberg is a big organization with around 20,000 employees and Tom Randall is one of the good guys, not related to the FUDsters attacking Tesla and Elon.

I am not impressed with Tom Randall's counting skills.

At first I wrote it off as just being personally miffed that he rejected my proposal to combine the in-the-wild spotted VINs with the solution to the German Tank Problem.

But then I read the other day how he first stated that Tesla had 2900 cars in transit into Q1 and then went on to expect 10k cars in transit into Q2. Fine.

But then he takes his Q1 production estimate and to arrive at a delivery estimate he subtracts the 10k transit estimate. Which is a glaring mistake, since one has to subtract the net transit number, i.e. 10k - 2900.

A cursory search didn't bring up this mistake, so maybe he fixed it retroactively.

German tank problem - Wikipedia

 

[mrdoubleb]

Be aware that this site includes registrations of pre-owned cars too, so that may account for some - it's obvious when you look at the VINs

Yes, thanks I am aware. January was 6 higher, February 5 cars higher (on a volume of ~1k) on the site than the official stats, so this should not have a big impact though.

There used to be a time when one could fairly easily extract the data from the site for several months back, but now I can only rely on my memory. Having checked the site every few days I may have seen like 10-20% of registrations being with the 100 battery, while the overwhelming majority was 75. even for Model x which traditionally had a much higher percentage of big battery buyers.

So regardless of how many of this is new sales, it still seems like we should not expect a major profit boost from 100 kWh sales in Norway. (As we were hoping that only having bigger batteries would compensate for the lower sales volume)

 

[Fact Checking]

BTW., a small data point:

Model 3 VINs​

If you switch the VIN registrations view to "Quarterly" and "North American/International", then it becomes clear that Tesla registered more North American VINs in Q1 (77,062) than in Q4 (75,640).

But 100% of Q4 production went to North America.

So even if we ignore all the international VINs, if Tesla kept their Q4 VIN allocation patterns (which is a big if) then production to North America alone might have increased over Q4, which would be amazing if true.

Tesla might have changed their VIN allocation pattern to be a lot less dense, but the question is why would they do it for North American production?

A mystery that might be resolved in about a day.

BTW., are there any known cases of Tesla assigning a lower "2018" VIN to Q1 cars? It might all be simply the effect of losing those existing VIN ranges and allocating much bigger VIN range batches for 2019.

 

[Artful Dodger]

I think it's way to soon to predict performance or technical specs of a future battery cell that benefited from Maxwell developments. I'm convinced that this is very much a long term investment that may even turn out to not yield much in terms of results in the end.

Really? Maxwell published specs for their 1st gen dry electrode Li-on cells on Jan 18, 2019 in this PDF (pg. 10 reproduced below):

Note the energy density targets for 2nd gen cells are identified as well (500 Whr/kg), in addition to 2x extended bty life, higher production density, 10-20% cost reduction, and numerous environment benefits.

Maxwell scientists also demonstrated their dry electrode technology at a recent energy research conference, as summarize in this paper:

"Maxwell dry electrode manufacturing demonstrative benefits"

1. environmentally benign due to elimination of toxic solvent
2. lower cost by reducing capital and operating expenditures, resulting from elimination of solvent recovery and recycling system
3. improved energy and power density afforded by unique dense high loading electrode microstructures.

"We have demonstrated":

• dry robust process capability, incorporating current and advanced chemistries such as
  • graphite
  • silicon
  • metal alloy
  • nickel-rich layered transition metal oxides
• roll-to-roll dry process scalability using common and advanced battery material chemistries

Cheers!

 

[AndyH]

March numbers in Sweden was lower than I hoped. M3 was only 5th best selling car.

M3 1005
MS 72
MX 29

 

[KarenRei]

Thank you Karen. I don't want or have the time to go into details but in the 70s I did NRC proceedings involving several TVA nuclear projects. I was not allowed to discuss the safety aspects but from a cost and projected generation capacity without back end of the fuel cycle these were totally white elephants. Fortunately most were not completed because 1not needed and 2 way too costly per Mw. Amazing that it still goes on.

Indeed. If there's one thing that can't be pointed out enough, it's this: nuclear is far more popular among ideologues on K-Street than it is among bean-counters on Wall Street.

And on top of that, just imagine if nuclear power plants had to insure themselves without the catastrophic liability indemnity offered by Price-Anderson. The entire industry would die overnight.

 

[Artful Dodger]

The good news is, it's not going to get a wide viewership. It has 24,000 views on YouTube as of right now, up from 19,000 earlier today. I figure it tops out at 50,000. And I doubt most people who did view it sat through all of it.

Take some comfort in this:

Joe Rogan Experience #1169 - Elon Musk
21,177,604 views

 

[KarenRei]

March numbers in Sweden was lower than I hoped. M3 was only 5th best selling car.

M3 1005
MS 72
MX 29

It's not absolutes that matter... how does it compare to past quarters?

 

[Don TLR]

Is that a Tesla board?

Board layout and routing could be done in house easily enough with just a few skilled employees (relative to actual fabrication of anything), though they can also be contracted out just as easily...

NOT a Tesla board and yes of course routing done in house but fab and assembly is sent out..

 

[KarenRei]

green on Twitter

Hm, new voice commands landed on my car a week ago, but only saw it now. Is this new? "enable sentry mode" / "disable sentry mode" ; "keep tesla safe" ; "turn sentry on/off". anybody tried those yet?

also noticed there's now a voice command handler for wipers on/off/increase/decrease speed, but no actual code words assigned yet. Coming soon I guess, for the benefit of the model3 owners?

ok, this is a fun rabbit hole to fall through. Here's what could be voice controlled in 19.8 (but no keywords assigned yet - assigning it does not require new firmware, just new "voice commands json" download from mothership):

car.glovebox
car.mirror.set
car.steering_column
car.wipers.on/off
car.wipers.increase/decrease
car.hvac.on/warmer/colder/off
car.hvac fan.more/less/absolute
car.hvac.intake.fresh/recirc/biohazard
car seat_heater.on/off/up/down

All coming soon?

Apparently none are active yet. But probably will be soon.

That would be a pretty cool "magic trick" to adjust the steering column with a voice command

 

[aubreymcfato]

It's not absolutes that matter... how does it compare to past quarters?

Basically, like this:

 

[lklundin]

What's in Savannah, GA?

If I am not mistaken, Elon Musk's plane is there now:

https://flight-data.adsbexchange.com/map?icao=A835AF&date=2019-04-01

(Please let me know if I am mistaken).

 

[Artful Dodger]

Weekend OT:

Watched Johnny English Strikes Again on DVD last night. There's a car chase scene where Rowan Atkinson is driving an 1981 Aston Martin Vantage while chasing a BMW i3 driven by Olga Kurylenko (Ophelia Bhuletova, a Russian spy and Johnny's love interest).

Johnny says "those electric cars sound like nose hair timmers". I near coughed up a hair-ball.

Cheers!

 

[Pezpunk]

BTW., a small data point:

Model 3 VINs​

If you switch the VIN registrations view to "Quarterly" and "North American/International", then it becomes clear that Tesla registered more North American VINs in Q1 (77,062) than in Q4 (75,640).

But 100% of Q4 production went to North America.

So even if we ignore all the international VINs, if Tesla kept their Q4 VIN allocation patterns (which is a big if) then production to North America alone might have increased over Q4, which would be amazing if true.

Tesla might have changed their VIN allocation pattern to be a lot less dense, but the question is why would they do it for North American production?

A mystery that might be resolved in about a day.

BTW., are there any known cases of Tesla assigning a lower "2018" VIN to Q1 cars? It might all be simply the effect of losing those existing VIN ranges and allocating much bigger VIN range batches for 2019.

i think it's a mistake to foment hope for a blowout first quarter based on something as flimsy as VIN registrations (unless / until they match other data points). the only reason to expect VIN registrations to linearly match production numbers is prior precedent, and that is a weak justification when it matches neither guidance nor other more concrete observations. the truth is Tesla could register any number of VINs for any reason.

they could be simply starting the year by replenishing the buffer that got depleted when 2018 expired. they could be changing their methodology for some behind-the-scenes reasons. perhaps they are creating several pools of VINs internally, each with their own buffers. we really don't know.

i think it's important to remember that there's no firm reason for VIN registrations to track within 1% of actual production, other than past performance, which --- say it with me -- is not an indicator of future results.