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

[AcesDealt]

They didn't manufacture enough cars.


They didn't manufacture enough cars.

This isn't complicated.

Had they sold a lot more cars there wouldn't have been a problem.

However the problem is not that they failed to produce enough cars. I agree it is not complicated. You are a disappointment neroden.

 

[Tslynk67]

You are correct. For some reason, I thought I remembered it as 7k total 3's. That makes his estimate more unlikely as you say.

But, nothing stops Elon from stating those wild projections. It's just who he is, and that's why the SP is now usually numb to any estimates he spits out on the Twittersphere.

I reluctantly have to agree. What Elon has achieved is amazing, but he's cried wolf too many times. The only way forward from here is actual results, as someone stated a few days "show me don't tell me" or even better, "Show Don't Tell", which happens to be a Rush song

 

[NicoV]

Yeah, I know about the M3 derived motor. But Tesla stated the S/X sales drop off was due to the line shutdown of the line for the changes. Seeing the changes announced it, I still do not see anything that would cause a long time shutdown of the line, especially a product that has high profit margins.

More automation. Tesla specifically said that they fired people working on the S/X lines. To bring the production to the level of Q4, they need more automation if they don’t hire again.

 

[Artful Dodger]

The video still plays if you click the vimeo link through Elon's twitter.

Not seeing Elon's tweet either. Could someone kindly post a link to the vimeo vid?

TIA.

EDIT:

"Chamath Palihapitiya On $TSLA" - YouTube (8:11)

 

[MikeC]

FMHR-chamath palihapitiya - Streamable

 

[Donatas]

Not seeing Elon's tweet either. Could someone kindly post a link to the vimeo vid?

TIA.

Vimeo video was removed. Here it is recorded:
Gofile - File sharing platform, anonymous and free
Lots of other links with different versions up here too

 

[FS_FRA]

I’m more disappointed with the Saudis state-backed killing of a journalist.

... not to mention the 36 recent beheadings

#middleages

 

[Fact Checking]

Probably you're right but doesn't give you the warm and fuzzies about the usefulness of Tesla's guidance does it.

Well, initial, broad 2019 guidance in the conference call(s) was "350,000 to 500,000 Model 3s", which shows a 30%-40% uncertainty with a 425k mid-point. Several major factors, such as Panasonic's cell output are outside of Tesla's control.

In the long run the exact trajectory won't matter - in the short run it's going to cause months of anxiety over supply and demand worries. Tesla is apparently the only company on the planet that is both supply and demand limited at the same time, depending on whom you listen to.

 

[uselesslogin]

UK orders are live with expected delivery in June.

 

[Matias]

Unfortunately I dont because since I'm not in Canada I always get redirected. Here's a screenshot someone posted on reddit: https://i.imgur.com/ddt9uxk.jpg

Tesla should also offer one in Finland, because the current cheapest Model 3 available in EU is slightly over finnish incentive price.

Actually selling SR in EU would be enough.

 

[carterm2]

This copy has good audio quality (Palihapitiya's speech on CNBC)
Gofile - File sharing platform, anonymous and free

Thanks for this... very helpful. Palihapitiya's thinking is clear and reasonable and I understand better what the problem has been with my expectations for the stock.... I have been rushing it and I did not realize who is in control of the narrative for the stock and why. His responses to the idiot interviewing him were really important to my view of the situation. Style vs substance.

I noticed a significant improvement in the Navigate on Autopilot today, (2019.12.114b1dd29) smoother, more aggressive, self lane change. It is nice to have a way to see the progress. I will be driving NOA everyday now.

Living in the Bay Area gives you the impression that Tesla has already won... there are multiple Teslas on the road with me, it is clearly a favorite car around here.

 

[lklundin]

I'm not sure how many times that Elon has to point out that mass production is far harder than most people give it credit for :Þ You don't just "change a motor" in a mass production line, as if it's some sort of weekend task.

On this topic, it is known whether the new Model S/X motor is interchangeable with one in the Model 3 - or is it a new variant of the Model 3 design?

 

[lklundin]

I loved what he said.

However, when asked about the Audi eTron, I do wish in addition to his great comment about Tesla already owning EV mindshare, he’d asked the interviewer:

“The Bolt didn’t win over the Model 3 just because it was first to market, so how will the $25k more expensive yet, slower and smaller eTron beat the Model Y?”

A small recap of who was interviewed and what was said would be greatly appreciated.

Thanks.

 

[Fact Checking]

7. Tesla tells a % of the fleet to collect detailed NN information when they encounter the situation at the top of the priority list. For example bikes on cars. They train a basic NN to recognise these pictures as a trigger to start collecting data.
8. Tesla start getting data back, they annotate this and feed it back to the recognition neural net. This makes the cars better at detecting the situation they are targeting, so the quality of data sent back by the fleet improves.

Note that the 'campaign' approach that @verygreen reported about a couple of months ago and which @KarenRei mentioned too, to automatically collect imagery from certain types of filtered-for circumstances, has the big advantage that it will collect a statistically valid sample of both negatives and positives - which data is going to train the networks that have both a low false positives and a low false negatives detection rate.

I.e. the end effect is the same: Tesla uses their fleet of hundreds of thousands of vehicles to train their neural networks, it's just that the 'campaign' method makes it smarter than just blindly collecting everything, or only collecting disengagement events.

For example the 'tunnel' problem the auto-wipers had is I believe a classic example of where the 'campaign' method helped: the auto-wiper NN attribute used to like tunnels and interpret them as "it's raining" for some reason (I speculate it's because of the "overcast sky" and the tunnel lighting also illuminates dirt on the windscreen that might be interpreted as raindrops) - so they probably sent out several rounds of "campaigns" to refine the NN's understanding that "entering tunnels does not mean it's raining".

I.e. Tesla's relationship to their NN is not a 100% data mining feedback mechanism like Google where all user data is recorded and sucked up, it's more like a teacher-pupil interaction, where the teacher is actively adapting to misunderstandings and is shaping the next lesson to help with those topics. The "lessons" are the @verygreen "campaigns".

 

[lklundin]

That makes sense too. Building the Y at Freemont where its close cousin the 3 is made and its supply chain already setup may make more financial sense then spending big money to physically expand GF1.

When Tesla started the Model S production they had a _lot_ of space at the Fremont site.

So I wouldn't be surprised if Tesla set up the motor production there without much concern for how much space it would take up.

So maybe the space freed up in Fremont by no longer making S/X motors there can make a significant contribution to the space needed for the Model Y production.

 

[AcesDealt]

Stepping back and looking at the big picture: Tesla lost 700M$. What is Tesla's fundamental problem in getting to profitability?

They had +215M$ in regulatory credits (pure margin) this last quarter. They have 3 extremely well selling vehicles with essentially no apples to apples competitor worth talking about. They have extremely high brand enthusiasm. They have a vertical sales structure where they don't share profits with dealers and also do most service work. They get revenue from the refueling component. They don't spend on advertising. They get thousands of revenue per unit from high-margin SOFTWARE. Most buyers have tax incentives that are worth 4k+$.

They've got so many huge advantages yet 700M$ loss. Tesla has screwed something up along the way here.

This post has 10 disagrees and 3 ironic funnies. The arrogance of this forum is astounding. You are losing against a 5 year market index. I didn't make the mistake neroden did thinking SolarCity was a good idea or what FC did thinking a hardware chip is an important advantage or what almost everybody did not doing the simple math about the implications of a model 3 that cost 38k to build after the dreadnought idea was proven bullshit.

After 15 years of investing in the market the truth remains that all online message boards have negative value.

 

[lklundin]

So how did YOU figure out that the road wasn’t actually closed?

With enough data of examples of this, labellers can tag clues that indicate the road isn’t actually closed. AI can be trained to recognize Road Closed signs that are not applicable.

Again, nobody else has the capability to gather enough data to do this except Tesla.

This is a bit of a corner case, and it will take a little while to work all of these corner cases out, but I believe Tesla is well-positioned to be able to handle this.

Tesla can collect plenty of this kind of training data from their customers in Italy...

 

[uselesslogin]

After 15 years of investing in the market the truth remains that all online message boards have negative value.

I hope you find it entertaining like I do because you seem to spend a lot of time here getting negative value.

Edit: Also my 19 years investing in the market trumps your 15.. I guess. Or maybe I'm just older than you.

 

[Tslynk67]

The video still plays if you click the vimeo link through Elon's twitter.

Gone from Vimeo too - obviously way too positive about Tesla for CNBC's liking...

 

[uselesslogin]

Gone from Vimeo too - obviously way too positive about Tesla for CNBC's liking...

I think it was more the pointing out that Chanos only makes money once a decade.