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

[ZeApelido]

All this talk of expected price targets reminds me we should have a poll. So here it is, please vote:

When will TSLA reach $2000 per share?


You know what's funny? I wrote out the same poll title 18 months ago (pre-split). Most of you didn't really think we'd even reach $400 ($2000 pre-split) even right now.

Bears

 

[jerry33]

Does anyone understand the reason for the odd data entry for # of shares owned? It only allows you to set your share number in strange increments. Why not just let it be an integer? And what is the purpose of having a slider to adjust the number? It bothers me when I don't understand how something odd came to be.

Feeping creatureism.

 

[MP3Mike]

Thank you at to all for the support, I'm not sure if everyone realizes how significant this is and how detrimental. Smashing the largest solar market in the US will impact all US solar markets and really impact Tesla negatively. PGE is a criminal enterprise plain and simple and it needs to stop.

This is sort of what Jack predicted and why he was pushing the "selfish solar" idea. It may push people to take their on-grid systems off-grid by buying a battery system with the extra money NEM3.0 would cost them. Sure it takes them mostly, or completely, off the grid but it also removes any excess power they produce from the grid, making the grid worse off. Personally I would love to cover my roof with solar panels and have a Tesla Powerpack installed, but it just doesn't make financial sense at this time. (I, also, think that Tesla doesn't like to, or plain won't, put a Powerpack in a residential install.)

 

[heltok]

As this is the Investor Discussions thread of TMC, I'll add my single post as to why I feel FSD is an important feature for _my_ owning a Tesla, and part of my investment thesis, and purchased FSD 5 years ago, even though my model S still does not have the cameras (yet) to even support the beta version.

Imagine having a medical emergency and having a FSD enabled Tesla drive you to the closest hospital and announce over the speaker (or from all automatically opened windows and the sound system for older cars without an outside speaker) that the driver is in need of medical attention. What's that worth? Priceless, to me.

While I don't know if Tesla envisions adding such a feature to FSD, to me, it seems a simple addition to the appeal of FSD.

Here's a quick Google search that shows the number of crashes, and lives, that could have been potentially saved given technology that could have made a difference.

I tore my rotator cuff a year ago. Was about 1h drive to the hospital, autopilot really helped me there. Imo this is something very underappreciated. I remember reading this story:

Tesla’s autopilot helps get man to the hospital during medical emergency

Joshua Neally’s Tesla Model X didn’t exactly save his life when he started having severe chest pains, but it helped him get most of the way to a hospital

www.theguardian.com

The 37-year-old was driving in his electric car from his law office in Springfield, Missouri, when the air was sucked from his lungs and he felt a sudden biting pain in his chest – a blocked artery in his lungs. Distracted by the pain and still in traffic, he let the car’s controversial autopilot carry him down the road toward a hospital.

“It was the most excruciating pain I’ve ever had,” he later told local KY3 news. “It was kinda getting scary. I called my wife and just said, ‘Something’s wrong.’”

“I just knew I had to get there, to the ER,” he said.

Tesla’s autopilot function requires a driver to touch the wheel every few minutes, and Neally was able to keep it active for 20 miles down the highway. He said that in the moment of crisis, he considered stopping for an ambulance but decided to trust the software, for fear that he would crash if the car were entirely in his control. Tesla’s software has been designed to guide cars to a stop on the side of the road if a driver stops responding.

“If something like that happens where I become unconscious or incapacitated while I’m driving,” he said, “I’m not going to cross over the interstate and slam into somebody or slam into one of the big rock walls.”

Eventually carried near a hospital, Neally gathered his strength and took over, driving off the highway exit ramp and the rest of the way to the hospital. He has since recovered and continues to receive treatment.

 

[Tyler34]

I tore my rotator cuff a year ago. Was about 1h drive to the hospital, autopilot really helped me there. Imo this is something very underappreciated. I remember reading this story:

Tesla’s autopilot helps get man to the hospital during medical emergency

Joshua Neally’s Tesla Model X didn’t exactly save his life when he started having severe chest pains, but it helped him get most of the way to a hospital

www.theguardian.com

The 37-year-old was driving in his electric car from his law office in Springfield, Missouri, when the air was sucked from his lungs and he felt a sudden biting pain in his chest – a blocked artery in his lungs. Distracted by the pain and still in traffic, he let the car’s controversial autopilot carry him down the road toward a hospital.

“It was the most excruciating pain I’ve ever had,” he later told local KY3 news. “It was kinda getting scary. I called my wife and just said, ‘Something’s wrong.’”

“I just knew I had to get there, to the ER,” he said.

Tesla’s autopilot function requires a driver to touch the wheel every few minutes, and Neally was able to keep it active for 20 miles down the highway. He said that in the moment of crisis, he considered stopping for an ambulance but decided to trust the software, for fear that he would crash if the car were entirely in his control. Tesla’s software has been designed to guide cars to a stop on the side of the road if a driver stops responding.

“If something like that happens where I become unconscious or incapacitated while I’m driving,” he said, “I’m not going to cross over the interstate and slam into somebody or slam into one of the big rock walls.”

Eventually carried near a hospital, Neally gathered his strength and took over, driving off the highway exit ramp and the rest of the way to the hospital. He has since recovered and continues to receive treatment.

Good thing they included examples of autopilot killing people in the article as well...Wouldnt want readers to get the wrong idea.

 

[Krugerrand]

All this talk of expected price targets reminds me we should have a poll. So here it is, please vote:

When will TSLA reach $2000 per share?


You know what's funny? I wrote out the same poll title 18 months ago (pre-split). Most of you didn't really think we'd even reach $400 ($2000 pre-split) even right now.

View attachment 754744

Bears

No VPN user voting allowed. Cat not stupid.

 

[juanmedina]

So what is Russ going to print now:

https://twitter.com/x/status/1481025114292441089

I just got blocked lol

 

[nativewolf]

Apologize that I'm not often around this Forum a place that was an important kick-start of my Tesla journey. It's been a time challenge sometimes but feel free to approach me at Twitter if urgent questions come up where you believe my support is needed.

Since you are here today...what do think about Berlin. Much to do about nothing ? Is Tesla actually ready to start producing serious volume? Any more news on the various battery related facilities in the EU?

 

[Singuy]

So what is Russ going to print now:

https://twitter.com/x/status/1481025114292441089

I just got blocked lol

Something about DMV shifting requirements for autonomous testing vehicles to report safety data in Ca.

 

[EVDRVN]

This is sort of what Jack predicted and why he was pushing the "selfish solar" idea. It may push people to take their on-grid systems off-grid by buying a battery system with the extra money NEM3.0 would cost them. Sure it takes them mostly, or completely, off the grid but it also removes any excess power they produce from the grid, making the grid worse off. Personally I would love to cover my roof with solar panels and have a Tesla Powerpack installed, but it just doesn't make financial sense at this time. (I, also, think that Tesla doesn't like to, or plain won't, put a Powerpack in a residential install.)

I agree, does not make sense for me either but I want to do it at a cost to stick it to PGE! I have friends that are planning to pull the plug. Would be great if Tesla enabled V2G on cars in CA to stick it to the utilities LOL

 

[JRP3]

Great way to find the clown show

View attachment 754738

You could have drawn a larger oval.

 

[WattsHappening]

All this talk of expected price targets reminds me we should have a poll. So here it is, please vote:

When will TSLA reach $2000 per share?


You know what's funny? I wrote out the same poll title 18 months ago (pre-split). Most of you didn't really think we'd even reach $400 ($2000 pre-split) even right now.

View attachment 754744

Bears

Man, I hope we are just as wrong this time also!

 

[BitJam]

How would I navigate it as a human the first time I do it? I would hopefully not need a map. I would drive it safely based on my view of the road and its conditions. In fact incorrect speed limit data from maps is a great example of the sort of problems maps offer currently.

I would also err on the side of caution when I don't know what's ahead. I agree with you there is a fusion problem using maps but there are some places where maps and/or memory seem to be needed. Dr. Know-it-all on YT is very pro-Tesla but he says he doesn't normally use FSD beta because of this problem with the lack of memory/user-input. In his case the road he takes into town is considered to be a parking lot by the car. How can you train FSD to work in places where people know they can't use it?

ISTM there are 3 options:

1) Drive unsafely in some situations
2) Always err on the side of caution (which gets very annoying after a while)
3) Have some memory so you only err on the side of caution the first time you are on a new road.

But as you point out, adding memory/maps opens up a can of worms. The mantra is that if humans can do it then Tesla can do it with neural nets. But that only works if the car has the same faculties as humans such as memory. This is why I think solving the last 1% of FSD is going to be really hard.

 

[2daMoon]

You know what's funny? I wrote out the same poll title 18 months ago (pre-split). Most of you didn't really think we'd even reach $400 ($2000 pre-split) even right now.

View attachment 754744

Bears

I guess you are doing this one pre-split too?

 

[Bet TSLA]

I would also err on the side of caution when I don't know what's ahead. I agree with you there is a fusion problem using maps but there are some places where maps and/or memory seem to be needed. Dr. Know-it-all on YT is very pro-Tesla but he says he doesn't normally use FSD beta because of this problem with the lack of memory/user-input. In his case the road he takes into town is considered to be a parking lot by the car. How can you train FSD to work in places where people know they can't use it?

ISTM there are 3 options:

1) Drive unsafely in some situations
2) Always err on the side of caution (which gets very annoying after a while)
3) Have some memory so you only err on the side of caution the first time you are on a new road.

But as you point out, adding memory/maps opens up a can of worms. The mantra is that if humans can do it then Tesla can do it with neural nets. But that only works if the car has the same faculties as humans such as memory. This is why I think solving the last 1% of FSD is going to be really hard.

Tesla has talked before about handling potholes. That requires some sort of collective memory. It's not hard to imagine that dynamically built maps are enhanced by location markers indicating at least: suspected pothole, hard to detect something (sign, dip, intersection, driveway, black ice, deep puddle, ...), unspecified danger. For lack of any other response, FSD could just go slow and be extra cautious in such places. After all, humans have memory and use it in addition to vision to improve safety when driving.

Admittedly, this is after hours, but surely this is a discussion for some other thread.

 

[GOVA]

Does anyone understand the reason for the odd data entry for # of shares owned? It only allows you to set your share number in strange increments. Why not just let it be an integer? And what is the purpose of having a slider to adjust the number? It bothers me when I don't understand how something odd came to be.

This report was done in MSFT Power BI. There is no free text input there, so what you expect needs to be done in different way. My guess is that this was quick 'table' of predefined values, used as slider which multiplies per shares.

Frankly, I like it. It is nothing special, but works fine for this.

 

[MC3OZ]

But that only works if the car has the same faculties as humans such as memory.

When a human remembers something, then sees something different, the human typically updates their memory.

When the human is operating in a situation with limited vision, and what they can see, matches their memory, the human has a lot more faith in what they can see. When the human is operating with limited vision, and what they can see doesn't match their memory, the human applies additional caution.

So I think maps might be useful sometimes, but only if they are automatically generated and updated. And in all cases what the car can see is the higher priority. Maps are more useful for predicting what the car currently can't see.

If they are using maps, these are basic vector maps, overlayed with speed limits, stop signs, traffic lights, school zones etc. Sometimes knowing in advance you might be approaching a school zone is useful. Or if you see one not on the map, it can be added.

Assuming I'm right about maps, then as cars drive around they are updating maps, somehow map updates would need to be captured and pushed to the fleet, But only the map changes, it is a reasonable assumption that most of the map rarely changes.

While I would like to see maps used in this way, I don't know if Tesla current does it, IMO it is 50/50

 

[The Accountant]

First, thanks for the detailed numbers and explanations you are providing. When I read your posts I feel smarter for having done so.

In July 2021 you predicted 2022 at $1,700 and additional prices out to 2026. Do you anticipate bumping the post-2022 numbers at some point?

Yes - I should have that 5 year forecast updated in a week or so.

 

[Knightshade]

While I would like to see maps used in this way, I don't know if Tesla current does it, IMO it is 50/50

They don't. Maps are currently from a 3rd party (see the myriad threads of folks frustrated by all the map errors that seemingly never get fixed)- and who the 3rd party is has changed at least 2 or 3 times over the years.

Like the earlier underwriting their own insurance thing it's something Tesla has talked about doing in the future though. I think they DO use fleet data to crowdsource traffic at least (when they also used to be only 3rd party data)- IIRC Green reported that change mid-2020.

 

[Thekiwi]

Tesla at present sells in the four largest global auto markets and has factories open(ing) in three of them. They sell in Japan (third) but probably will not build there any time soon.

India (5th) has been discussed here and is near to selling and probably some form of factory (most likely a CKD of the design coming from China). After all the biggest seller in India is Maruti Suzuki, that is Suzuki design and partly supplied from outside India.

Brazil (6th) was planned in 2015 and Model 3 reservations were taken at launch. I still have one of those.
Were the actual market be considered it would be Mercosur, since most OEMs have factories in Brazil, some have in Argentina, but all serve the whole fo Mercosur from one of those. They are a single automotive market. The Brazil, Argentina, Chile and Uruguay car sales in 2020 were 3,018, 342, which makes the market slightly larger than Germany, hence the 4th largest automotive market.

Lest we ignore the actual market, in 2020 three Mercosur nations of Brazil, Argentina and Chile sold 430,626 vehicles of typical ASP of US$80,000 or above, although the typical US ASP for the same vehicles is around US$50,000.

When we're thinking of next steps for Tesla we should realize that every existing country is far from saturated, even California. Then the markets globally are all barely covered. Even Norway has a huge installed base of ICE that need to be replaced.

For cars and for TE the prospects are amazing. It would be well for us to realize that the faster other BEVs arrive, the more the market will grow. In short we should hope the Tesla market share declines quickly, for that means the addressable market is increasing.

Thus, in Brazil right now there are BEV's from Audi, BMW, BYD, Citroën, Chery, Fiat,JAC,Jaguar, Mercedes, Mini, Peugeot, Porsche, Renault and Volvo. I've probably missed some. Sales are small but BEV sales in 2021 rose 77%. 2022 will grow much more, because these are beginning to move perceptions.
All we need is Tesla to build Mercosur wide Superchargers to drive adoption far more quickly. FWIW, I ddi not list the array of commercial vehicles that is rapidly entering the market. Nor did I mention anything about the rapid growth of utility level wind,solar and battery storage. Nor the phenomenal growth of residential photovoltaics.

This is not specific market advocacy. It is a reminder that we are always understating the prospects of growth. How bullish can one be? Never quite enough!

Recently there were a few posts talking about Tesla‘s new battery as being one of the biggest changes in the auto industry ever. I actually think the gigacasting is a much more fundamental change, and the new battery format/tech just adds to it.

At first Elon thought massively increasing the jobs done by robotics in body construction was the answer to the question of “how do we build a car body” - and that didn’t quite workout, and they ended up needed more humans on the line. But then coming up with giga-casting was a real stroke of genius and the clear “first principles” answer, and that reduction in needed factory space - along with the vastly reduced footprint required for 4680 cell production - I think may set the scene for some smaller integrated tesla auto/cell factories for places like Brazil, Japan and maybe the UK, where something like a ~250k annual unit volume may be more appropriate, at least initially. If they can figure out a way to make a line that can swap between 3, Y, the smaller hatch and/or van with minimal downtime then that would work great in these smaller volume markets.