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

[betstarship]

That Isaacson book drop of his biography on Elon Musk is on September 12th...might have relevance on TSLA?

 

[Captkerosene]

OT my neighbor was razzing me about a news report that a couple of Teslas burned after being submerged in salt floodwaters around the Tampa area. Super annoying.
Please excuse the interruption, I just needed to vent.

Consider the possibility that your neighbor is just sick of you bragging about your Tesla. Just sayin'.

 

[ItsNotAboutTheMoney]

[CUE: debate over if trains or individual robocars are the best mass transit solution...]

That would of course depend on the definition of mass transit. You can't get much mass in a robocar.

 

[2daMoon]

That would of course depend on the definition of mass transit. You can't get much mass in a robocar.

ItsNotAboutThe number of people that fit inside. It is about the cost and time to move one or more people a mile (or more).

A train will have a cost per person carried and a distance carried on a train with average occupancy. That 'data' (you may be unfamiliar with this term) can be measured against the cost of a four seat robotaxi carrying *one person the same distance. (* just to make the numbers interesting)

You may not have heard of Tony Seba, nor of Ark Invest, nor have seen any of the videos and reports where the overall cost of providing this service could end up as low as $0.12/mile for an autonomous electric robotaxi. (those colored words are what we call "links" and can be clicked on to provide additional information)

Not Sure what the cost per person per mile is for commuter rail (or bus, subway, whatever). It would be interesting to see the data on that.
This report might be a good starting point.

Any such study would not further encompass how the robotaxi will take the passenger privately to the exact place they want to go, rather than to a station where they may need to employ another transportation medium in order to reach their destination. Silly as it seems, some people might actually find such convenience more practical than traditional mass transit.

This is about as Bullish as it could be

 

[Fred42]

Shipping out CT even on a Holiday.

https://twitter.com/x/status/1698747787087577158

Why would they cover the CT's in the parking lot and then ship them uncovered? They'll get dirty on the truck trailer.

 

[2daMoon]

Why would they cover the CT's in the parking lot and then ship them uncovered? They'll get dirty on the truck trailer.

They are hiding them from the drones...

 

[scaesare]

That would of course depend on the definition of mass transit. You can't get much mass in a robocar.

<trading_holiday_pedantic_mode>

mass != volume

</>

 

[scaesare]

Why would they cover the CT's in the parking lot and then ship them uncovered? They'll get dirty on the truck trailer.

Not to mention the rock chips in the paint!

 

[2daMoon]

Lucid’s Peter Rawlinson is highest paid auto CEO in 2022 with $379 million compensation

The highest-paid automotive CEO of 2022 was Lucid Motors CEO Peter Rawlinson, who received a total compensation of $379 million.

www.teslarati.com

Elon's still not taking a salary, right?

Is there some sort of inverse-squared relationship between salary and performance?

 

[hobbes]

 

[JoRoMo]

That would of course depend on the definition of mass transit. You can't get much mass in a robocar.

Think «swarm transit», perhaps.

 

[2daMoon]

Think «swarm transit», perhaps.

Or, in Oprah terms...

EVERYBODY GETS A CAR!​

 

[petit_bateau]

So in the past we have discussed the JPR007 valley of death graphic.

Spoiler: Valley of death Summer 2023 version

View attachment 970947

we've covered multiple issues with it. My first step at improving it was to fact check the historical data. I found widely varying numbers (and could find some that matched) but settled on using Cars from 2022 statistics | www.oica.net and prior years

Spoiler: historical global car production per OICA

View attachment 970949

The peak is in the same year (2017) but the scale is smaller because I excluded commercial vehicles. I'm not sure what OICA considers a commercial vehicle so that might have to be revisited.

I didn't bother fact checking the EV production numbers. I might revisit that also.

This is a first attempt at a new "valley of death" chart based on the OICA numbers. I've got what are likely very poor guesses for 2023 to 2030. I kept the 2025 as the low production year like JPR007 had it and just fudged the numbers until the curve looked interesting.

But if you would like to critique, I'll make it prettier once we settle on content.

Spoiler: very rough draft of a new valley of death chart

View attachment 970951

Interesting, nice to see others put numbers on their forecasts. Our models are broadly similar. Setting aside two issues ....

- commercial vs cars;
- future 'steady-state' demand level;

.... I think that data-this-far suggests you are slightly over-optimistic. My graph below is on the basis of (all) total vehicles per Sales Statistics | www.oica.net and I assume steady state of 94m/yr which would be 28m commercial and 66m passenger cars. The point being that in 2030 the S-curve data thus far suggests an outcome of 6m/y of ICE, i.e. 7% ICE remaining at that point vs 87m BEV, 93%. Hence my thought that you are being more optimistic than the data thus far would tend to support. You are suggesting a 100% transfer by 2030 whereas I think it will be a 93% transfer by 2030.

If there is a change in the commercial vs passenger build ratio then that might affect matters, though I don't think that likely to stray far from the 68-70% norm. Of course the US-ians with their distorting "light trucks" are the fly in the ointment here, but Tesla Cybertruck is about to show the way. That in turn suggests to me that the existing distortion will remain unchanged.

If there is a different steady-state-ish mfg volume then that could affect things. A higher volume (which is what you are suggesting as you are only counting passenger cars) would really tend to delay the switch, i.e. the reverse of the mkt share ICE/BEV that you suggest. There might be various complex dynamic issues in play at that point which on balance I think would tend to suppress demand. The 2016-2018 total peak was 95-97m/yr so that informed my view of 94m/yr given that thought. I don't have a formal validated dynamic model of this issue so I am driving on my own judgement in this respect.

I tend to agree that all PHEV, HEV, and FCEV are distracting rounding errors that will have largely washed through by then.

Year by year the data flows in and we can reflect this in our models.

 

[mongo]

Lucid’s Peter Rawlinson is highest paid auto CEO in 2022 with $379 million compensation

The highest-paid automotive CEO of 2022 was Lucid Motors CEO Peter Rawlinson, who received a total compensation of $379 million.

www.teslarati.com

Elon's still not taking a salary, right?

Is there some sort of inverse-squared relationship between salary and performance?

Elon was awarded the last 4 tranches of the 2018 CEO plan in 2022, so that article is wrong.
Around 100 million shares and $20 Billion pre tax value.
Nothing further currently in place.

 

[dhanson865]

Interesting, nice to see others put numbers on their forecasts. Our models are broadly similar. Setting aside two issues ....

- commercial vs cars;
- future 'steady-state' demand level;

.... I think that data-this-far suggests you are slightly over-optimistic. My graph below is on the basis of (all) total vehicles per Sales Statistics | www.oica.net and I assume steady state of 94m/yr which would be 28m commercial and 66m passenger cars. The point being that in 2030 the S-curve data thus far suggests an outcome of 6m/y of ICE, i.e. 7% ICE remaining at that point vs 87m BEV, 93%. Hence my thought that you are being more optimistic than the data thus far would tend to support. You are suggesting a 100% transfer by 2030 whereas I think it will be a 93% transfer by 2030.

If there is a change in the commercial vs passenger build ratio then that might affect matters, though I don't think that likely to stray far from the 68-70% norm. Of course the US-ians with their distorting "light trucks" are the fly in the ointment here, but Tesla Cybertruck is about to show the way. That in turn suggests to me that the existing distortion will remain unchanged.

If there is a different steady-state-ish mfg volume then that could affect things. A higher volume (which is what you are suggesting as you are only counting passenger cars) would really tend to delay the switch, i.e. the reverse of the mkt share ICE/BEV that you suggest. There might be various complex dynamic issues in play at that point which on balance I think would tend to suppress demand. The 2016-2018 total peak was 95-97m/yr so that informed my view of 94m/yr given that thought. I don't have a formal validated dynamic model of this issue so I am driving on my own judgement in this respect.

I tend to agree that all PHEV, HEV, and FCEV are distracting rounding errors that will have largely washed through by then.

I've been working on the formatting and slightly adjusting the curves in projections, the most recent version I had saved has 1M ICE/Hybrid in 2030 so not 100% conversion by then just 98.75% and that is with excluding commercial vehicles which might be slower to convert.

This is the slightly more pretty version I made this afternoon. I'm just not so sure about how deep the valley will be. So I lessened the dip and slowed the ICE draw down to have some straggler content in 2030.

I'm following JPR007s lead in assuming that pure EVs will increase demand past the high of 2017, but unlike him I'm not assuming growth all the way to 2040 and I'm not expecting growen above the 2017 peak to be as steep as he did. For now I'm leaving 2031-2040 out of it. I think a 6 year projection is far enough into the unknown to get the point across.

For now I've taken his EV data and inserted it as a portion of, not in addition to, any historical data I got from OCIA, if you think that is wrong please let me know.

so for example 2012 was 63.1 million in the OCIA data and I graphed it as 63.0 in one color and 0.1 in the other.

Also his 2021 number was obscured so I made a pure guess of 4.8 based on the height of the bar. Not having source data to work with.

If anyone can give me an open URL for source data for global EV sales that I can use I might switch the historical EV bars.

2012​	63.0​	0.1​
2013​	65.6​	0.1​
2014​	67.6​	0.2​
2015​	68.2​	0.3​
2016​	71.6​	0.5​
2017​	73.0​	0.5​
2018​	69.1​	1.4​
2019​	65.4​	1.7​
2020​	53.5​	2.3​
2021​	52.3​	4.8​
2022​	54.9​	6.7​

 

[Gigapress]

Inner city to city? 100% trains. I haven't needed a car in 15+ years, and wouldn't trade my bikes/trains for a car even if it had FSD v44.5.

Every time I accept a ride with friends (to make them happy), I end up losing my time in traffic. PS: I know its a privilege (I live in Paris, France) but it's a way of life too and wouldn't be that difficult to rebuild countries designed for people (instead of cars)

NB: markets are closed today right?

If Boring company achieves their goals then Loop systems will outperform urban trains on every metric of merit. It’s not even a trade-off. Loops with autonomous EVs win by a wide margin on:

• speed
• cost (both upfront and ongoing)
• safety
• waiting time for vehicle
• capacity (including flexibility too, since vehicles can be dynamically routed to wherever demand is highest at any given time)
• station density per neighborhood
• ride comfort
• reliability
• arrival time predictability
• privacy
• user simplicity (no routes or timetables to learn; just tell Loop desired destination)
• integration with existing surface road infrastructure
• construction ease and risk
• noise pollution
• energy efficiency
• resiliency against disasters, natural or manmade
• communicable disease risk

The best transportation solution is eco-friendly, grade-separated personal rapid transit, and the technology to enable PRT is finally arriving this decade. For PRT to work well, we need smartphones, mobile internet, cheap AEVs, cheap narrow tunnels, and an influential leader and city government who are crazy enough to try the idea. Almost all of these elements are now in place. The century-old cars vs. trains debate will be moot if this works. A string of robotaxis in a tunnel effectively is a subway train except the vehicles are dynamically linked with software instead of statically linked with steel, and with headways of mere seconds instead of minutes. Dynamic linkages have massive benefits.

(By the way, I like the train system in Paris and I had only a bike for transportation most of my adult life thus far.)

Here’s a good summary:

Reddit - Dive into anything

reddit.com

 

[Cosmacelf]

If Boring company achieves their goals then Loop systems will outperform urban trains on every metric of merit. It’s not even a trade-off. Loops with autonomous EVs win by a wide margin on:

• speed
• cost (both upfront and ongoing)
• safety
• waiting time for vehicle
• capacity (including flexibility too, since vehicles can be dynamically routed to wherever demand is highest at any given time)
• station density per neighborhood
• ride comfort
• reliability
• arrival time predictability
• privacy
• user simplicity (no routes or timetables to learn; just tell Loop desired destination)
• integration with existing surface road infrastructure
• construction ease and risk
• noise pollution
• energy efficiency
• resiliency against disasters, natural or manmade
• communicable disease risk

The best transportation solution is eco-friendly, grade-separated personal rapid transit, and the technology to enable PRT is finally arriving this decade. For PRT to work well, we need smartphones, mobile internet, cheap AEVs, cheap narrow tunnels, and a wealthy leader crazy enough to try the idea. Almost all of these elements are now in place. The century-old cars vs. trains debate will be moot if this works. A string of robotaxis in a tunnel effectively is a subway train except the vehicles are dynamically linked with software instead of statically linked with steel. Dynamic linkages have massive benefits.

(By the way, I like the train system in Paris and I had only a bike for transportation most of my adult life thus far.)

Here’s a good summary:

Reddit - Dive into anything

reddit.com

While I agree with you, let’s not get this thread derailed (ha!) into a train vs Loop discussion please! It is yet another third rail type of topic.

Μod: completely agree, this was thrashed to death a couple of years ago. --ggr

 

[M|S|M MYP]

Im sure this was not intentional on CNBS:

 

[wipster]

If Boring company achieves their goals then Loop systems will outperform urban trains on every metric of merit. It’s not even a trade-off. Loops with autonomous EVs win by a wide margin on:

• speed
• cost (both upfront and ongoing)
• safety
• waiting time for vehicle
• capacity (including flexibility too, since vehicles can be dynamically routed to wherever demand is highest at any given time)
• station density per neighborhood
• ride comfort
• reliability
• arrival time predictability
• privacy
• user simplicity (no routes or timetables to learn; just tell Loop desired destination)
• integration with existing surface road infrastructure
• construction ease and risk
• noise pollution
• energy efficiency
• resiliency against disasters, natural or manmade
• communicable disease risk

The best transportation solution is eco-friendly, grade-separated personal rapid transit, and the technology to enable PRT is finally arriving this decade. For PRT to work well, we need smartphones, mobile internet, cheap AEVs, cheap narrow tunnels, and a wealthy leader crazy enough to try the idea. Almost all of these elements are now in place. The century-old cars vs. trains debate will be moot if this works. A string of robotaxis in a tunnel effectively is a subway train except the vehicles are dynamically linked with software instead of statically linked with steel. Dynamic linkages have massive benefits.

(By the way, I like the train system in Paris and I had only a bike for transportation most of my adult life thus far.)

Here’s a good summary:

Reddit - Dive into anything

reddit.com

Lost Wages should be a great test for the above... not to mention great marketing for Tesla!

"You mean I could own one of these things? For how much? Damn, sounds like a deal!"

 

[MartinAustin]

Tesla in Frankfurt today (in Euros)