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

[Artful Dodger]

"The Coming Dawn of Self-Driven Cars with Oliver Cameron" | ARK Invest (55 min)

Cheers!

 

[Bet TSLA]

Dilemma, dilemma. What to do? Getting a MCU2 for my Model X or two TSLA shares which is about the same price (tax included). And no, can’t do both

My approach was to write a few puts (about 5% above the current price) during the mid-morning dip and buy them back later in the day. Made around $5500, which after taxes (income and sales) covers the upgrade. Should work nicely on a Monday, although of course failure is an option.

 

[M|S|M MYP]

I wonder what will happen once Q3 is finished and Tesla posts their first profit ex-credits, because for the first time ever they will have had 2 factories both operating at full capacity. And then in less than 2 years, they will have doubled the number of factories to 4.

But I guess this is falling into the trap of "Tesla is a CAR COMPANY and should be VALUED LIKE ONE!" though, so that's probably the next goalpost movement that's coming up.

I wonder what my TSLA will be worth once Tesla is getting revenues from 4 factories. Hopefully a little more, my portfolio is not quite at the stage where I can think about retiring yet so maybe a little more share price appreciation would help me out.

The share appreciation will definitely come...how soon? Given the economic environment...probably longer than it should. I do plan on holding $TSLA for a very long time, at least 10 more years until i retire...unless of course it gets to $5k/share in 3 years and then i will consider selling maybe 2 or 3 to buy a nice bottle of whisky

 

[M|S|M MYP]

Could we discuss likely scenarios for S&P inclusion?

We can expect the various players to look after their own interests. The behavior of one party may have an effect on another party.

Once the announcement is there, sellers instantly want more for their chairs. Preferably the most. A squeeze is what they will like best. they will want to sell at the peak.

MMs will prefer to have multiple peaks. They know when they manipulate so they know that the SP going down is their own doing. They will scare sellers who think it is the peak to jump ship.

I would expect index funds to do dollar cost averaging. They could take advantage of an induced price drop, but if they are too eager there isn’t much of a price drop. So, I expect them to stick to their strategy.

this allows the MMs to allow the price to go up again. Now, will they let the peak price go higher than the previous peak? I’d think they would make the second peak lower and the third peak higher than the first one. The second peak is to shake out those that gambled that the first peak would be higher and Now have an opportunity to still make a nice profit. The third one to shake out the stubborn sellers that can pride themselves in getting a bit more out of it.

But what do I know.

What we do know is that they have a lot of $$$ and with the anemic volume being traded lately...they have free reign to do what they want.

 

[Knightshade]

Add they will be able to add 1M Robo-taxi capable cars to the fleet each year.

This one confuses me a bit too.

There's only a few hundred thousand taxi and limo drivers in the entire US right now.

Nobody needs, or wants, "millions" of robotaxis here.


Maybe, worldwide, there's like 4 millionish taxis.... but like 2/3rds are in India or other less developed countries where there's no superchargers and no Tesla presence at all... plus China- where both the fleet and the supercharger network will need a LOT more buildup to replace taxis.


Here in the US there's already more Teslas on the road than demand for taxis- let alone future sales between now and when robotaxis are actually ready.

I've actually seen a number of folks on twitter (including at least one exceedingly prominent Friend of Elon) telling everyone they can have free cybertrucks thanks to autonomy and robotaxis... as if all the million+ teslas each year will all have infinite demand for renting themselves out- and it's just not so.


Doubtless there'll be some folks whose circumstances make it possible for them to cease owning their own car and rely on robotaxis, but lots, lots, more, will continue to prefer (or require) their own vehicle.....

After all if all everyone cared about was going A to B safely in a car they'd just sell 80 million corollas a year instead of 80 million vastly different, often far more expensive, types of vehicles from vastly different companies, right?



Don't get me wrong- Robotaxis would be a large cash cow for Tesla- and some percentage of owners who are someplace with enough demand relative to the # of Teslas in the area... Likewise it'd make all the Teslas NOT used as taxis hugely more appealing to folks who want their own cars to take them places.

But probably not quite the "everyone can earn huge money letting their free-to-them car drive around while they sleep and traffic will be solved!" panacea some folks imagine.

For Tesla there is one simple question:-
1) What are you doing about cleaning?

Charging's a question for Tesla as well- with no driver in the car they'll either need to retrofit every supercharger with an automated system that doesn't appear to currently exist- or they'll have to skim some of that sweet FSD money to pay for a supercharger attendant at every location to recharge the robotaxis as they come in and out.

The other issue is the current SC network is largely built around enabling long-distance travel via major highways and interstates.

So they're exactly the opposite locations from where a lot of taxis would be operating- local driving in largely urban non-highway areas.

I suppose it's not a huge deal if the car drives itself to have to travel a bit further from downtown-wherever to do it...but the roundtrip, plus charge time, will cut into the $/hr the robotaxi earns unless they're planning to alter future supercharger rollout to make more sense for this application.

(and that's not even covering the fact folks letting their cars robotaxi 12+ hours a day means they're gonna be hitting SCs multiple times a day- putting a LOT more demand on the network than what it was built for)

None of that's insolvable at all- but it will take money to do it.

I feel like you don't have a good understanding of this space. I'd suggest reading the Autonomy section of this blog I wrote.

I'm curious on what basis you feel you have a good enough understanding given when someone pointed out a real-life shortcoming of Teslas current system you've mentioned-

I don't own a Tesla, and I don't even have a driver's license, so maybe I don't fully understand the problem you're describing

And went on to admit you were unaware the car often dropped out of NoA (and sometimes AP entirely) in bad weather.


I don't disagree their "solve vision" approach is the ideal approach- and they've certainly got the largest opportunity for data collection- but your blog makes some thus-far "facts not in evidence" claims on the hardware side... (which should probably go to that new thread the mods created).

 

[Steve m]

This one confuses me a bit too.

There's only a few hundred thousand taxi and limo drivers in the entire US right now.

Nobody needs, or wants, "millions" of robotaxis here.

Uber stated that there are over 750,000 Uber drivers in the US in 2017.

Uber has more work to do winning over drivers

 

[FrankSG]

I'm curious on what basis you feel you have a good enough understanding given when someone pointed out a real-life shortcoming of Teslas current system you've mentioned-



And went on to admit you were unaware the car often dropped out of NoA (and sometimes AP entirely) in bad weather.


I don't disagree their "solve vision" approach is the ideal approach- and they've certainly got the largest opportunity for data collection- but your blog makes some thus-far "facts not in evidence" claims on the hardware side... (which should probably go to that new thread the mods created).

I definitely don't have a deep understanding of the current state of Tesla's autopilot, although I do somewhat keep up-to-date through things like the Dirty Tesla YouTube channel.

There are also people far more knowledgeable than me when it comes to the tech and how to solve autonomy, although I believe I have a pretty decent big picture understanding through my research, and a little bit of a background in Machine Learning, Neural Networks, and Computer Vision.

However, what I think I have quite a good view of is the competitive landscape of this industry, because that's what is the most important to understand for me as a Tesla investor, so that's where I focused my research. At the same time, I do acknowledge I could be overestimating my understanding. I'm always open to being proven wrong given solid evidence.

Regarding the "facts not in evidence" claims on the hardware side, what I quoted was an excerpt. There is more explanation on why Tesla is at least 1-2 years ahead in terms of Hardware chip. ARK usually says more years than that, but based on the research I did, it seemed more like 1-2 years to me.

 

[dc_h]

I'm in the mood to speculate (this overview based on a video frame grabbed from Jason Yang Aug 6, 2020)

View attachment 573815

Clockwise from top-left:

1. Motor Workshop - shared output for Models 3+Y
2. Battery Workshop - shared output for Models 3+Y
3. Gigapress / Stamping - (near end of Phase 1) possible location for Gigapress*
4. Gigapress / AL Casting - (Phase 2 standalone bldg) 2nd possible location for Gigapress*
5. Body Shop - Model Y (note smaller bldg size made possible due to AL casting of underbody)
6. Additional Stamping - Model Y (could be anything, but have to make a guess here!)
   
7. Paint Shop - Model Y (larger shop for 2x volume vs Model 3 / Phase 1 paintshop)
8. Air Handling / Paint shop (27 large roof-mounted air handling units for large throughput)
9. Offices / Engineering (no windows in this building)

Note that there is one more building under construction, which is obscured in this photo, on the far side of the building I've labeled "Gigapress / AL Casting". It is a smallish, 1-story building I'm tentatively calling a Logistics Office, for managing shipping in/out of Phase 2.

I also would not be surprised to see that eventually there will be at least one more factory building added to the Phase 2 complex, where the blue-tarped yard is now between the Motor Workshop and the Electrical substation and Utilities building. I'd expect Tesla will want to manufacture more of their own components onsite, and I haven't yet identified a location for a Seat Workshop for instance. At any rate, items intended for use in General Assembly would have a prime location for production in this space.

All of the above is subject to revision as more info becomes available. Not advice for building your Gigafactory.

Cheers!

Great post. I was thinking the additional stamping site would be a local seat factory. I think the gigacast would only be needed next to the original stamping building at the end of the Model 3 building. They have a an overhead connection between the new gigacast building and the Y assembly building. Cool if they are already installing a new gigacast for the front of the car, but it seems too soon. They should build for the model 3 before expanding for Model Y.

 

[Bullish_Bear]

I think you can read it here: Tesla's Profits From Emission Credit Sales Will Disappear - ValueWalk

Thanks! It's like dear Mark would hide his face in the pillow sometimes...there is no fund performance in this article

 

[MC3OZ]

There's only a few hundred thousand taxi and limo drivers in the entire US right now.

Nobody needs, or wants, "millions" of robotaxis here.

I think the true answer lies somewhere between the most bullish predictions and your answer..

Robo-taxis will displace some public transport use, and some private car use.

In theory each Robo-taxi takes up to 10 private cars off the road...but the number of trips take in Robo-taxis will be much higher.
A large fleet size is needed when we consider customer's likely expectations:-

1. A Robo-taxi arrives with in 5-10 minutes of being ordered
2. If ordered 15 minutes in advance a Robo-taxi always arrives on time...
3. The is close to zero chance 1. & 2. are not met.. perhaps a late Robo-taxi means a free trip.

What matters most of all in terms of meeting the public service expectation, is growing the fleet rapidly, until it provides the ideal service level.

or they'll have to skim some of that sweet FSD money to pay for a supercharger attendant at every location to recharge the robotaxis as they come in and out.

I think this will be the initial solution. bit large fleet size is helping here, most cars will have 300 miles of range, and need Supercharging after say 250-270 miles. if a busy city location Tesla can have a purpose built recharging station and an attendant will simply walk around a rectangular block, plugging or unplugging as needed.

Keeping in mind cars with lower battery levels can be scheduled for shorter trips only. So there is a way of spreading trips over the total battery capacity of the fleet.

In more remote locations they will simply use the regular Supercharger station, but Tesla needs to arrange for it to be plugged in.

Long distance trips can be a different hire, and the passenger can plug in the car as needed, if required many customers will not find plugging in a big hassle. If they don't plug in, they need to move their luggage to a new car,,

The big advantage of the Supercharger network is Tesla knows where they can plug in and what they will pay... and some of what they will need is already built.

In future fully automated charging is possible, it is just isn't a requirement for Day 1 of operation..

 

[Curt Renz]

Tesla Daily - half hour ago:

 

[FrankSG]

This one confuses me a bit too.

There's only a few hundred thousand taxi and limo drivers in the entire US right now.

Nobody needs, or wants, "millions" of robotaxis here.


Maybe, worldwide, there's like 4 millionish taxis.... but like 2/3rds are in India or other less developed countries where there's no superchargers and no Tesla presence at all... plus China- where both the fleet and the supercharger network will need a LOT more buildup to replace taxis.


Here in the US there's already more Teslas on the road than demand for taxis- let alone future sales between now and when robotaxis are actually ready.

I've actually seen a number of folks on twitter (including at least one exceedingly prominent Friend of Elon) telling everyone they can have free cybertrucks thanks to autonomy and robotaxis... as if all the million+ teslas each year will all have infinite demand for renting themselves out- and it's just not so.

I don't think these numbers are accurate. Pretty sure the number is much much higher than 4 million. A quick Google search turns up an 18M number, but I'm not sure how accurate that is either.

Either way, taxis are quite expensive as of right now, and robotaxis will be much much cheaper. Comparing the current market size for taxis to the future market size of robotaxis, is like comparing the market size of a Model S to the market size of a Toyota Camry.

Doubtless there'll be some folks whose circumstances make it possible for them to cease owning their own car and rely on robotaxis, but lots, lots, more, will continue to prefer (or require) their own vehicle.....

I don't want to incorrectly assume your background, and I know very little about North Carolina, but is there a possibility that your opinion is influenced by how you've always experienced transportation over roads?

I know that in urban areas transportation on demand can be just as convenient as personal car ownership, if not more convenient. I can usually get a (Grab)taxi in under a minute, and by the time I get down to the lobby of my condo, it will be waiting for me. If I owned a car myself, I'd have to go down further to the parking garage, drive the car out of the garage, and of course I would also deal with parking when I get to my destination.

More importantly, a robotaxi will be much cheaper than personal vehicle ownership, safer, and nobody will have to drive. It's simply a better product, and I don't see any reason why people wouldn't completely switch to robotaxis in urban areas, except for perhaps being afraid of change.

Remote non-urban areas could be different, because people might have to wait much longer to get a robotaxi, so convenience may be much worse compared to private car ownership.

After all if all everyone cared about was going A to B safely in a car they'd just sell 80 million corollas a year instead of 80 million vastly different, often far more expensive, types of vehicles from vastly different companies, right?

Uber, Grab, etc. also offer different types of comfort with very basic cars and more luxury cars.

But the biggest reason for this is that a vehicle purchased for private ownership often has to fulfill many different use cases. Some people only need a simple small car, because all they use it for is to commute to work. Others, however, need a vehicle that can not only bring them to work, it also needs to be able to haul furniture, go on a yearly family road trip, carpool 7 kids to soccer, etc.

Different people use their car for different things. A robotaxi can be optimized to do just one thing:

• A 1-2 person cheap urban trip to or from work
• A 1-2 person urban trip in luxury with drinks and WiFi
• A 6-8 person group on its way to a party/restaurant/conference/etc
• A 4-5 person family going on a vacation. Wintersport, Vegas, visit family, etc.
• Transporting cargo/furniture/packages/letters

Don't get me wrong- Robotaxis would be a large cash cow for Tesla- and some percentage of owners who are someplace with enough demand relative to the # of Teslas in the area... Likewise it'd make all the Teslas NOT used as taxis hugely more appealing to folks who want their own cars to take them places.

But probably not quite the "everyone can earn huge money letting their free-to-them car drive around while they sleep and traffic will be solved!" panacea some folks imagine.

We definitely see things differently. Time will tell. It'll be exciting to find out what the future will look like in 2030

 

[PeterJA]

There's only a few hundred thousand taxi and limo drivers in the entire US right now. Nobody needs, or wants, "millions" of robotaxis here.

This reminds me of a quote (apparently apocryphal) attributed to an IBM chairman, saying that the world market for computers is five machines.

Personal computers are not mainframes. Robotaxis are not taxis.

Make robotaxi rides cheaper and more convenient than taxis and you will get more customers. Make them cheaper and more convenient than Uber and you will get more customers than Uber.

 

[StealthP3D]

I suppose it's not a huge deal if the car drives itself to have to travel a bit further from downtown-wherever to do it...but the roundtrip, plus charge time, will cut into the $/hr the robotaxi earns unless they're planning to alter future supercharger rollout to make more sense for this application.

(and that's not even covering the fact folks letting their cars robotaxi 12+ hours a day means they're gonna be hitting SCs multiple times a day- putting a LOT more demand on the network than what it was built for)

This analysis shows a distinct misunderstanding of exactly how many miles a Robo-taxi is capable of racking up working a typical urban environment. They would be doing well to average 22 mph (average speed throughout the day). That means in a 12 hour shift they could rack up 264 miles. These are urban miles which typically get the rated range or a bit better thanks to regen braking and slow average speeds.

In other words, no, they would not typically need to hit Superchargers multiple times in a 12+ hour day, once would likely cover them all day. Alternatively, they could spend their idle time at a central taxi zone with multiple 48 amp connections available and one attendant to keep the windows clean, check for debris in the cabins and plug them in until they had another call.

While you raise a lot of objections, none of them would prevent autonomous EV's from crushing the cost structure of traditional ride hailing services. And that is what will make it a real winner.

 

[MC3OZ]

I'm a retired software guy. I have great confidence that Tesla is far ahead of everybody else based on a single thing: they have pretty much thrown out all their software and rewritten it from scratch. Twice. Having the courage to do that is rare. But with what is essentially a research problem it is the only way to succeed. You have to build a real system, then throw it out when you finally understand enough to know how you got it all wrong. That takes money and time, both of which are almost never available in sufficient supply.

The only reason this is possible is because Elon is a software guy (among his many other talents).

Now this is no guarantee that Tesla will solve the problem. But I'm reasonably sure that nobody else will get there before them.

I'm a software 1.0 guy myself, and this is one big reason why I'm optimistic.

The other reason is I don't assume I know more a about software 2.0 than Elon and Karpathy.

I sort of understand software 2.0 and their general approach, what I like most of all (I can't l stress this too highly) is that it learns from its mistakes.

So within the limitations of what I can fully understand about the solution, I like what I see. So if Elon and Karpathy are confident, I'm confident, as they are far better placed to judge ...

It seems to me many people think driving a car qualifies them to comment on what software 2.0 can or can't do.

My other takeaway is we can't complain about a lack of effort form Tesla, a low budget or a lack of data and resources... there is a bunch of very smart people, trying very hard, and backed by whatever resources they seem to need.

So it will be a surprise if another company is first, simply because I can't nominate another project with the same amount of data, an equally qualified team and equivalent resourcing.. Leaving out data, the best of the completion perhaps goes close. But everyone claims data is all important for software 2.0.

 

[azaz]

Tesla Daily - half hour ago:

any new info?

 

[StealthP3D]

Spiegel is so obsessed with Elon/Tesla that someone in his vicinity should take action and seek counseling for him. This is not healthy anymore. I used to laugh at it and draw the comparison coyote/roadrunner like in the cartoon. But some of his investors or next of kin should take action..

An intervention would be the right thing to do. But I'm pretty sure he won't go quietly:

 

[Joshker]

RoboTaxi charging solution simple... Tesla knows who’s at the superchargers at all times. As the RoboTaxi arrives you send a message on the MCU and App to everyone already there asking for a plug in, you do the same for an unplug. In return you offer whomever accepts and does it within a minute or two minutes of the request a small supercharging credit. Software is amazing isn’t it?

 

[SpaceCash]

any new info?

No.
Basically, "S&P 500 inclusion will happen when it happens, which might be soon, or might be medium-soon, or could be much later, or possibly never, depending on whatever, because the committee literally makes up their own rules on the fly."

But my personal take is that the longer they put it off, the quicker the S&P themselves become irrelevant.

 

[Krugerrand]

The OP 'chose' to sell some of his TSLA and gave his reasons why. I may have missed it where he was trying to scare people (FUD) into selling theirs or indicating that others should simply follow his lead.

Enlighten me if I missed the part where he was encouraging others to sell. He was just giving his reasoning. I am not going to sell TSLA based on his reasoning but he makes a solid case for diversification.

Mark Cuban may advocate concentrated/one stock investing but most people I know/listen to like some diversification.

I will respect your final word on this discussion but for my part this is my last one on the subject.

I quoted you specifically. You said:

Selling makes sense when personal experience and/or belief/needs dictates.

Not a peep about choice in that statement of yours that I commented on.

I said it wouldn’t make sense to sell if the ‘belief’ was flawed.

I wouldn’t need to have the last word, if you hadn’t changed the context of the discussion.