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

[petit_bateau]

https://twitter.com/x/status/1630014622412550146

Thread here:
Tesla Compact Hatch Coming?

Ferragu sets out the 2/Z platform pretty well. We all know it is coming, but not when. But it has to come soon if Tesla is to have any chance of getting to Tesla's stated aim of 20m cars/yr in 2030.

However the kicker is in the last sentence,

"Moreover, at $25k MSRP, Model 2 addresses 80% of the car market and paves the way for Tesla to exceed 12m units and $400bn auto revenues in 2030..."

i.e. even Ferragu doesn't see a way to exceed 12m/yr in 2030, which is a long way off the 20m/yr that Tesla has often stated. And that in turn is the difference between $400bn auto revenues and $700bn auto revenues.

The reality is that Tesla needs to reclaim and recapture sliding market share by vehicles and by battery metrics if it is to get to 20m/yr and $700bn/yr by 2030. Which makes quite a difference to fair value share price, now.

 

[Artful Dodger]

Teslas built like Mattel "Hot Wheels"?
(Sorry if posted before....)

No, that's the video CtD posted 2 days ago. The one that was posted yesterday, then flagged private, is this one:

Tesla Gen 3 Platform Engineering Secrets Revealed - YouTube | Connecting the Dots (1h r35min)

This video seems to be back to 'publicly viewable' status now.

 

[jerry33]

I think if you change the time before reroute to more it doesn't reroute you for 10 min gain on not drive you on littles road on my system my reroute is set to only if 20 minutes gains try it

The default is five minutes, I've found that ten minutes works well for me.

 

[Artful Dodger]

The reality is that Tesla needs to reclaim and recapture sliding market share

Haha, no. Tesla only needs to bring Robotaxi to market. If Robotaxi hardware is sold for a 30% margin, Tesla can sell 7M of while keeping 3M themselves for the Tesla Network.

So there's the 20M:

1. 1M total S/X CT Semi and Roasters from the U.S.A.
2. 3M 3/Y from 4 plants worldwide
3. 6 Model 2 from 'x' plants worldwide
4. 10M Robotaxis from localized production

This has been obvious for literally years, ever since Elon made 20M/yr the goal. Tesla can exceed this if necessary by 2036 if other Automakers fail to produce EVs in quantiy, or simply fail as going concerns.

P.S. Tesla's 'Market Share' is not sliding, it's growing. And with zero-profits and looming bankrupcies for other automakers (including BYD in China), this is likely to get better for Tesla.

 

[Nolimits]

Haha, no. Tesla only needs to bring Robotaxi to market. If Robotaxi hardware is sold for a 30% margin, Tesla can sell 7M of while keeping 3M themselves for the Tesla Network.

So there's the 20M:

1. 1M total S/X CT Semi and Roasters from the U.S.A.
2. 3M 3/Y from 4 plants worldwide
3. 6 Model 2 from 'x' plants worldwide
4. 10M Robotaxis from localized production

This has been obvious for literally years, ever since Elon made 20M/yr the goal. Tesla can exceed this if necessary by 2036 if other Automakers fail to produce EVs in quantiy, or simply fail as going concerns.

P.S. Tesla's 'Market Share' is not sliding, it's growing. And with zero-profits and looming bankrupcies for other automakers (including BYD in China), this is likely to get better for Tesla.

But wouldn’t 10M robotaxi eat away at sales of other tesla models? Part of the idea of robotaxi was to replace the need for more cars

 

[Max Plaid]

Ferragu sets out the 2/Z platform pretty well. We all know it is coming, but not when. But it has to come soon if Tesla is to have any chance of getting to Tesla's stated aim of 20m cars/yr in 2030.

However the kicker is in the last sentence,

"Moreover, at $25k MSRP, Model 2 addresses 80% of the car market and paves the way for Tesla to exceed 12m units and $400bn auto revenues in 2030..."

i.e. even Ferragu doesn't see a way to exceed 12m/yr in 2030, which is a long way off the 20m/yr that Tesla has often stated. And that in turn is the difference between $400bn auto revenues and $700bn auto revenues.

The reality is that Tesla needs to reclaim and recapture sliding market share by vehicles and by battery metrics if it is to get to 20m/yr and $700bn/yr by 2030. Which makes quite a difference to fair value share price, now.

"sliding market share", what are you talking about? Tesla increases it's share of the global car market with every passing day

 

[Mengy]

This is still tracking for more than 2.0M vehicles this year assuming factories have normal uptime and Berlin and Texas ramps continue. If Ber/Tex each average 4k/week for the whole year then that's 2M, but they should keep ramping from here, and possibly could ramp explosively. Shanghai tripled its production from 4k/week to 12k/week over about a 12-month period. If something similar happened this year for Ber/Tex, we would get about 2.3M.

The big wildcard is whether there will be an extended production outage. Macro disruptions to supply chain or COVID shutdowns could happen again, but more importantly there might be huge upgrades to one or more of the factories that might take months to get back to full speed.

I think one thing many Tesla bulls don't take into account is converting the M3 to the Highland version this year. That is going to introduce M3 downtime which will end up decreasing M3 production numbers and sales for the year, and this might be why Tesla is only officially guiding for 1.8 million cars in 2023.

Sure they might do more, but I would not expect it because Tesla knows more than we do, and they have guided for 1.8 million.

 

[jerry33]

"sliding market share", what are you talking about? Tesla increases it's share of the global car market with every passing day

Sliding market share is just another instance of how ridiculous the concept of EV Market is. The only market is the automobile market.

 

[petit_bateau]

Haha, no. Tesla only needs to bring Robotaxi to market. If Robotaxi hardware is sold for a 30% margin, Tesla can sell 7M of while keeping 3M themselves for the Tesla Network.

So there's the 20M:

1. 1M total S/X CT Semi and Roasters from the U.S.A.
2. 3M 3/Y from 4 plants worldwide
3. 6 Model 2 from 'x' plants worldwide
4. 10M Robotaxis from localized production

This has been obvious for literally years, ever since Elon made 20M/yr the goal. Tesla can exceed this if necessary by 2036 if other Automakers fail to produce EVs in quantiy, or simply fail as going concerns.

P.S. Tesla's 'Market Share' is not sliding, it's growing. And with zero-profits and looming bankrupcies for other automakers (including BYD in China), this is likely to get better for Tesla.

Run that by me again.

The evidence is that the competition is growing their relevant market share faster than Tesla however you want to cut it. Before you say that the relevant competition is the entire vehicle market, note that the access gate to that vehicle market is cell supply, where Tesla's share is sliding. Yes loads of other vehicle companies may go bust en route, but those cells will still get to market in someone's vehicles.

Regarding dreams of Robotaxi, that would be nice, but the march of 9s has a long way to go and humans and the natural environment are very stupid and create so many edge cases that I am sceptical this will appear before 2030 at the earliest. What is more I expect that by the time it might appear there will be at least one close-follower, and once there is a close-follower then that follower will also go straight into competing RT fleets. So any additional margins due to RT operation will be rapidly eroded. The additional margins due to FSD might (or might not persist) as that is the equivalent of the Apple/Android situation in mobiles. But there isn't anywhere I've been that taxi is anything but a knife fight in a phone booth over margins, and do you really care who has marginally better software up front if both are good enough.

As a long-only Tesla shareholder I don't like to see this, but they are the indisputable facts we have in front of us.

 

[Singuy]

Run that by me again.

The evidence is that the competition is growing their relevant market share faster than Tesla however you want to cut it. Before you say that the relevant competition is the entire vehicle market, note that the access gate to that vehicle market is cell supply, where Tesla's share is sliding. Yes loads of other vehicle companies may go bust en route, but those cells will still get to market in someone's vehicles.

View attachment 911904

View attachment 911905

Regarding dreams of Robotaxi, that would be nice, but the march of 9s has a long way to go and humans and the natural environment are very stupid and create so many edge cases that I am sceptical this will appear before 2030 at the earliest. What is more I expect that by the time it might appear there will be at least one close-follower, and once there is a close-follower then that follower will also go straight into competing RT fleets. So any additional margins due to RT operation will be rapidly eroded. The additional margins due to FSD might (or might not persist) as that is the equivalent of the Apple/Android situation in mobiles. But there isn't anywhere I've been that taxi is anything but a knife fight in a phone booth over margins, and do you really care who has marginally better software up front if both are good enough.

As a long-only Tesla shareholder I don't like to see this, but they are the indisputable facts we have in front of us.

Can we stop using marketshare as some kind of credible measurement? It's only useful until after equilibrium has established. Before that date, it's kind of pointless.

 

[mongo]

Seems pretty clear to me this was the case.

Tesla has had this in the wings for some time. They were just hanging onto it until they could get some concessions when it was clear how chaotic getting funding through the states was going to be.

A week after Biden announces Tesla is participating in this arrangement and 3 weeks after Musk sits down and meets with the Biden administration; Tesla starts rolling out this solution to locations which are clearly not compliant with NEVI requirements.

This is a non-trivial effort. Tesla isn’t going to break even on these installations without some kind of government concessions.

Non-trivial from a modification point of view, or from sheer numbers?

From what I've seen, upgrades don't seem super intensive. Remove shroud, disconnect NACS only socket and wiring, install Magic Dock (some pedestal mods may be needed), and replace shroud.
Two teams of two doing retrofits:
Tesla prepares more Superchargers for Magic Dock installation

 

[CHGolferJim]

I am missing a discussion here on the implications of what CIA-director Bill Burns said yesterday.
CIA is sure that China is contemplating sending weapons to Russia, but that the decision in China hasn't been made yet.
In my opinion sending weapons would mean a step-up in the conflict, where USA will feel the need to respond, etc.
The military aspect of this doesn't belong here, but what it would mean for TSLA stock surely does.

Now, to say that I am nervous about the consequences for TSLA stock is a little too much said, but at this point the effects are unclear.
However, I am pretty sure TSLA will take a dive when this happens.
I wonder what effect we will see on the stock exchange today.
Hoping that the love of the Chinese for having a bustling economy will be put in first place.

My theory is this is a related strategy to releasing intel about Russia just prior to the Ukraine invasion.……that there are sufficient indications of the possibility, and the administration is publicizing it to increase international outcry against it as a deterrent.

 

[TrendTrader007]

I'll call it a mere conjecture, but one I consider to be well worth considering....it seems extremely likely that, as @Ogre points out, most of the "$27 to 40 billion worth of stock...even close to $60 billion in a single day" is indeed merely HFT, where each institution's algorithms are merely trying to constantly balance their value of shares held vs options held/sold at various price points rather than trying to increase / decrease shares to be held by the institution for any significant period of time. It's been well established that the volume of TSLA options (both volume outstanding and volume traded daily) greatly exceeds other security options trading; this HFT trading is indeed the "options tail wagging the security dog". If one conjectures that the bulk of the TSLA daily trading volume each day is typically such algorithmic balancing HFTs, then the *actual* volume of TSLA daily trading volume from humans (be they retail or be they fund managers) who for whatever reason want to buy or sell some shares may be significantly smaller. It may be shocking to some just how low that volume is.

Based on the above mere conjecture, @TrendTrader007's observation that "selling TSLA with market order in any sizeable order is a losing proposition" makes total sense. Those trades may seem small compared to the overall daily TSLA volume, but compared to the other actual human (retail or fund manager) buy/sell orders on the books, it's sufficient to move the overall price. It is a very easy to notice phenomena, frankly...even submitting market buy orders for 10k shares or 20k shares which should be easily matched instead moves the price quite noticeably, and based on commentary I assume your "sizeable order" size is far larger than my preferred block sizes.

This will likely be the object of many an MBA case study for decades to come...

in my opinion, 10,000 or 20,000 share blocks is the way to buy or sell TSLA with LIMIT orders. even market orders for as little as $2 million worth of tsla stock is heavily manipulated and gets poor fills. so, yes, i agree that HFTs are trading most of tsla stock with real alive human investors doing only a relatively minor fraction.
it becomes real easily clear with zero doubt if you panic sell 100,000 shares or more of tsla in a market order. then you get robbed so fast., it's not even funny
never doing that again.
plus, large orders are broken up into numerous, and i truly mean numerous fills and it takes like forever to fill those. so liquidity is only for those who are patient enough to put in no more than 10,000 share block orders with LIMIT otherwise good luck panic selling a larger market order. more sharks on wall street than pacific ocean. Liquidity is a mirage unless you are patient and not in a huge hurry to buy or sell TSLA
that is why i stay away from TSLL. even one of my smaller accounts, it will take me all day to slowly and painfully build up a position. i am just a very small individual investor/trader. imagine the plight of institutions. those guys, takes days to weeks to months to build up position in TSLA otherwise you end up running the price upon yourself. slow and painful and not for impatient.
that is why trading TSLA everyday is a total nightmare unless you are trading smaller accounts. i do find LEAPS to be much easier to buy or sell- never had any problems with size although it still moves the price but not as much as common stock in tesla. but pros and cons of LEAps merits another long post
not financial advice

PS: my CPAs literally laughed at me this friday when i submitted 225 pages of trading in 2022 and that was a bad performance year. 2023 will be better and lot more work for CPAs

 

[Mengy]

Run that by me again.

The evidence is that the competition is growing their relevant market share faster than Tesla however you want to cut it. Before you say that the relevant competition is the entire vehicle market, note that the access gate to that vehicle market is cell supply, where Tesla's share is sliding. Yes loads of other vehicle companies may go bust en route, but those cells will still get to market in someone's vehicles.

View attachment 911904

View attachment 911905

Regarding dreams of Robotaxi, that would be nice, but the march of 9s has a long way to go and humans and the natural environment are very stupid and create so many edge cases that I am sceptical this will appear before 2030 at the earliest. What is more I expect that by the time it might appear there will be at least one close-follower, and once there is a close-follower then that follower will also go straight into competing RT fleets. So any additional margins due to RT operation will be rapidly eroded. The additional margins due to FSD might (or might not persist) as that is the equivalent of the Apple/Android situation in mobiles. But there isn't anywhere I've been that taxi is anything but a knife fight in a phone booth over margins, and do you really care who has marginally better software up front if both are good enough.

As a long-only Tesla shareholder I don't like to see this, but they are the indisputable facts we have in front of us.

Marketshare does not matter, production and deliveries matter. If Tesla sells 20 million EV's per year and everyone else combined sells 200 million then Tesla only has a 9% marketshare, BUT they would still be selling 20 million cars per year!

Marketshare goes down as more EV models enter the market, that's just math and not a big deal.

 

[petit_bateau]

Can we stop using marketshare as some kind of credible measurement? It's only useful until after equilibrium has established. Before that date, it's kind of pointless.

GWh/yr is hardly a non-trivial measurement.

Marketshare does not matter, production and deliveries matter. If Tesla sells 20 million EV's per year and everyone else combined sells 200 million then Tesla only has a 9% marketshare, BUT they would still be selling 20 million cars per year!

Marketshare goes down as more EV models enter the market, that's just math and not a big deal.

On this trajectory Tesla gets to 12m/yr, not 20m/yr.

That is the point that Ferragu is making. And so am I.

It is quite relevant to the TSLA share price.

 

[mongo]

Run that by me again.

The evidence is that the competition is growing their relevant market share faster than Tesla however you want to cut it. Before you say that the relevant competition is the entire vehicle market, note that the access gate to that vehicle market is cell supply, where Tesla's share is sliding. Yes loads of other vehicle companies may go bust en route, but those cells will still get to market in someone's vehicles.

You're assuming a zero sum game.

If Tesla were to drop 18650s and someone else started using them, that use reduces Tesla's % of total cell usage, but doesn't impact Tesla's production plans.

Someone else using Crapium (tm) cells does not impact Tesla beyond raw mineral demand (in which case, expand to all global Li, Ni, Co, ... usage). Only in the case where Tesla's cell supplier allocates supply preferentially to a competitor is there a potential hit to their plans.

Also, percentage growth is easier when starting with a small number.

 

[cliff harris]

Regarding dreams of Robotaxi, that would be nice, but the march of 9s has a long way to go and humans and the natural environment are very stupid and create so many edge cases that I am sceptical this will appear before 2030 at the earliest. What is more I expect that by the time it might appear there will be at least one close-follower, and once there is a close-follower then that follower will also go straight into competing RT fleets.

I totally agree with your general scepticism regarding robotaxi before 2030. I definitely think Elons dream of everyone giving up car ownership for a flawless robotaxi servcie instead is not going to happen without major improvements to both the sensor suite and FSD computer.

But I do disagree that there will be any fast-follow.​

If we agree that the path to true robotaxi is with a machine-learning large-dataset model, then there simply is NOBODY even approaching the level of data collection in the real world that tesla have, and continue to accumulate. The big auto firms are paying lipservice to self-driving but not even equipping their vehicles with the sensors and connections required to collect the data (even assuming they had any staff trained to work with it).

It will be a decade, minimum, between a large rival equipping its fleet with a sensor suite and OTT collection back to a central location, and them actually achieving robotaxi, even assuming infinite budget and compute.

The elephant in the room is the possibility that robotaxi is achieved by some AI breakthrough not dependent on large-0data collection, but that seems to get less likely each year. And even then, I suspect elon & Tesla would realize this before anybody else.

Because I see zero risk of fast follow, I see FSD as nothing but upside, with only a slight risk of refund liability if FSD cannot be delivered soon.

ring-fenced LIDAR based mapping silliness like waymo/cruise is an unscalable joke that can be easily ignored imho.

 

[jw934]

Tesla wins the “Overall Loyalty To Make” award, overtaking Ford who has won that award every year for the past decade.

How Tesla overtook Ford in annual brand loyalty awards | Automotive N…

archived 27 Feb 2023 13:29:41 UTC

archive.is

 

[AndrewZ]

I totally agree with your general scepticism regarding robotaxi before 2030. I definitely think Elons dream of everyone giving up car ownership for a flawless robotaxi servcie instead is not going to happen without major improvements to both the sensor suite and FSD computer.

But I do disagree that there will be any fast-follow.​

If we agree that the path to true robotaxi is with a machine-learning large-dataset model, then there simply is NOBODY even approaching the level of data collection in the real world that tesla have, and continue to accumulate. The big auto firms are paying lipservice to self-driving but not even equipping their vehicles with the sensors and connections required to collect the data (even assuming they had any staff trained to work with it).

It will be a decade, minimum, between a large rival equipping its fleet with a sensor suite and OTT collection back to a central location, and them actually achieving robotaxi, even assuming infinite budget and compute.

The elephant in the room is the possibility that robotaxi is achieved by some AI breakthrough not dependent on large-0data collection, but that seems to get less likely each year. And even then, I suspect elon & Tesla would realize this before anybody else.

Because I see zero risk of fast follow, I see FSD as nothing but upside, with only a slight risk of refund liability if FSD cannot be delivered soon.

ring-fenced LIDAR based mapping silliness like waymo/cruise is an unscalable joke that can be easily ignored imho.

Do you know what the data collection is actually used for and where the concept originated? Mobileye + AWS were collecting data from a claimed 70million cars around the world as of the middle of 2022. NVIDIA has hardware in who knows how many vehicles. I think the big auto firms are mostly using various implementations of tech from Mobileye and NVIDIA, not their own in-house stuff. For example, Mercedes Drive Pilot is actually an implementation of NVIDIA's Drive Orin.

Tesla is currently using NVIDIA H100s for the actual neural net training. This is not Tesla versus just the OEMs, it's Tesla's vertically-integrated approach versus the OEMs + Mobileye, NVIDIA, and whoever else.

 

[philw1776]

All this talk about automobile sales #s and revenues by 2030 misses an essential point. Tesla calls themselves an AI company. As silly as the Optimus bot appears today, before 2030 derivative products will be creating large markets and producing substantive revenue. Tesla's leadership does not think in quarters or even years while planning strategy. It's decades.