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

[Wnuk]

Has any one heard of a SR (not plus) being delivered yet ?

Thought there was a report recently of the first 500 being produced but I have seen delivery reports yet.

 

[mongo]

And some day, the nvidia hardware will not be fast enough for the latest neural net. That’s when progress will stop for ap 2.x hardware, just as it stopped for my ap 1.0 car not very long after I bought it. At least ap2.x users will be able to upgrade.

HW2.5 is already too slow for the latest NN. With SW2.0 (development methodology), Tesla can run the NN training with different target topologies(HW2.5 and HW3). Due to the limitations, these targets differ in BN suze/shape, frame rate and image cropping along with feature set to achieve a high enough success rate against listed functions. This is likely part of why there is now the reduced function AP / full function FSD split. It reduced functionality to the level that HW2.x can perform well.

You & I are old-timers here. The youngsters are forgiven, but we know that Tesla promised before that they'd smooth out the S transit cycle many moons ago. They tried one quarter and gave up pretty much right away after. If all it took is one quarter to overcome it, Elon wouldn't list it as a reason to take the company private. He knows it is going to repeat over and over and over again. And if by chance there is a quarters with a little less severity then suddenly there'll be a fiscal change in some country and import tariff that expires, a new product that launches.

I am baffled that it is controversial to claim that Tesla is not in a steady state of operation. You guys are arguing nothing less than the short thesis : that this is the one exceptional quarter where they pull out all the stops to get sales and then it won't happen again. Newsflash. Not going to happen. If Tesla squeezes by today's all-out-effort than tomorrow that's the new normal that the company is expected to exceed. By its customers, by its shareholders, by its employees and most certainly by its executive team.

Are you referring to the Norway region surge reduction? I thought the use of RORO ships was part of the plan to reduce transit time to ease the delivery crunch.
Sure, there will be short term demand influencers. EoQ crunch will always depend on when the last ship arrives, unless they can somehow get ahead of demand and build up a buffer with some of their extra cash such that the last ship is there to refill their inventory. This may not be possible for a long time if word of mouth advertising grows faster than they can increase shipment. Good problem for a company to have, at least.

 

[EVMeister]

Has any one heard of a SR (not plus) being delivered yet ?

Over in the SR/SR+ waiting room thread it seems people are having their SR model delivery delayed - SR/SR+ Waiting Room.

 

[Fact Checking]

So I took one for the team and sampled today's $TSLAQ talking points:

• : Video being enthusiastically forwarded showing a Tesla not centering on a lane but swinging left and right. The TSLAQ genius peddling this story claimed to have "visual confirmation" that the car was on Autopilot before and after the video, but curiously didn't have the video storage space to actually capture this on the video itself... (Link.)
  
• : skabooshka of "I have documentary proof that Tesla cannot make more than 3,000 cars per week at Fremont" just a day before Tesla indeed made 4,997 Model 3's in a week in-fame started complaining that Twitter shadowbanned him, and fellow TSLAQ cult members are giving him support by retweeting. Since 'shadowbanning' on Twitter is one of the crazier conspiracy theories which Twitter doesn't do and never did, not the least because it would be singularly stupid by Twitter to implement on a public forum since it's so easy to prove that it's being done. Also, not unsurprisingly, none of his tweets seem to be shadowbanned to me - sadly! This incident too lends further support to the observation that TSLAQ cult members like stupid ideas, and that they are far out to 'la-la-land'. There's even a #freeskabooshka tag - I kid you not! (The tag is not trending at all, which suggests that Twitter as a whole is still an order of magnitude more intelligent than the TSLAQ cult.) (Link.)
• : I'd like to highlight the second word of skabooshka's message "Friends, fud-sters". Is he really asking for the help of fellow FUDsters, kind of giving away the game that he's only out there to get out the truth, while his brother might or might not be working at Volkswagen PR? I guess he could plead the 5th if ever confronted with that message - the right against self-incrimination is strong, at least as long as he stays on U.S. soil.
  
• : The usual TSLAQ FUD around the pricing gyrations of Tesla's 2025 bonds. The underlying trend is that these bonds are sensitive to SEC related FUD, as is clearly visible in September and October of 2018, when the SEC uncertainty forced bond traders out of positions of this otherwise pretty illiquid instrument. The TSLAQ cult tries to spin this kind of SEC uncertainty into solvency uncertainty, but it's only a halfhearted effort. It's hard to spin a company with $3.6b cash on hand in Q4 and no intention to raise cash from the bond market at all as a solvency risk after all. Yawn. (Link.)
  
• : Brand new TSLAQ Shorty Air Force video showing idled Tesla car carriers! Must be proof of no demand, no supply, fraud, in no particular order. The TSLAQ geniuses are again missing the small detail that there's only 7 days left from the quarter, and that much of Tesla's car carriers are back already from trips to faraway delivery centers, that 90% of the quarterly push is over already, and that only a small number of cars are left to be made and delivered to west coast delivery centers. But "logic" and "reason" are taboo words in $TSLAQ conversations so conspire away they do. (YouTube link and Twitter link for double the fun.)
• : TSLAQ cult members are never giving up the hope that their savior might be Elon stepping back from Tesla, voluntarily! (Link.)
• : Jim Cramer has gone mostly silent about Tesla since he attacked the federal judge hearing "SEC vs. Musk". Maybe his lawyer called him and explained to him that given the ... overly honest videos circulating where Cramer is admitting how short sellers are manipulating the market it's not the ... brightest of ideas to troll federal judges? Just in case she might be hearing ... his case in the future? (Link.)
  
• : Jim Chanos again peddling the flawed logic that because ICE incumbents with a 3 trillion dollars dying ICE industry firmly tied around their necks and not growing are valued at 0.5x revenue multiple, Tesla with its hyper-growth path and leadership position in the new 3 trillion dollars EV market must trade at a discount valuation too! Because a Tesla is just like a Ford Focus, just more expensive! (Link.)
• : Jim Chanos is the gift that keeps giving: in a tweet he proves it beyond doubt that he knows squat about the economies of scale of neural networks and billions of miles driven by a fleet of 600,000+ cars. A necessary consequence of his mistake is that his $TSLA valuation is off by at least $1,000. We are all rooting for Chanos to cover his short position as late as possible, for the general welfare of all TSLA longs. (Link.)

That's it for today folks, this was again a very entertaining waste of time over in TSLAQ-land! Could someone please sample the freshest Tesla FUD from Seeking Alpha? Your sacrifice will be remembered!

TL;DR: no new information worth mentioning. Sorry, should I have mentioned this first?

 

[Pras]

HW2.5 is already too slow for the latest NN. With SW2.0 (development methodology), Tesla can run the NN training with different target topologies(HW2.5 and HW3). Due to the limitations, these targets differ in BN suze/shape, frame rate and image cropping along with feature set to achieve a high enough success rate against listed functions. This is likely part of why there is now the reduced function AP / full function FSD split. It reduced functionality to the level that HW2.x can perform well.

.

Ability to process multiple cameras vs one or two will be big differentiator with hw3

 

[EVMeister]

Thanks for dipping your head into the cesspool so we don't have to, Fact Checking.

 

[mongo]

Ability to process multiple cameras vs one or two will be big differentiator with hw3

I'm pretty sure (but can't find a quote at the moment) that HW2.x is using all the cameras (part of NoA). However, the input data is cropped and running at a reduced frame rate.

 

[TheTalkingMule]

The more analyst downgrades of Model 3 production estimates the better.

 

[SteveG3]

What’s going on in premarket? TSLA is down 1.4% to $260.xx. Macros are not great, but other stocks are down a lot less.

I thought the news over the weekend was neutral/good.

First Update on the neighbor i-Pace, relative to “how well does this compete with MS/MX/M3/MY” of interest here, and so far I have learned a bit about the purchase decision:

• The buyer (a smart guy and career techie) purchased the car because he is a Brit and always wanted a Jag. He considered Tesla but believed the FUD, he worried about bankruptcy and Elon who is “clearly a genius but does all those crazy tweets” (...snip, bolding above added)

• For years now, literally the overwhelming majority of mainstream media has tried to undermine public confidence in Tesla, Elon Musk and their products... even create distaste for them.
  
  (...)
  
  Elon/Tesla can't stop it.
  
  We can't stop it.
  
  The best we can do, is get the message at least as widely out there that the source of this attempt to deligitimize Elon/Tesla & the products IS NOT LEGITIMATE JOURNALISM.
  
  (...)
  
  The best we can do via social media is get the point across it is not a competition of Tesla vs. ICE mfg products, but, rather,
  
  like the proverb of "teaching a man to fish rather than giving him a fish," it is a question of whether the public just takes for granted the BS about Tesla the media has flooded us all with, or actually learns directly about what Tesla is offering and how it can improve our world. Never mind rebutting the million false narratives blasted out to virtually every human not living under a rock... get it well known to as close to as many people as possible that their perception of Tesla/Musk/Products has been polluted by a massive machine pumping out false narratives. This can be done with simple concise messaging.

  
  and it’s not just FUD with a target of consumer confidence (real world impact to Tesla sales reflected in the second quote above)... supplier and investor confidence are targets as well.

Recommendation to most effectively respond to this in the last few paragraphs of the last quote included above (just click on “expand” if bolder portion of the quote is not visible).

 

[Pras]

I'm pretty sure (but can't find a quote at the moment) that HW2.x is using all the cameras (part of NoA). However, the input data is cropped and running at a reduced frame rate.

Right . It’s the FPS that is the qualifier. But for a vision based system, that will be such a big improvement (like using glasses for weaker eyes)

 

[Carl Raymond]

This I find a common error with non-native English speakers, here in Belgium, the up-coming Tuesday would be referred to as "next Tuesday", where as in English this indeed refers not to 26/3, but rather 2/4.

Correct usage is: Tuesday, next Tuesday, the Tuesday after next.

Feel free to nit-pick my FR, NL, D and DK too...

Tuesday, Tuesday week, Tuesday fortnight, is unambiguous.

 

[Dan Detweiler]

So I took one for the team and sampled today's $TSLAQ talking points:

• : Video being enthusiastically forwarded showing a Tesla not centering on a lane but swinging left and right. The TSLAQ genius peddling this story claimed to have "visual confirmation" that the car was on Autopilot before and after the video, but curiously didn't have the video storage space to actually capture this on the video itself... (Link.)
  
• : skabooshka of "I have documentary proof that Tesla cannot make more than 3,000 cars per week at Fremont" just a day before Tesla indeed made 4,997 Model 3's in a week in-fame started complaining that Twitter shadowbanned him, and fellow TSLAQ cult members are giving him support by retweeting. Since 'shadowbanning' on Twitter is one of the crazier conspiracy theories which Twitter doesn't do and never did, not the least because it would be singularly stupid by Twitter to implement on a public forum since it's so easy to prove that it's being done. This incident too lends further support to the observation that TSLAQ cult members like stupid ideas, and that they are far out to 'la-la-land'. There's even a #freeskabooshka tag - I kid you not! (The tag is not trending at all, which suggests that Twitter as a whole is still an order of magnitude more intelligent than the TSLAQ cult.) (Link.)
• : I'd like to highlight the second word of skabooshka's message "Friends, fud-sters". Is he really asking for the help of fellow FUDsters, kind of giving away the game that he's only out there to get out the truth, while his brother might or might not be working at Volkswagen PR? I guess he could plead the 5th if ever confronted with that message - the right against self-incrimination is strong, at least as long as he stays on U.S. soil.
  
• : The usual TSLAQ FUD around the pricing gyrations of Tesla's 2025 bonds. The underlying trend is that these bonds are sensitive to SEC related FUD, as is clearly visible in September and October of 2018, when the SEC uncertainty forced bond traders out of positions of this otherwise pretty illiquid instrument. The TSLAQ cult tries to spin this kind of SEC uncertainty into solvency uncertainty, but it's only a halfhearted effort. It's hard to spin a company with $3.6b cash on hand in Q4 and no intention to raise cash from the bond market at all as a solvency risk after all. Yawn. (Link.)
  
• : Brand new TSLAQ Shorty Air Force video showing idled Tesla car carriers! Must be proof of no demand, no supply, fraud, in no particular order. The TSLAQ geniuses are again missing the small detail that there's only 7 days left from the quarter, and that much of Tesla's car carriers are back already from trips to faraway delivery centers, that 90% of the quarterly push is over already, and that only a small number of cars are left to be made and delivered to west coast delivery centers. But "logic" and "reason" are taboo words in $TSLAQ conversations so conspire away they do. (YouTube link and Twitter link for double the fun.)
• : TSLAQ cult members are never giving up the hope that their savior might be Elon stepping back from Tesla, voluntarily! (Link.)
• : Jim Cramer has gone mostly silent about Tesla since he attacked the federal judge hearing "SEC vs. Musk". Maybe his lawyer called him and explained to him that given the ... overly honest videos circulating where Cramer is admitting how short sellers are manipulating the market it's not the ... brightest of ideas to troll federal judges? Just in case she might be hearing ... his case in the future? (Link.)
  
• : Jim Chanos again peddling the flawed logic that because ICE incumbents with a 3 trillion dying ICE industry firmly tied around their necks and not growing are valued at 0.5x revenue multiple, Tesla with its hyper-growth path and leadership position in the new 3 trillion dollars EV market must trade at a discount valuation too! Because a Tesla is just like a Ford Focus, just more expensive! (Link.)
• : Jim Chanos is the gift that keeps giving: in a tweet he proves it beyond doubt that he knows squat about the economies of scale of neural networks and billions of miles driven by a fleet of 600,000+ cars. A necessary consequence of his mistake is that his $TSLA valuation is off by at least $1,000. We are all rooting for Chanos to cover his short position as late as possible, for the general welfare of all TSLA longs. (Link.)

Folks that's it for today, this was again a very entertaining waste of time over in TSLAQ-land! Could someone please sample the freshest Tesla FUD from Seeking Alpha? Your sacrifice will be remembered!

TL;DR: no new information worth mentioning. Sorry, should have started with this?

I bet you feel dirty after sampling all of that. I know I do. lol!

As to the self proclaimed "fellow FUD-sters" comment. There is no need for them to deny it. Everyone knows what they are doing. Nobody will hold them accountable...and the worst thing, it's working.

Dan

 

[Fact Checking]

Not much TSLA related news today, so far:

JMP Securities reiterated their 'Market Outperform' rating with a bit lower price target: from $406 to $394.

Their comments were:

"We are still buyers of the stock – current challenges do not change the underlying competitive story, in our opinion. It is worth reiterating that our investment stance on Tesla has always been based on the potential the company has to make competitive gains over time. The undeniably challenging environment that Tesla faces at the moment is not enough to impact our fundamental stance on the company and its prospects. Our methodology continues to be based on 2020 target multiples of 2.5x EV/Revenue, 15x EV/EBITDA, and 35x price/free cash flow, discounted back to the present at 10% per year. We note that 10% is higher than the 8% discount rate we normally use, which reflects the relatively greater level of risk that we associate with an investment in Tesla."​

For finance geeks here's the underlying calculations I can see:

• Revenue multiples: 2.5x revenue multiple makes $29b in 2019 map to $72.5b Enterprise Value:
  • which with 172.72m shares outstanding is $420/share, I'm sure they didn't want to mention that outcome,
  • fortunately GAAP-diluted shares are at around 178m, which gives a more politically correct $407/share valuation.
• EBITDA multiples: Taking @ReflexFunds's Q4 EBITDA of $1.48b and JMP Securities's 15x multiple we TTM EBITDA of $5.92b and Enterprise Value of $88.8b: around $514/share valuation.
• Cash multiples: They are using a 35x price/FCF multiple, which with @ReflexFunds's adjusted FCF of $0.66b makes $93b of Enterprise Value, or about $539/share.

How did JMP Securities get to around $400? They either took the lowest one (the revenue multiple), which is conservative without adding a growth premium, or they are assuming a weaker Q1 than Q4 which I don't think is going to happen, and this is dragging their valuation numbers down a bit by about 10-20%.

Anyway, bullish AF just based on financials, without having to assume much about the "Tesla Story", which is a new thing after Q3 and Q4.

(Another piece of recent news is that yesterday Ars Technica's Timothy B. Lee managed to spin Elon's sur-reply in a negative light - which his readers were, unsurprisingly, lapping up. Took him two days to come up with the twisted logic to present the sur-reply in a negative light while technically not lying about it overly obviously, so he was late to the party and there's no need to link to the article. I'm wondering how he's going to spin Elon's eventual court win negatively - but he seems very eager when it comes to Tesla FUD so I'm sure he'll find a way. )

Update:

A predictable TSLA "downgrade" just in from a Tesla perma-bear:

RBC Capital analyst Joseph Spak lowered the price target on Tesla (NASDAQ: TSLA) to $210 (from $245) while maintaining an "Underperform" rating after lowering their Q1 delivery estimates by 8.5%.

Spak cut their 1Q19 Model 3 delivery forecast to 52.5K from 57K, versus the consensus 58.4K. The analyst cited "meager demand and some M3 delivery issues abroad."​

Firstly, Joseph Spak only ever had 'hold' and 'sell' ratings on Tesla, which is perhaps not surprising given RBC's occasionally reported Tesla short position.

Secondly, his 52.5k deliveries expectation is in stark contradiction with:

• Elon's email suggesting a record Q1 delivery push which includes North America,
  
• Gigafactory sources reporting consistent 6k-7k Model 3 battery pack production,
• 16 ships carrying somewhere between 38k and 52k cars, and all but one ship having been unloaded already and the final ship due to arrive in about 30 minutes at Tianjin,
  
• Alpha Hat measuring 11.5k January U.S. deliveries alone,
• Bloomberg projecting ~75k production in Q1.

BTW, we now know that with about a week to go until the Q1 production & deliveries report current Wall Street consensus for deliveries is 58.4k units.

 

[anon]

This I find a common error with non-native English speakers, here in Belgium, the up-coming Tuesday would be referred to as "next Tuesday", where as in English this indeed refers not to 26/3, but rather 2/4.

Correct usage is: Tuesday, next Tuesday, the Tuesday after next.

Feel free to nit-pick my FR, NL, D and DK too...

As a native English speaker, I actually say next Tuesday to mean 26/3, but I have occasionally heard people give your interpretation too. I'm curious what other people think about this, as it can sometimes cause confusion and I'd like to get a better idea of the consensus. It may be one of those things that's different even in different parts of the UK, let alone other English-speaking countries.

 

[SteveG3]

Update:

A predictable TSLA "downgrade" just in from a Tesla perma-bear:

RBC Capital analyst Joseph Spak lowered the price target on Tesla (NASDAQ: TSLA) to $210 (from $245) while maintaining an "Underperform" rating after lowering their Q1 delivery estimates by 8.5%.

Spak cut their 1Q19 Model 3 delivery forecast to 52.5K from 57K, versus the consensus 58.4K. The analyst cited "meager demand and some M3 delivery issues abroad."​

Firstly, Joseph Spak only ever had 'hold' and 'sell' ratings on Tesla, which is perhaps not surprising given RBC's occasionally reported Tesla short position.

Secondly, his 52.5k deliveries expectation is in stark contradiction with:

• Elon's email suggesting a record Q1 delivery push which includes North America,
  
• Gigafactory sources reporting consistent 6k-7k Model 3 battery pack production,
• 16 ships carrying somewhere between 38k and 52k cars, and all but one ship having been unloaded already and the final ship due to arrive in about 30 minutes at Tianjin,
  
• Alpha Hat measuring 11.5k January U.S. deliveries alone,
• Bloomberg projecting ~75k production in Q1.

BTW, we now know that with about a week to go until the Q1 production & deliveries report current Wall Street consensus for deliveries is 58.4k units.

Maybe we could quickly crowd source a list of the 4 or 5 notes out the past week pushing out forecasts of 45K to 53K Model 3 deliveries. We could try via social media (Reddit, Twitter, comment sections of articles on these gamed forecasts...) to get these calls out there with the context re 1) how far off are these forecasts? (10%? 20%? 30%? 40%? 50%?) 2) will the media report on how vastly inaccurate these bearish calls have been when actual come out next week? 3) if these analysts are really providing expert info, why can “amateurs” forecast with such dramatically better accuracy? 4) have you ever thought about the sheer volume of unflattering but incorrect messaging re Tesla and its products coming at us from nearly all directions?


again,

The best we can do via social media is get the point across it is not a competition of Tesla vs. ICE mfg products, but,

like the proverb of "teaching a man to fish rather than giving him a fish," it is a question of whether the public just takes for granted the BS about Tesla the media has flooded us all with, or actually learns directly about what Tesla is offering and how it can improve our world. Never mind rebutting the million false narratives blasted out to virtually every human not living under a rock... get it well known to as close to as many people as possible that their perception of Tesla/Musk/Products has been polluted by a massive machine pumping out false narratives. This can be done with simple concise messaging.

 

[K_Dizzle]

The more analyst downgrades of Model 3 production estimates the better.

Go on

 

[KarenRei]

Go on

They're saying that they want the expectations to be low so that there's a big spike afterwards.

 

[AlMc]

You make the case that the tweet was ill advised. You have made no case that the tweet was 'wrong' (i.e. material. i.e. in breach of the SEC agreement).

SEC'''Short Enhancement Commission'...funny, probably true but ill advised...again, IMO. After the 'settlement'..just produce and let surrogates take on the SEC via twitter and just 'like' posts you agree with.

 

[Carl Raymond]

​

As a native English speaker, I actually say next Tuesday to mean 26/3, but I have occasionally heard people give your interpretation too. I'm curious what other people think about this, as it can sometimes cause confusion and I'd like to get a better idea of the consensus. It may be one of those things that's different even in different parts of the UK, let alone other English-speaking countries.

Next Tuesday will always confuse somebody, it’s clear as mud. Say “Tuesday” or “Tuesday week” or “Tuesday fortnight”.

 

[Unpilot]

Next Tuesday will always confuse somebody, it’s clear as mud. Say “Tuesday” or “Tuesday week” or “Tuesday fortnight”.

How about Next Tuesday (--/--/2019