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

[Doggydogworld]

There are three good reasons to use a P/S ratio:

1) Sales revenues is a very good indication of the demand for the company's products. Regardless of the gross profit margin, it indicates that the company is making things that the buying public wants.. A company with declining gross revenues may be profitable but in a declining industry or producing products that are no longer in favor. You may make golf clubs and have massive gross margins but if fewer people play golf then what is the future like?

2) Sales revenue is the easiest number to audit. A company may use "creative accounting" to brighten the bottom line - as many Tesla bears have accused Tesla of doing - insufficient warranty reserves etc. but it is very easy to arrive at the gross sales revenues by multiplying the number of vehicles delivered by an estimate of the ASP. Tesla couldn't add an extra $2 billion to the automotive revenues in a quarter without someone noticing that the ASP had soared to an unreasonable level.

3) Like point 1) - a company with growing revenues is a good candidate for a takeover if a buyer perceives the existing management team as being ineffective...so Tesla's gross revenues might attract a buyer who thinks that they have a better handle on the logistics of selling globally.

EV/Sales is a better metric. P/S makes highly leveraged companies look much cheaper than they truly are. As I found out the hard way a few decades ago (anyone remember Service Merchandise?).

P/S is also pretty useless for comparing across industries, e.g. Apple looks 17x more expensive than car dealership chain Autonation.

 

[humbaba]

I actually agree, but I also think his inability to post anything without screaming/crying/whining about "service hell" was getting really old. It was only his occasional informative posts that kept him off the ignore list for me.

I felt that they were more than occasional and respect his intelligence. Some people have an exaggerated expression online and are more reserved in person. I recently met @neroden and he was as quick to be informative there as here, while also being quite vigorous in his contentions regarding Tesla, $TSLA and related discussions.

While people will differ in how much (or little) good they derived from his posts, I for one am sorry to discover that he is gone. Still, if you recall he self-banned himself earlier this year but just couldn't keep himself from posting and came back. Which I appreciated, but perhaps a short (or long) stay will be what he needs. Well, I know this much, I'll be happy to read his posts if he is unbanned at some future point.

Amazing. Because in Maryland I see far more model 3’s then any other electric. Three model 3’s at work and zero leafs. (And two model S!)

Tesla is heavily concentrated in the US while Leaf is more global. A few months ago I'd projected when the M3 would take the crown from Leaf so I'm not surprised in the least.

Amusingly, despite the number of Tesla's I spotted recently while in the Philadelphia/New Jersey area there seems to be little awareness, but much more openness than near Detroit (where I was also recently). Near Detroit there was much greater awareness, but much derision. "Tesla Autopilot, isn't that supposed to be something like Super Cruise?"

What surprised me the most was the claim that GM's entire lineup would be steering wheel free in the next model year. When I asked for clarification, this is for planning model years and referred to vehicles made in 2025. When I expressed continued skepticism as to autonomous vehicles I was given the "you don't know what GM's got" shrug. Fine, maybe they'll shock the world, but when they can't even keep their 2019 deadline (which they finally admitted last month) I think I'm entitled to some skepticism that their entire lineup will be autonomous in six years.

Granted, this bit about no GM vehicle having a steering wheel starting in 2025 was far from official. And it may have been intended as nothing more than a barb (I drove up in a Tesla after all). But it did come from GM employee with enough seniority to lend plausibility to the claim as a plan.

One final thought: I drove around 2800 miles, almost entirely on autopilot. Twice it disengaged suddenly and without warning simply saying "autopilot is not available" though it started working again within short minutes. There was a navigation failure where I missed getting back on the interstate because it had stopped reducing the distance to the next junction so I thought it was still five miles away. It still takes most exits too fast and broadly (pray there isn't a guardrail at the edge of the lane). It still misbehaves at inconvenient times, forcing a disengagement which then results in missing an exit because autopilot wasn't engaged for that one second. It still mysteriously slams the brakes (highway shown as 70mph with cruise set at 75mph, but braking sharply to <50mph for no discernible reason). Other times when it would suddenly decide that the posted speed limit was 45mph and force speed to 50mph.

There are many faults to be had when driving the distances and roads and interchanges that I did. And yet, it is no exaggeration to say that I could not have made the trip without having an Autopilot enabled Tesla. Prior to getting my Tesla I'd stopped practically all non-local driving, but autopilot takes me from being the driver to the monitor -- it is a much easier role. Even more, halfway through the trip I became significantly ill and still I was able to complete the journey because it is that much easier to drive a car with autopilot.The last leg was in fact planned as a two-day drive to ease the per-day load, but I did it in a single day instead. When autopilot disengaged it wore on me so quickly that I wasn't sure I could continue -- but then it was able to resume. When I finally got back home I was quite tired, but I haven't been able really travel for nearly twenty years.

I get that not everyone is into Autopilot, but it is an amazing driving assistance package that has gotten more amazing over the seven months I've had my Tesla. The fact that it isn't perfect just means that there's still room for it to become even more amazing. And my car just finished updating...

 

[Beltsbear]

Nissan Leaf is a car that has been selling globally for years, whereas model 3 was very concentrated in the US initially.

Nissan deserves some kudos at least for what they have done with the leaf. It has been producing a 5-door EV with ~$30k base price since 2012, even if only at fifty thousand units a year. For anyone who is only after a economical, environmentally friendly city runabout as a 2nd car, a used 2016 Nissan Leaf with low mileage at $10k-$12k is a pretty good deal.

If the low priced leaf could charge at 100kw I would have bought one. Finally the plus versions can charge as almost as fast as a 2012 Tesla.

 

[EVDRVN]

Anyone get the $3K "sale price" for FSD via email today for EAP owners? Too bad it's $1k more than the previous offer.

 

[Joshker]

Anyone get the $3K "sale price" for FSD via email today for EAP owners? Too bad it's $1k more than the previous offer.

Agreed, or the $2k for AP on my non AP 3.

 

[Mokuzai]

Anyone get the $3K "sale price" for FSD via email today for EAP owners? Too bad it's $1k more than the previous offer.

My friend got that email even though they already had FSD. I purchased FSD on the last sale and got the following email which looks like they’re trying to entice current owners to upgrade. And honestly the psychological benefit of the free supercharging is a big factor for me if I ever do want to upgrade.

I also have another email address in Tesla’s marketing radar that isn’t associated with a Tesla account and got this email.

 

[BBone]

My friend got that email even though they already had FSD. I purchased FSD on the last sale and got the following email which looks like they’re trying to entice current owners to upgrade. And honestly the psychological benefit of the free supercharging is a big factor for me if I ever do want to upgrade.

View attachment 438604

I also have another email address in Tesla’s marketing radar that isn’t associated with a Tesla account and got this email.
View attachment 438605

Yes very tempting, so much that I pulled the buy trigger tonight. Our 2016 Model X required real world 3 charges for 820 km or so. The new 2019 Raven X should be able to do that with 1 charge and about 10 hours of driving. Very much looking forward to the higher charge rate, improved features, and of course the greater range. And FUSC/MCU2/AP3/FSD is great!

 

[ZeApelido]

I felt that they were more than occasional and respect his intelligence. Some people have an exaggerated expression online and are more reserved in person. I recently met @neroden and he was as quick to be informative there as here, while also being quite vigorous in his contentions regarding Tesla, $TSLA and related discussions.

While people will differ in how much (or little) good they derived from his posts, I for one am sorry to discover that he is gone. Still, if you recall he self-banned himself earlier this year but just couldn't keep himself from posting and came back. Which I appreciated, but perhaps a short (or long) stay will be what he needs. Well, I know this much, I'll be happy to read his posts if he is unbanned at some future point.


Tesla is heavily concentrated in the US while Leaf is more global. A few months ago I'd projected when the M3 would take the crown from Leaf so I'm not surprised in the least.

Amusingly, despite the number of Tesla's I spotted recently while in the Philadelphia/New Jersey area there seems to be little awareness, but much more openness than near Detroit (where I was also recently). Near Detroit there was much greater awareness, but much derision. "Tesla Autopilot, isn't that supposed to be something like Super Cruise?"

What surprised me the most was the claim that GM's entire lineup would be steering wheel free in the next model year. When I asked for clarification, this is for planning model years and referred to vehicles made in 2025. When I expressed continued skepticism as to autonomous vehicles I was given the "you don't know what GM's got" shrug. Fine, maybe they'll shock the world, but when they can't even keep their 2019 deadline (which they finally admitted last month) I think I'm entitled to some skepticism that their entire lineup will be autonomous in six years.

Granted, this bit about no GM vehicle having a steering wheel starting in 2025 was far from official. And it may have been intended as nothing more than a barb (I drove up in a Tesla after all). But it did come from GM employee with enough seniority to lend plausibility to the claim as a plan.

One final thought: I drove around 2800 miles, almost entirely on autopilot. Twice it disengaged suddenly and without warning simply saying "autopilot is not available" though it started working again within short minutes. There was a navigation failure where I missed getting back on the interstate because it had stopped reducing the distance to the next junction so I thought it was still five miles away. It still takes most exits too fast and broadly (pray there isn't a guardrail at the edge of the lane). It still misbehaves at inconvenient times, forcing a disengagement which then results in missing an exit because autopilot wasn't engaged for that one second. It still mysteriously slams the brakes (highway shown as 70mph with cruise set at 75mph, but braking sharply to <50mph for no discernible reason). Other times when it would suddenly decide that the posted speed limit was 45mph and force speed to 50mph.

There are many faults to be had when driving the distances and roads and interchanges that I did. And yet, it is no exaggeration to say that I could not have made the trip without having an Autopilot enabled Tesla. Prior to getting my Tesla I'd stopped practically all non-local driving, but autopilot takes me from being the driver to the monitor -- it is a much easier role. Even more, halfway through the trip I became significantly ill and still I was able to complete the journey because it is that much easier to drive a car with autopilot.The last leg was in fact planned as a two-day drive to ease the per-day load, but I did it in a single day instead. When autopilot disengaged it wore on me so quickly that I wasn't sure I could continue -- but then it was able to resume. When I finally got back home I was quite tired, but I haven't been able really travel for nearly twenty years.

I get that not everyone is into Autopilot, but it is an amazing driving assistance package that has gotten more amazing over the seven months I've had my Tesla. The fact that it isn't perfect just means that there's still room for it to become even more amazing. And my car just finished updating...

I don't know of another investor forum for a company with such quality of information provided by posters.

As an investor, I need that intelligent criticism to boost my confidence in TSLA valuation. Neroden provides key input.

I'd rather have annoying posts on service than massive stock losses.

 

[kbM3]

I believe this was already posted, but I just watched and wanted to add a couple of thoughts.

I was impressed by George, and more importantly Lex Fridman is very impressed with him. For those that don’t know, Lex is a professor of AI at MIT and has had top people from Cruise Automation and Waymo speak to his class and seemed to think very well of them.

George basically says that everyone’s approach to self driving aside from Tesla’s is wrong and just will never work. My impression was that he convinced Lex of this fact as well, though obviously Lex did not come out and say it.

George also said that everyone else’s approach stems from that early code that successfully passed the DARPA challenge early last decade (which I didn’t know). It kind of explains why Tesla is taking a different approach from everyone else, and all the others may just kind of be locked in.

There was also a funny anecdote where he just spouted out an idea to Elon about having a separate CPU with each camera, and Elon immediately called him on it and pointed out why it was a bad idea. George said that almost never happens because he can usually just dazzle people and then come back and correct later, if he’s wrong.

It just seems like there’s more and more evidence piling up that Tesla really is alone in the self-driving lead, contrary to appearances and industry consensus.

 

[Narrow Den]

I don't know of another investor forum for a company with such quality of information provided by posters.

As an investor, I need that intelligent criticism to boost my confidence in TSLA valuation. Neroden provides key input.

I'd rather have annoying posts on service than massive stock losses.

+100%

Never look a gift horse in the mouth

 

[kengchang]

My friend got that email even though they already had FSD. I purchased FSD on the last sale and got the following email which looks like they’re trying to entice current owners to upgrade. And honestly the psychological benefit of the free supercharging is a big factor for me if I ever do want to upgrade.

I also have another email address in Tesla’s marketing radar that isn’t associated with a Tesla account and got this email.

That was one thing Elon addressed during earning call, car might looks the same but it is very different car than 3-4 years ago. This email is the result of that.

 

[AlMc]

It is one great advantage of this forum that we want to learn, and reading comments from thoughtful opinionated people who disagree with us is something I treasure. I wish @neroden were back. For sure i'll disagree with much of what he posts, and will be irritated with some of it. Still I'll always learn something and that is the reason I am here.

Please mods, let his irritating self come back!!

Thank you for expressing your opinion, that while irritating at times (he knows he is), that you would like @neroden back.

I have also expressed that opinion but can't find my previous posts to that effect.

 

[RobStark]

Amazing. Because in Maryland I see far more model 3’s then any other electric. Three model 3’s at work and zero leafs. (And two model S!)

LEAF does much better in Europe and Asia than the US.

Inside the US LEAF's best market is the urban West Coast, Denver(Colorado has a $5k BEV incentive) plus for a time Atlanta (when Georgia had their $5k BEV incentive.)

 

[Tim S]

Tesla tweets pic and says "So, Gigafactory Shanghai is going well..."

Electrek story...
Tesla releases picture of Gigafactory Shanghai, latest drone video shows new structures - Electrek

 

[Cosmacelf]

But is he really?

“Narrow Den”, really? New account as of yesterday. No, not suspicious AT ALL. Carry on.

 

[Silent Ludicrosy]

Amazon has TSLA shorts!

Amazon.com: tsla shorts

Short squeeze secured.

 

[Fact Checking]

There was also a funny anecdote where he just spouted out an idea to Elon about having a separate CPU with each camera, and Elon immediately called him on it and pointed out why it was a bad idea.

Yeah, he doesn't explain in the interview why it's a bad idea to use GPUs locally at each camera, which makes it a nice riddle to figure out:

• Even 60 fps 1280×720p HD RGB raw camera feed bandwidth isn't that high: around 55 mpixel/sec per camera - which can be easily routed around a car from camera to the central GPU in a star topology. So there's no necessity to perform camera local GPU processing.
• For self-driving cars the major economic constraint is power use: if a Tesla consumes about 20 kWh to drive 100 km for about two hours in an urban environment, that's a continuous power consumption of about 10,000 Watts. A ~250W power consumption self driving system (100W central GPU, 50W cooling from a 200% heat pump, plus all sensors) will thus be an efficiency and range hit of about 2.5%, which is probably the pain threshold customers would tolerate.
• 8x weaker camera local GPUs will use significantly more power than a more powerful central unit, and it's also significantly more expensive to install: liquid cooling has to be duplicated 8 times, etc.
• Failover redundancy is easier to provision centrally as well.
• But the killer factor in favor of central GPU processing is the inherent data overlap in the actual pictures: if you composite the 8 frames into a "super-frame" and feed that large frame to a single neural network, a lot of the redundant data (same or similar objects visible in multiple cameras at the same time, etc.) can be discovered and shared by a large, unified, camera agnostic NN. This effectively compresses the data from the 8 cameras at a relatively early stage. With 8 distributed GPUs this kind of data sharing is less natural.

So yes, Elon is right, the proper approach for FSD vision is what the human brain does: aggressive centralization of data.

George said that almost never happens because he can usually just dazzle people and then come back and correct later, if he’s wrong.

Elon's pretty good at first principles physics, computing and engineering. This also makes him bull-sugar proof.

It just seems like there’s more and more evidence piling up that Tesla really is alone in the self-driving lead, contrary to appearances and industry consensus.

Nice quote:

George Hotz: "Elon talked about disengagements on the 405 [road] down in LA, [...] the lane marks were kind of faded and the MobileEye system would drop out" [30:45]​

This should put to rest the idea that Tesla isn't generating rich, fine-grained, geotagged AutoPilot disengagement data: they were already doing it back in the MobilEye days, with Elon being aware of special cases on specific road sections that would trigger disengagement.

Money quote for Tesla investors:

George Hotz: "Oh Tesla's gonna win level 5, they really are." [39:20]​

 

[RobStark]

My friend got that email even though they already had FSD. I purchased FSD on the last sale and got the following email which looks like they’re trying to entice current owners to upgrade. And honestly the psychological benefit of the free supercharging is a big factor for me if I ever do want to upgrade.

View attachment 438604

I also have another email address in Tesla’s marketing radar that isn’t associated with a Tesla account and got this email.
View attachment 438605

So refresh on Jan 1, 2020?

 

[Antares Nebula]

Wow! Big news. Means no more S,X updates this year; no fall refresh as many anticipated. Means "big changes" coming to Fremont, referred to by Jerome previously, likely refers to Model Y tooling/lines (although it could refer to new S,X lines if updates being released early next year).

Kinda makes sense as Raven update was recent, and a fall refresh would piss off those who purchased the Raven update. A delay into 2020 also makes sense from a tax drop off perspective. Introducing a major refresh after the final tax drop off would be a major way to boost demand.

BUT!! In the meantime, this delay will continue to affect ongoing demand of the high end S,X — as we can already see by the return of the Unlimited Free SuperCharging promo. Raven was a good first step but not enough to boost demand to the full market potential of the S,X. This is because of S,X inferior battery tech relative to Model 3 (no V3 max SuperCharger speed), as well as other inferior tech, see for example:

EM states Tesla's future is tied to the M3 and its cousin the Y. Articles like these who just how much engineering went into the M3 and why in many cases its a much more advanced vehicle than the S.

Tesla engineers share Model 3 steering, drivetrain, and suspension secrets

I'm starting to think that Model 3 packs simply cannot be burned.

テスカス on Twitter

I mean, any car can have an upholstery fire or whatnot. But if the pack doesn't ignite from damage like this, what does it take?

The pack is extremely good at containing single cell fires. It probably takes 70mph into a pole to get that far in. I think you will see small fires from crashes like that here and there but at 1/5 the rate of the S/X.

2019 was the year of the model 3, but it was also the year Tesla's flagships failed to shine, and how this affected high end sales revenue and margins and ultimately the bottom line, which in turn caused the stock to suffer precipitously. A MAJOR UNFORCED ERROR. Huge Tesla Bull, but I have to call out Tesla on this. I know others say the focus is on 3 and Y. But letting the flagships lag is not good marketing and leads to brand deterioration.

Well no point crying over spilled milk. Damage is done. Hopefully Tesla's many other upcoming catalysts, like Y and pickup, will keep up consumer and investor enthusiasm in the meantime. We get through this year, and next year with the Y and hopefully upgraded S,X and other new products and software on the horizon, we can look forward to brighter days.

Proceed with the downvotes (if I get as unpopular as Neroden, I'll take it as a badge of courage. Thank you!).

 

[mongo]

Wow! Big news. Means no more S,X updates this year; no fall refresh as many anticipated.

Best update you'll get != only update they'll get. (Unless you buy 2 S/X this year)

But letting the flagships lag is not good marketing and leads to brand deterioration.

S, like Roadster, was a means to an end. That end being the 3 (and future more affordable cars). If Tesla will forevermore be defined by the S (or the X which was not in the original plan) then they fail in a way.