TSLA Market Action: 2018 Investor Roundtable

[kbM3]

Lol, I love this company and have quite a bit of my net worth invested into it. But do you own a tesla? They do amazing things, but I have yet to feel as though they've given me exactly what I was promised. The only "more" might be the occasional Easter egg.

S,X and Model 3 all have more features and are just generally better than what was shown during the unveil. This is over-delivering.

 

[defc0n]

Q1 and Q2 are still too scary for me to jump back in right now. I thought we’d see 260 again sometime the first half of this year. Lot of my money is tied up in LIT still and it hit a 5 year high last month but has taken a 16% beating since then.

I think my new plan is to keep holding LIT and buy TSLA if we see 260 from now until Q3. Ill decide whether or not to jump in then based upon events. Lower number of S and X deliveries and questions about Tesla raising capital for model Y etc. might freak people out these upcoming few months.

I’m sure I’ll be wrong about everything

 

[myusername]

Don't kid yourself, anyone in the stock market is gambling. Some take higher risks for greater reward. Others take much higher risks for much greater reward. I don't think you can truly say that some are gambling and others are not.

From Wikipedia:
Gambling is the wagering of money or something of value on an event with an uncertain outcome with the primary intent of winning money or material goods. Gambling thus requires three elements be present: consideration, chance and prize.

Some take higher risk on lower capital while others take lower risk on lots of capital which in the end is the same. I prefer the former. And selling options on a high risk growth stock is madness.

 

[Ulmo]

So ......

 

[Ulmo]

They just posted an 8-K clarifying Elon's comments on the earnings call. Their Q1 target of 2.5k/week is NOT dependent on the new Grohmann line arriving in March. The new Grohmann line will add 2-2.5k of additional module capacity.

Just added some more after hours...

That sounds more like a correction than a clarification.

Anyway, even though I think that, there is a tie in that makes it less of a surprise for those who had listened to prior quarterly conference calls regarding the same topic: Elon explained in those earlier quarters (was it just 3 months ago? or 6? I think 3 --- the very prior one(?) where he explained the Model 3 battery pack assembly bottleneck) that there were different quality and speed of machines for that line coming; I think he mentioned at least 4 manufacturing systems in the pipeline that would come online soon if not already online. From my rather poor memory:

• Temporary manual
• The not-working machine
• Some better machines not yet online
• Some really super better machines not yet online.

I guess the German machines are one of the latter two, or similar. By the way, this conference call included a new category of semi-manual that I don't recall from the prior explaination, but certainly makes sense (and I may have just forgotten him mentioning it before). In addition, in the prior conference call, there was a fifth category: they're putting a huge focus on that bottleneck to fix it no matter what as well as can be done in a time efficient way. So, in time order but counterrespectively, that could be called the "no matter what machines" and the "human-machine hybrid machine (let's just call it the cyborg battery pack assembly subline), which in my view, the latter is one of the former. I wouldn't be surprised if the German machines were also influenced to some degree by the no-matter-what emphasis, but for the sake of good quality, that pressure should have been presented to them as an opportunity to have fun creating an excellent solution, not as an oppressive impossible burden. I especially like the note that the German machines already work. If they stop working in what Hollywood depicts as the sandy state, it shouldn't be hard at all to figure out why and cause them to start working properly again.

Everybody needs to chill. Starman had a great hint with ‘Don’t Panic’.

FYI, Elon Musk on Twitter, which if like me you had read Douglas Adams' relevant book(s), would have explained that sign. It was there back when that was tweeted, so not directly time related.

The fun part is that me as a non native speaker did not read out of the CC what all you heard.

I find it fun that you had somehow heard that interpretation more "accurately", and you are from the place in question, whether or not they're related. Of course, this requires me to go back and listen carefully, as a California native, just out of politeness, so as not to make accusation, regardless of all of that. (I'm not trying to put you down in any way, regardless.)

Here's my transcript: All quotes are Elon unless otherwise noted. I use parenthesis to mean square brackets since TMC coding uses square brackets for other uses. Elipses are skipping. Time point is in the conference call at IR.Tesla.Com. Until the very last "paragraph" (after my "bingo"), I hurredly copied, so it is more error prone. You can see I put some detail into the last paragraph.

6:19: “Don’t read too much into this.”
…
“There are 4 zones in module production.”
…
“Of the 4 zones, 2 of them … didn’t work.”
…
“18 month development cycle for production system of that scale and complexity and try to do that in basically 6 months, maybe a little … 6 to 9 months.”
…
“All of Grohmann’s machines have been to spec.”
…
“(line) 1 & 2 … from Grohmann” “a single Tesla Grohmann line to be equivalent to 3 if not 4 of the current lines we have and be smaller.”
…
“We have a semi-automatic line … series of automated stations manned by people … remarkably effective”
…
“ability of people to adapt … has been remarkable.”
…
“semi-manual semi-automatic is exceeding all 3 automatic line right now”
...
“not artisanal”
​

I’d love to be one of the investors to see it. Can I be part of that arrangement?

(JB) “intense focus”

…
​

I didn’t hear any specific run rate pre or post Grohmann. It makes sense that 2,500 could be with the encumbered ones without Grohmann.

​

bingo; I spoke too soon:

(re: Grohmann machines) 11:58: “It’s not a question of whether it works or not; it’s just a question of disassembly, transport and reassembly. Um, (JB: “exactly”), so that’s, you know, the, yeah, um, and so so we expect to alleviate that constraint, um that; uh ok, with alleviating that constraint, that is that’s what gets us to roughly uh two to two and a half thousand unit per week, uh, ra- production rate, uh, The next constraint would be material conveyance" (unrelated topic)
​

I can see a new paragraph being formed at the comma of the Um after "reassembly" in Elon's mind, which means that the 2k...2.5k/week comment was regarding the whole discussion of the battery bottleneck, not the Grohmann machines. If there was no new paragraph after "reassembly" and before "we expect", then another interpretation is with the same outcome: "we expect to alleviate that constraint", which even with the "so so" before that (an Elon stuttered "so", not English phrase "so-so"), cleanly does not glue Grohmann into the alleviation, but nor does it specifically exclude it. He did raise his tone slightly, enough to bridge past back in time to the higher topic pre-Grohmann subroutine, said bridge starting with "so", but not necessarily so just from tone (it was too slight to be certain, but I don't know Elon that well, so my lack of knowing him would encumber that bridge determination). The fact he then immediately went to an unrelated constraint shows that that new speech is completely raised past the battery module bottleneck, so just by that one step change in topic alone, the prior sentence could have been, by logic, any level between Grohmann (inclusive) and material conveyance (noninclusive), which does slightly pressure the interpretation of production rate to be less weighed down to the lower levels of Grohmann.

That's not to misinterpret Grohmann as being lower quality; in fact, I was very impressed with the ~" all the machines from Grohmann perform to spec "~ comment.

 

[Reciprocity]

That sounds more like a correction than a clarification.

Anyway, even though I think that, there is a tie in that makes it less of a surprise for those who had listened to prior quarterly conference calls regarding the same topic: Elon explained in those earlier quarters (was it just 3 months ago? or 6? I think 3 --- the very prior one(?) where he explained the bottleneck) that there were different quality and speed of machines for that line coming; I think he mentioned at least 4 manufacturing systems in the pipeline that would come online soon if not already online. From my rather poor memory:

• Temporary manual
• The not-working machine
• Some better machines not yet online
• Some really super better machines not yet online.

I guess the German machines are one of the latter two, or similar. By the way, this conference call included a new category of semi-manual that I don't recall from the prior explaination, but certainly makes sense. In addition, in the prior conference call, there was a fifth category: they're putting a huge focus on that bottleneck line to fix it no matter what as well as can be done in a time efficient way. So, in time order but counterrespectively, that could be called the "no matter what machines" and the "human-machine hybrid machine (let's just call it the cyborg battery pack assembly subline), which in my view, the latter is one of the former.

I think temporary manual solution is the semi automatic solution and always was. It's a mix of the machines that never fully worked and people. It's the only thing that makes sense as they wouldn't develope a new half solution machine then start over again to develope the new machines. If there is now a semi automated solution, it must be using the less then ideal original machines that never fully worked. They probably fixed them as much as they could with programming and then used people to fill in the gaps, which appears to mostly be moving materials between different processes.

This new update is great but now we need to worry about the automated parts conveyance system. Luckily, that seems to be a programming problem, though described as a very complicated software solution. I have zero concern that they wouldn't be able to improve that system in steps so that it can support 1k/w, then 2, then 3 and so on. To me those 2 major risks are way more ideal then issues with the automated welding or other production line issues, though we don't know what we don't know. My guess is that they have been able to run the line in burst mode to work through those issues over the past few months. I believe Elon even stated that parts of the line had been shown to support up to 6k/w already back in the last quarterly CC in an answer about capex requirements to go from 5k-10k. It's certainly a very complex line with many unique sub stations that all must run at very high speeds and very low error rates. I find it hard to believe that no one has thought to just run everything faster. My guess is that the software is what makes it possible. The timings must be very complicated and there must be a great deal of feedback and maybe even vision recognition where they are using NN to learn and keep everything in sync as well as image welds and other parts to verify they are good. With enough data feedback, the system could automatically adjust to many different variables that might occur during the process. For instance, if part of the line must slow down waiting for parts. This software would not be an after thought, but something simulated well before the hardware showed up. Tesla probably also worked with kuka to confirm the speeds the robots could operate at. But even then, the software could adapt to a kuka that slowed due to wear and warn the maintenance crew to needed repairs.

If they could just get over this damn hurdle, the next 500K is just copy and paste. Hopefully Y is just a software revision and then copy/paste the hardware to New factories (way over simplified but you get my drift).

 

[Ulmo]

I can't stand this TMC 60 minute time limit on edits.

I repeat in blue, and add in green:

I can see a new paragraph being formed at the comma of the Um after "reassembly" in Elon's mind, which means that the 2k...2.5k/week comment was regarding the whole discussion of the battery bottleneck, not the Grohmann machines. If there was no new paragraph after "reassembly" and before "we expect", then another interpretation is with the same outcome: "we expect to alleviate that constraint", which even with the "so so" before that (an Elon stuttered "so", not English phrase "so-so"), cleanly does not glue Grohmann into the alleviation, but nor does it specifically exclude it. He did raise his tone slightly, enough to bridge past back in time to the higher topic pre-Grohmann subroutine, said bridge starting with "so", but not necessarily so just from tone (it was too slight to be certain, but I don't know Elon that well, so my lack of knowing him would encumber that bridge determination). The fact he then immediately went to an unrelated constraint shows that that new speech is completely raised past the battery module bottleneck, so just by that one step change in topic alone, the prior sentence could have been, by logic, any level between Grohmann (inclusive) and material conveyance (noninclusive), which does slightly pressure the interpretation of production rate to be less weighed down to the lower levels of Grohmann. But, the strongest interpretation by far is that Elon likes to conclude a topic, so the stuttered "so so" gave him time to contemplate and put forth his concluding statement of the biggest bottleneck topic, with the biggest bottleneck run rate statement, making that statement fully raised to the battery bottleneck topic level. RTS from Grohmann back to lines, RTS back from lines into module assembly, RTS back from module assembly back to biggest assembly bottleneck ("Module production is fundamentally the limiting factor on Model 3 output"), which resumes with its summary (run rate prediction) (a subroutine right under "day to day battles" which is under "Churchill"). (Funnily, "limiting factor" topic was created after "module pack", even though it was father to the earlier topic, inserted between "day to day battles" and "module pack" essentially as an afterthought edit, so "limiting factor" topic became parent node to "module pack" and child of "day to day battles".)

Ok, I relent; it is more likely it was actually a clarification, even though it sounded like a correction when I first saw it.

 

[Ulmo]

This new update is great but now we need to worry about the automated parts conveyance system.

Ha! I've been saying I've been wondering if the thing would work better if parts should be thrown by robots to other robots in the factory (even dropped from above), and cars should be in assembly zones, rarely moved, with parts flowing to them to robots that move (slightly from car to car) in and out to each car as the multifunction robots are needed for the various parts. I've always thought it was crazy a multiton product was the item on the conveyer belt rather than its components. Elon did talk about airflow for parts transport. But there are some rather cubic buildings for inventory on both sides of the Fremont factory. Elon mentions this topic as a constraint, and I believe him, but I think it would be fantastically fun to solve given Elon's thoughts on methods to solve it; if he isn't reigned in to failure by the Board, this topic should come with some variety of possibilities for good efficiency improvement by orders of magnitude. It occurs to me that the conveyer-belt-is-for-product paradigm doesn't even need to be shifted in order to execute the throwing parts robot sports project, and that makes me realize any type of parts flow speedup ideas could be easily intertwined into the 4 dimensional flow. For instance, they could put safety cables on safely attached parts, and then stream them down into their places with a semi-throw semi-catch system (the cables would be an annoyance, but would reduce potential harm to humans as long as all mistakes end up away from humans in catchments away from their exposure). Or, they could have 100% walls between human areas and airborne parts that are sufficient to stop any type of penetration. Airborne part could be a naked part flowing with aerodynamics, and/or specifically "drone" carried. For instance.

I wonder if they intend simply to 1-dimensionally speed up the parts robots they currently have. I consider that very in-box thinking. I'm not opposed, as long as they can do it safely, but seriously think they're going to try to at least 3 dimensionalize it to a degree, since the 1 dimensional approach seems foolish and super-limited. I prefer the 4 dimension approach, however.

 

[phil0909]

IIRC, in the cc before this, Elon was himself programming robots in the battery line because contractor had dropped the ball, but everything was just about ready.

Yes, Elon bragged that he had completed 40 man-years of programming work in 2 weeks, or some such. The only other man I know of capable of such a feat is Kim Jong-Un.

 

[phil0909]

In the conf call yesterday, the term "product" was used at least once to describe building factories. Before I just took this to mean they are trying to look at the factory as a product, in order to make it work really well. But for some reason yesterday I got to wondering if they are thinking of licensing the factories themselves. It sounds kind of crazy, but if you think back to Tesla's mission "to accelerate sustainable transport" and you think about how Tesla wants competition, not a monopoly, it kind of seems like it might make sense.

No, it's all just gibberish, doesn't mean a damn thing. I wouldn't worry about it if I were you.

 

[phil0909]

TSLA should just invite Investors to GF and have them look at operations, so community knows EM is not BSing.

Yes, absolutely. Tesla should totally do this!

Unless...

 

[RedMS]

I’m poor so I picked up a couple at 297 as well. I think it’s a good buy, especially in the long run

 

[DaveT]

Was surprised to see sub-$300 today. Decided to pick up some LEAPs.

Some thoughts on TSLA volatility over past 2 days,
Why is TSLA getting beat up so much more than the rest of the market?

 

[Off Shore]

I was on the bus from Managua for the best part of the clearance sale today, so I had to settle for $309. But it was less than my previous buy-on-dip, so I lowered my cost basis Rage on, Elon!

 

[GoTslaGo]

My bottom feeding buy limit in the 295s triggered today.

Interestingly it was filled in three separate orders. It implies to me that it was hard for my broker to fill that order in a single go around... which implies to me that getting shares at those prices maybe getting to the bottom.

That being said, I've got another set of bottom feeding buy limit orders on the ready...this one I won't post in deference to those on margin.

 

[mongo]

Yes, Elon bragged that he had completed 40 man-years of programming work in 2 weeks, or some such. The only other man I know of capable of such a feat is Kim Jong-Un.

Where/ when did Elon claim he personally did the SW rewrite?

From Q3 2017 transcript available here:

We had to rewrite all of the software from scratch and redo many of the mechanical and electrical elements of zone 2 of module production. We've managed to rewrite what was about 20 to 30 man years of software in 4 weeks, but there's still a long way to go.

It's -- because the software work can be fast with the electromechanical elements that need to be fabricated and installed, and that -- getting these items in place and rebuilt is, unfortunately, a lot longer, and that's -- it's probably more external constraints than software. This is what I spent many a late night on the Gigafactory working on.

J.B. has been here constantly, and we've -- we allocated many of our best engineers to fundamentally fixing zone 2 of the module line and then most problem behind that it is zone 1.

 

[mongo]

Ha! I've been saying I've been wondering if the thing would work better if parts should be thrown by robots to other robots in the factory (even dropped from above), and cars should be in assembly zones, rarely moved, with parts flowing to them to robots that move (slightly from car to car) in and out to each car as the multifunction robots are needed for the various parts. I've always thought it was crazy a multiton product was the item on the conveyer belt rather than its components. Elon did talk about airflow for parts transport. But there are some rather cubic buildings for inventory on both sides of the Fremont factory. Elon mentions this topic as a constraint, and I believe him, but I think it would be fantastically fun to solve given Elon's thoughts on methods to solve it; if he isn't reigned in to failure by the Board, this topic should come with some variety of possibilities for good efficiency improvement by orders of magnitude. It occurs to me that the conveyer-belt-is-for-product paradigm doesn't even need to be shifted in order to execute the throwing parts robot sports project, and that makes me realize any type of parts flow speedup ideas could be easily intertwined into the 4 dimensional flow. For instance, they could put safety cables on safely attached parts, and then stream them down into their places with a semi-throw semi-catch system (the cables would be an annoyance, but would reduce potential harm to humans as long as all mistakes end up away from humans in catchments away from their exposure). Or, they could have 100% walls between human areas and airborne parts that are sufficient to stop any type of penetration. Airborne part could be a naked part flowing with aerodynamics, and/or specifically "drone" carried. For instance.

I wonder if they intend simply to 1-dimensionally speed up the parts robots they currently have. I consider that very in-box thinking. I'm not opposed, as long as they can do it safely, but seriously think they're going to try to at least 3 dimensionalize it to a degree, since the 1 dimensional approach seems foolish and super-limited. I prefer the 4 dimension approach, however.

A few points to concider:
The vehicle does not have any cabling attachments to the GF, robots do.
A small Kula robot KR 360 2830 has a mass of 2,385 kg, the finished 3 is 1,740 kg or so.
To work without added vision requirements, the robots and vehicle need to have a fixed common reference coordinate system.
Each robot has a specialized end effector for its task.
The vehicles will start final assembly after the paint booth. They need to move from there and eventually get to the exit.

A conveyor system can move more than one type of part on one belt (time or position multiplexing).

Autonomous carts (and belts) can carry large numbers of parts with specialized racks. (Like Kiva bots at Amazon)

 

[Tslynk67]

Don't kid yourself, anyone in the stock market is gambling. Some take higher risks for greater reward. Others take much higher risks for much greater reward. I don't think you can truly say that some are gambling and others are not.

From Wikipedia:
Gambling is the wagering of money or something of value on an event with an uncertain outcome with the primary intent of winning money or material goods. Gambling thus requires three elements be present: consideration, chance and prize.

Sure, all the above accepted, but options are, by definition, time-limited, so one is at the mercy of chance, as we saw in the last week, much more than riding-out the storm.

 

[AlMc]

Was surprised to see sub-$300 today. Decided to pick up some LEAPs.

Some thoughts on TSLA volatility over past 2 days,
Why is TSLA getting beat up so much more than the rest of the market?

Excellent synopsis, as usual, Dave. I am not sure if the piece acknowledges the latest info in the 8K. Their existing equipment/people at the GF is capable of getting to 2.5k/week battery model production by end of Q1 *without* the Grohmann fix. The Grohnman fix is needed to help them get to 5k exiting Q2. This info is materially different than what many people took away from the recent CC.

 

[Krugerrand]

Seriously? When has Elon ever "over delivered" on a production target? Listen to CC, even Elon has his doubts. There will be another excuse when this target is not met and there will be fewer folks willing to prop up this loss making enterprise with New money.

Model S. Originally planned for a max of 20k per year worldwide. Way over delivered on that puppy, eh?