General Discussion: 2018 Investor Roundtable

[abasile]

While it'd be great to be proven wrong, I'm skeptical that Model 3 reservations will meaningfully increase going forward, and I'm not sure that the number of M3 reservations is going to continue to be a useful metric for evaluating demand. The number of outstanding orders should be more relevant, if Tesla choses to share this.

At this point, at least in the US and Canada where many LR/PUP cars have already been delivered, there seems to be less advantage to the consumer in having held a reservation. Today, anyone can initiate a new reservation and then immediately order a car with LR, PUP, and possibly AWD.

When the cheaper variants of the M3 become available, having an existing reservation might afford some priority in being invited to order and perhaps being able to claim a reduced US tax credit while it's still available. I do hope that Tesla will try to give priority to its early reservers when it comes time to sell the $35k car. Perhaps a new reservation today could give a budget-minded buyer a better shot at a $3750 or $1875 US tax credit. Personally, I would recommend that anyone interested in a Model 3 initiate a reservation now. On the other hand, $1000 is usually not a trivial amount for a mainstream consumer to park for months, and I think most people will simply wait for the car to be available to order (or drive off the lot).

 

[RubberToe]

PG&E replacing 3 gas power plants with 2,270 MWh battery storage...

PG&E to replace 3 gas plants with world's biggest battery projects

 

[Unpilot]

Wouldn't it be nice if as everyone is nitpicking about 5k a week, TE is signing a lot of these mega project's

 

[mongo]

Not in real real world. Tesla loaded up to complete final inspection on 5000 cars last week. This takes production from the week before and the week following the burst week. How much maintenance was done during burst week? How many S/X were painted?

Musk said they would exit 2017 making 5000/week and produce a half a million vehicles in 2018. At that time I said they would produce about 250,000 vehicles in 2017, that Alien Dreadnaught was largely b.s. Using Musk as a source of predicting production is repeatedly proven to be faulty.

A good guess at Tesla's real run rate today is 3000 - 3500. They will improve as fast as they can. They can't know how quickly they can improve.

To quote Scotty:
you canno change the laws of physics.
Cars bodies take space, they can't leave 100's of S/X sitting around post paint. Or have thousands of 3 sitting around to game the numbers,

 

[aubreymcfato]

New article by Alex Roy, from The Drive:
How the Tesla Effect Is Hurting Tesla
I'm not particularly fond of him, but I think articles like these are very important for that particular audience that like cars but are annoyed by Musk celebrity and Tesla hype, bears VS bulls, etc.
Tesla makes better cars, and makes for a better car ecosystem: this should be a main talking point everytime.

 

[winfield100]

PG&E replacing 3 gas power plants with 2,270 MWh battery storage...

PG&E to replace 3 gas plants with world's biggest battery projects

".....because the utility would own the 182.5 MW, 730 MWh project it is building with Tesla......"

 

[Waiting4M3]

PG&E replacing 3 gas power plants with 2,270 MWh battery storage...

PG&E to replace 3 gas plants with world's biggest battery projects

Those are some large numbers...

 

[neroden]

New article by Alex Roy, from The Drive:
How the Tesla Effect Is Hurting Tesla
I'm not particularly fond of him, but I think articles like these are very important for that particular audience that like cars but are annoyed by Musk celebrity and Tesla hype, bears VS bulls, etc.
Tesla makes better cars, and makes for a better car ecosystem: this should be a main talking point everytime.

Best article I've seen in a while. No analysis of the financial situation, of course -- he doesn't do that -- but he's got the fundamentals right on tons of details.

 

[winfield100]

Those are some large numbers...

"....The project would be three times the size of the current world record: Tesla's 100 MW, 129 MWh battery in Australia...."
(is Puerto Rico next? there are the hints about 11,000 projects there)
Tesla has ‘about 11,000’ energy storage projects underway in Puerto Rico, says Elon Musk

 

[Foghat]

Those are some large numbers...

3 gas plants replaced. The significance of this trend can not be overstated. I think this steps on some fairly big toes since gas has been heavily invested in as the “future.” One can begin to see how these interests would try to mitigate risk of this accelerating given the real possibility of stranded assets.

 

[electracity]

To quote Scotty:
you canno change the laws of physics.
Cars bodies take space, they can't leave 100's of S/X sitting around post paint. Or have thousands of 3 sitting around to game the numbers,

It's not that simple, but they certainly have space to let a couple of hundred car bodies sit around. The paint shop will far more space compared to today to do 500K cars a year. Cars spend maybe 18 hours in the paint shop.

The "tent" is likely about future space utilization. They didn't want to build a GA line and then need to move it in a year.

Production is often skewed to achieve an accounting result. Everyone who has worked professionally in a large auto plant or similar manufacturer knows this. There is zero percent chance that Musk didn't game the burst week. I would in his situation, and my base level of truthiness is considerably higher than Musk's.

Why do you think Musk mentioned "the last car passing the final QC check". He was defining what he meant by 5000/wk. This allows his claim to be indisputably true but not a complete picture of production. A real run rate of 5000/wk is proven by doing it week after week.

What is also true is that Tesla is pushing as many expenses out of Q3 as possible. Q2 financials are a one time disaster, so deeper losses probably don't matter. Q4 financials will have increased M3 sales compared to Q3, so profitability is more certain. Again, I'm not aware of any public company that doesn't manipulate financials to achieve goals. Often the reason is to smooth results between quarters and years.

 

[mongo]

It's not that simple, but they certainly have space to let a couple of hundred car bodies sit around. The paint shop will far more space compared to today to do 500K cars a year. Cars spend maybe 18 hours in the paint shop.

If they are in the paint shop line, then they are not being pre or post buffered. 500 Model S require over 1.5 miles to store if placed bumper to bumper. It would also require 500 extra frame holders, unless they just placed the painted frames on the ground or purpose built racks. Purely for reference, BMWs paint line is 4 miles and takes 12 hours.

The "tent" is likely about future space utilization. They didn't want to build a GA line and then need to move it in a year.

Production is often skewed to achieve an accounting result. Everyone who has worked professionally in a large auto plant or similar manufacturer knows this. There is zero percent chance that Musk didn't game the burst week. I would in his situation, and my base level of truthiness is considerably higher than Musk's.

Why do you think Musk mentioned "the last car passing the final QC check". He was defining what he meant by 5000/wk. This allows his claim to be indisputably true but not a complete picture of production. A real run rate of 5000/wk is proven by doing it week after week.

I do not think gaming was required nor implemented.
The count needs a reference point, Final QC check is a known point in the process that is representative of production (as opposed to vehicles shipped which is impacted by outside forces). One could argue that they started with a fully filled line and ended with an empty one, but that would need data to back it up. It would also only impact the total rate number by the line buffer size (and is a really hard scenario to create).

What is also true is that Tesla is pushing as many expenses out of Q3 as possible. Q2 financials are a one time disaster, so deeper losses probably don't matter. Q4 financials will have increased M3 sales compared to Q3, so profitability is more certain. Again, I'm not aware of any public company that doesn't manipulate financials to achieve goals. Often the reason is to smooth results between quarters and years.

Sure, If they can get the spend in Q2 and make Q3 more positive, why not.

 

[Zaxxon]

I did some quick guesstimation on the Model 3 ramp given the updated parameters (2,645 produced in 2017, 1k/week entering Q1 and ending Q1 at 2,500/week, and 2,500/week entering Q2 ending at 5,000/week) and assuming a fairly linear ramp tilted slightly to EoQ to remain conservative. I then also ran the same guesstimations through the end of the year assuming a similarly linear ramp to 10k/week at the end of December, minus the final week of the year for possible factory closure.

It's in a Google Sheet here for anyone interested.

Short summary:
Since these are effectively Tesla's official guidance at this point and their modus operandi is not to exceed official guidance significantly...
-I expect the Model 3 production by quarter to be around 22k in Q1, 49k in Q2, 83k in Q3, and 107k in Q4.
-I expect total Model 3 produced count by EoY not to exceed 265k.
-I'll hereby stick to my somewhat facetious 1/1/2018 guess posted here and say that we'll see 201,800 Model 3s produced in 2018. And my 'that'd sure be great' hope is that the 265k comes to pass.
-Any hopes folks had of Tesla approaching 500k vehicles produced in 2018 died today. If we assume reasonable growth of S/X ending 2018 at 125k between them, that's 390k total for the company when you add in 265k Model 3s. I think 400k would be extremely optimistic.

And to be clear before I get jumped on, I've thought all along that we'd be in at best the 350k-400k range for 2018 given conversion from Elon Time™ to Gregorian Time. So today's ramp change announcement doesn't surprise or particularly faze me. It's basically what I expected to see. I'm happy that 2017 Model 3 production exceeded 2,600 as I was guessing 2,000-2,500 after the Q3 earnings call. And 101k S/X is great.

Now back to waiting for my configuration invite...

Boy, did this (even my somewhat facetious guess) prove to be optimistic... Since that post on 1/3, I've re-run the numbers each time we've gotten an update from Tesla--in late Jan when the 10k EoY goal was shifted and then most notably on 4/3 after Q1 production was announced (revised my 2018 3 production estimate from 265k to 187k) and now on 7/3 after Q2 production was announced (revised down from 187k to 173k). At least I got my car in May! That does make this sort of analysis a lot less painful.

Since there has been some discussion over maintaining end-of-quarter production rates:
-Tesla entered Q1 at 1,000 3s/week. Q1 average production was 750/week, or 75% of Q4 exit rate.
-Tesla entered Q2 at 2,000 3s/week. Q2 average production was 2,200/week, or 110% of Q1 exit rate.
-Taking the midpoint of those two numbers--92.5%--and the 5,000/week Q2 exit rate and a[n admittedly conservative] 6,000/week Q3 exit rate, I predict 60,100 3s produced in Q3 (4,600/week average). The 100,000th Model 3 would be produced in the last week of Q3.
-Taking a 6k/week entry rate to Q4 and 92.5% maintenance, and another conservative assumption of a 6,750/week exit rate, I have Q4 production at 72,150 (5,550/week average).
-This results in a 2018 Model 3 number of 173k and change. Add 100k S/X to that (currently 44,100 delivered, plus significant overhang in transit) and we're at about 275,000 vehicles. 2017 was 103k, so that's a 167% increase YoY. I'll still take that every day of the week and twice on Sunday.

I realize that this will likely come off as far too conservative to many here. As a response I'd simply point out that the last two times I posted my guesstimates, I was also called too conservative, and yet those guesstimates turned out to be optimistic. I do think that Q4 especially could turn out to be too conservative, but given my need to revise downward twice so far, I am shooting to be on target or pleasantly surprised this time.

 

[stef]

Boy, did this (even my somewhat facetious guess) prove to be optimistic... Since that post on 1/3, I've re-run the numbers each time we've gotten an update from Tesla--in late Jan when the 10k EoY goal was shifted and then most notably on 4/3 after Q1 production was announced (revised my 2018 3 production estimate from 265k to 187k) and now on 7/3 after Q2 production was announced (revised down from 187k to 173k). At least I got my car in May! That does make this sort of analysis a lot less painful.

Since there has been some discussion over maintaining end-of-quarter production rates:
-Tesla entered Q1 at 1,000 3s/week. Q1 average production was 750/week, or 75% of Q4 exit rate.
-Tesla entered Q2 at 2,000 3s/week. Q2 average production was 2,200/week, or 110% of Q1 exit rate.
-Taking the midpoint of those two numbers--92.5%--and the 5,000/week Q2 exit rate and a[n admittedly conservative] 6,000/week Q3 exit rate, I predict 60,100 3s produced in Q3 (4,600/week average). The 100,000th Model 3 would be produced in the last week of Q3.
-Taking a 6k/week entry rate to Q4 and 92.5% maintenance, and another conservative assumption of a 6,750/week exit rate, I have Q4 production at 72,150 (5,550/week average).
-This results in a 2018 Model 3 number of 173k and change. Add 100k S/X to that (currently 44,100 delivered, plus significant overhang in transit) and we're at about 275,000 vehicles. 2017 was 103k, so that's a 167% increase YoY. I'll still take that every day of the week and twice on Sunday.

I realize that this will likely come off as far too conservative to many here. As a response I'd simply point out that the last two times I posted my guesstimates, I was also called too conservative, and yet those guesstimates turned out to be optimistic. I do think that Q4 especially could turn out to be too conservative, but given my need to revise downward twice so far, I am shooting to be on target or pleasantly surprised this time.

Which means a year exit rate just shy of 400k. What did Elon say? Tesla is doomed... lol

 

[sub]

Hey Elon, about that short burn thing....

 

[Zaxxon]

Which means a year exit rate just shy of 400k. What did Elon say? Tesla is doomed... lol

Exactly. I like to run these estimates to help hone my 'Tesla Time' translator, but also to show myself that even when you take everything with a large grain of salt, the accomplishments are still pretty incredible. 275,000 all-electric vehicles in 2018? Who else is doing more than 10-15% of that? GM just reported 3,483 Bolt sales in the US in Q2. Leaf is at 4,114. Take them both together, and the 3 already more than doubles their total, with a far higher selling price.

 

[stef]

WallStreet still mad a Elon and in full revenge mode I guess...
Well hope he’s feeling good knowing that he may not need them anytime soon.
All will be well in the end...

 

[electracity]

If they are in the paint shop line, then they are not being pre or post buffered. 500 Model S require over 1.5 miles to store if placed bumper to bumper. It would also require 500 extra frame holders, unless they just placed the painted frames on the ground or purpose built racks. Purely for reference, BMWs paint line is 4 miles and takes 12 hours.


I do not think gaming was required nor implemented.
The count needs a reference point, Final QC check is a known point in the process that is representative of production (as opposed to vehicles shipped which is impacted by outside forces). One could argue that they started with a fully filled line and ended with an empty one, but that would need data to back it up. It would also only impact the total rate number by the line buffer size (and is a really hard scenario to create).



Sure, If they can get the spend in Q2 and make Q3 more positive, why not.

Shouting CEO, changing rules: inside Tesla's Model 3-building sprint

 

[Yuri_G]

Shouting CEO, changing rules: inside Tesla's Model 3-building sprint

Hiring 400 people a week should help take the edge off. It's obvious the q2 goal was important to Elon. Now he'll make it sustainable and the media will have to find something else to report on daily.

 

[mongo]

Shouting CEO, changing rules: inside Tesla's Model 3-building sprint

Whether Tesla can do it week in and week out - and without relying on overtime and extra hands - is another question, and one that weighed on investors Monday, as shares slumped 2.3 percent.

Which is why they announced hiring 400 people a week...

Because of the focus on the Model 3, the S line is about 800 cars behind, the worker said.

Asked about the potential S and X impact, Tesla said it also produced 1,913 of those vehicles during the last week of the quarter along with its Model 3s.

So Tesla was planning on making 2,713 S that week? Even if you allow for that it should have been S&X, they never had a 32k quarter (2,713 * 12 weeks). Highest to date was 25,708 (2,142/wk * 12 wk).

My take away: the factory/ line can produce 5k 3's and almost 2k S/X in a week, and the ramp outran the staffing. That's better than, "we had people, but couldn't get the cars through automated body line"