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

[neroden]

Interesting bit.

If 50% of the volume is short selling and the price is stable, then it means that no investors were selling.

True, but the source of the data can double or triple the volume attributed to short-sellers -- it can be used as a solid relative indicator (as it goes up or down), but it is a bad absolute indicator. We went through this in Papafox's thread a year or so back.

 

[Remus]

I remember right after earning call lots of tslaq folks went on and on about why it's impossible for Tesla to offer insurance.

Now that the insurance is getting more real, why the q folks so interested in that?

 

[neroden]

7k/w was Fremont only w/o China.
"Model 3 production volumes in Fremont should gradually continue to grow throughout 2019 and reach a sustained rate of 7,000 units per week by the end of the year"

But given the new Q2 estimates of 90-100k total deliveries, assuming 20k S/X(lines still ramping up) that's 70-80k of 3s, plus adding 10-20k to the pipeline, thats 80-100k produced or assuming 12 weeks in a quarter, 6.7k - 8.3k/week.

This does look somewhat... unrealistic. They need sustained 7k/w right now, not at EOY. Maybe that's the case, but I wish he didn't throw these numbers into the letter. Why put pressure on yourself.

Yah. I think there are 13 weeks in the quarter now (no major retooling or major holidays). But even if they do no inventory add, get S/X up to 20K, and have 13 weeks in the quarter, the low end estimate requires 5385/week and the high end requires 6153/week. Adding 10K of inventory build to the high end estimate requires 6923/week.

So yeah, they really need to be at 7000/week NOW. I suspect they aren't, though I'd love to be proven wrong. If they're at 5500/week, they can hit their low-end estimates but they're still struggling. If they're at 7000/week, they're good. If they're below 5500/week, they have real problems.

 

[Remus]

The nearest alternative route adds 15 minutes.

Roads are a wild and wooly space. For budgetary reasons, I don't see them being standardized any time soon.

For the specific case you mentioned, customer delayed by 15min, so what? Pretty hard to make a negative headline out of it.

 

[neroden]

OT

2015 is ancient. About time you got a new one.

I'm hanging onto my 2013 until it goes to a museum. (Actually, I may actually donate it to a museum if I can find a suitable one. I may have the last one which wasn't upgraded to v9 software, which makes it a valuable museum piece. Interested museums which will agree to not update the software, and maintain it as an example of Tesla's old software, may contact me. )

 

[M3Rider]

We don't know how much automation is required to hit 7000 Model3 per week, and how much automation Tesla has implemented. It could be that 5000/week is the limit with human labor, and automation is needed for further production jumps. We have already seen that automation is very tricky.

On the positive note, we have to consider that 90-100k deliveries estimate was given almost a month into the quarter. So, I really hope this is based on production volume they already have seen and extrapolated, plus it assumes that cell issues on the Panasonic side were already resolved a while ago.
So, hoping this estimate can be more realistic than "tiny profit" estimate in Q1 vs a big loss.

My issue though is that I could somewhat believe 7k/w, but the higher end of that estimate that translates to 8k+/w does not make sense, Elon haven't talked about such numbers for a long time, since some 10k by EOY 2018 prediction.
Maybe I'm counting wrong and they only plan to put 10k of 3s into pipeline and if S/X deliveries are higher including prior inventory sold, maybe 30k, but

we are in the process of increasing production backup over the course of Q2. So, just for the pervasive expectations, we will exit Q2 at a higher production rate than we did in Q1 on S and X and then return back to a more normal volume in Q3

Then 60-70k 3s delivered + 10k in-transit = 70-80k or weekly (assuming down time for maintenance, so 12 weeks) = 5.8k - 6.7k/w which is more realistic. Maybe that's the answer - more S/X, less in pipeline.

Actually at the end of Q1, they had a lot of international inventory that they shipped but could not deliver. So, the 20k includes both domestic (2 weeks worth, so about 5 to 7k) and international. Ofcourse you keep selling inventory and replenishing the inventory - to form the "pipeline".

What I was referring to was - they have more in S&X inventory than they probably want to keep. So, they will reduce that inventory and that money can be used to build 3 inventory. So, on the whole this would be cash neutral.

International inventory they had at Q1 end is delivered. They will deliver what they ship in April-May too. Only the cars they put on ships in June will become the part of the pipeline that hurts the bottom line. So, you can't say right now what that will be. If we assume ~ 60-70k 3s delivered plus some extra in transit =70-90k total. We don't know international %. Out of 50k delivered in Q1, U.S. was 24k, so roughly 50%. If the % trend continues, then 35-45k will be shipped internationally in Q2. So, approximately a third of this stuck on ships is ~12-15k. Don't know if U.S. inventory of 3s would go lower or not.

 

[Navin]

This Adam Jonas guy at Morgan Stanley makes me angry. He basically kicked Tesla today for no good reason other than to get on tv and push his bear / short narrative. Im pulling my account with Morgan Stanley and encourage other folks who own Tesla’s and / or shares who park money with Morgan Stanley to do the same.

 

[Steve m]

Yes, production is hard, you are preaching to the choir. I agree wholeheartedly. Go tour Marysville and see how Honda puts together the Accord (even not taking account the ICE motor), its impressive.

Not going to beat a dead horse, IMO, the changes announced to S/X should not be a reason for a long delay in manufacturing. There were more substantial reasons than a line changeover for the S/X dropoff in sales. Was the cutoff to the change done prematurely? Who knows.

Ah would guess that if you are going to introduce a significant upgrade that you would wind down your manufacturing of the existing version and try to sell as much of your existing inventory...

(I use extra dots for dramatic effect)

Before anouncing your improved product....

Where you ramp production up again.

 

[neroden]

So how did YOU figure out that the road wasn’t actually closed?

(1) People were driving on it.
(2) People were parked at the roadside.
(3) The construction equipment was all parked at the roadside.
(4) No construction workers were visible.

With enough data of examples of this, labellers can tag clues that indicate the road isn’t actually closed. AI can be trained to recognize Road Closed signs that are not applicable.

Very, very tricky.

Again, nobody else has the capability to gather enough data to do this except Tesla.

True.

Nobody else has a chance. Tesla has a very hard project ahead of it, but nobody else is anywhere.

 

[Prunesquallor]

Yah. I think there are 13 weeks in the quarter now (no major retooling or major holidays). But even if they do no inventory add, get S/X up to 20K, and have 13 weeks in the quarter, the low end estimate requires 5385/week and the high end requires 6153/week. Adding 10K of inventory build to the high end estimate requires 6923/week.

So yeah, they really need to be at 7000/week NOW. I suspect they aren't, though I'd love to be proven wrong. If they're at 5500/week, they can hit their low-end estimates but they're still struggling. If they're at 7000/week, they're good. If they're below 5500/week, they have real problems.

The InsideEVs number is typically US only. Do they try to extrapolate to total?

 

[neroden]

This Adam Jonas guy at Morgan Stanley makes me angry. He basically kicked Tesla today for no good reason other than to get on tv and push his bear / short narrative. Im pulling my account with Morgan Stanley and encourage other folks who own Tesla’s and / or shares who park money with Morgan Stanley to do the same.

MS are crooks anyway... good on you for discontinuing business with them.

 

[neroden]

The InsideEVs number is typically US only. Do they try to extrapolate to total?

No, they don't. We have to do a guess on international.

 

[anthonyj]

If delivery #'s are good for April I'm sure Tesla/Elon will put out a statement saying so

 

[AudubonB]

So...um...my persona comes across as a bald, grumpy old man waiting for the fisticuffs to commence, huh?

Okay, I can live with that.

Oh.
I thought the comment was referring to those falcon-wing ears.

 

[AudubonB]

Merrci biene, monsour, und zo witer. Believe me, my pronunciation is trop zolie, even worse than spelling....

Sawadi krap.

 

[Steve m]

(1) People were driving on it.
(2) People were parked at the roadside.
(3) The construction equipment was all parked at the roadside.
(4) No construction workers were visible.


Very, very tricky.


True.

Nobody else has a chance. Tesla has a very hard project ahead of it, but nobody else is anywhere.

The way I deal with this type of problem is...

Hey my road is closed. So I pull over and think what now? If a couple of other cars come along and ignore the road closed signs then I would follow them.

Now, what if all there are, are AI drivers...

At some point, the state highway department is going to start collaborating with the car companies and the AI will know ahead of time the road isn't really closed. Oh and they will repaint the roads where cars get confused.

 

[Beaver]

I know inventory car numbers don't mean much in terms of overall sales but one I found interesting is 90210. I checked that postal code for fun last Friday and for the Model 3 inventory was listed as "99+". It is now in the teens. Maybe the 3 is the hot graduation gift of 2019.

In Chicago we went from 40 to 10 in the same time period. What I found interesting, and good, was that we had a decent amount of long range AWD and there is only one left.

 

[Prunesquallor]

No, they don't. We have to do a guess on international.

So what would be a “good” InsideEV April number?

 

[DocZ]

Not sure why people are focusing so much on InsideEV Numbers. They are not accurate.

 

[EVNow]

International inventory they had at Q1 end is delivered. They will deliver what they ship in April-May too. Only the cars they put on ships in June will become the part of the pipeline that hurts the bottom line. So, you can't say right now what that will be. If we assume ~ 60-70k 3s delivered plus some extra in transit =70-90k total. We don't know international %. Out of 50k delivered in Q1, U.S. was 24k, so roughly 50%. If the % trend continues, then 35-45k will be shipped internationally in Q2. So, approximately a third of this stuck on ships is ~12-15k. Don't know if U.S. inventory of 3s would go lower or not.

I'm basically trying to figure out production, delivery and inventory levels - now that Tesla is saying they will end the wave. The basic assumption is they want about 1 month sales in inventory.

BTW, bigger pipeline doesn't hurt the bottomline (except for interest) - but it hurts the cash position.

Anyway, this is my somewhat optimistic estimate.