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

[Fact Checking]

Tesla guided to around 10k Model 3's being in transit for Q1 right?

Almost: they guided for Model 3 production to outpace deliveries by about 10,000 units, due to filling up the international logistics pipeline.

This would translate to ~11k Model 3's in transit in Q1, but only if Tesla's "fleet" inventory stays constant, which it might not: in Q4 the fleet increased to about 7,000 units, 5,000 of which were probably common configurations for the U.S. market.

I'm still trying to get my head wrapped around the tiny profit for Q1.

That's still probable: profits depend on units fully delivered by March 31, and European deliveries have only just begun.

The depth of the reduction in Q1 Model S/X deliveries is also unclear to me:

• There's a seasonal dip of ~3k units,
• plus a new production constraint of about 20k units per quarter (due to the removal of the 75 kWh pack and the standardization on the 100kWh battery packs),
• plus the removal of the S/X production night shift at Fremont,
• and a leak of unknown reliability suggests a 50% reduction in S/X parts supplies.

This puts Q1 S/X deliveries into a very broad range: 13k-22k deliveries are all possible in principle. S/X deliveries below 20k would obviously be taken badly by the market.

 

[mongo]

Now, satellites use SAR (Synthetic Aperture Radar), because on the downside, they're very small relative to the distances that they have to measure from, but on the upside, they're moving extremely fast (SAR creates a large virtual aperture not by having a large spacing between multiple antennas, but rather relying on the fact that by the time the signal returns, the satellite has moved, so a single transmitting antenna can act like second receiving antenna at a significant distance away. For vehicular uses, you'd use two physically separated phased arrays. But the physics is the same.

That's not how SAR works. If the satellite were able to receive its pulse from the original location and the new one, then that would be long baseline interferometry. However, since it only receives the pulse in one location (unless it has multiple antennas), it acts like a single antenna system.

In SAR, the range dimension is done like any chirped radar. The pulse is a linearly increasing frequency, the return is mixed with the the same linear sweep (time shifted), that causes distance (time) to be transformed into frequency. An FFT on that gives you return strength vs distance. Cool thing: distance to the target only impacts received power, not resolution, the frequency range and slope determines that. FWIW in LEO, a satellite radar pointed at the earth moves about 100 m in the time it takes to receive a transmitted pulse. This means the antenna beam width needs to be wide enough to pick up what is sent, not a problem when you are 2000km from the target (6 degrees about does it). To boost data quantities, multiple pulse can be in flight at the same time (~130 for a 10kHz pulse repetition frequency).

Now to azimuth: in a convention radar, the larger the dish, the tighter the beam and the better the resolution. In a SAR, it uses data collected while moving instead. The satellite does not have a tight antenna beam (and any beam spreads with distance), that means all objects the same distance from the antenna show up as the the same data point (ie if you were 100m away from two telephone poles and located in their midpoint, they would only present what looks like a single return. To overcome this, the radar system sends out multiple pulses as it travels the sky. As it moves the relative distance to every point changes (i.e. as you drive past telephone poles, the ones in front get closer, and the ones behind get further away). So each object becomes a curve of points.

The processing unit takes all of the compressed range pulses, then shifts them based on the imager's movement (motion compensation) and range to target, finally it runs another FFT on the data which compresses it into specific points. It does this once for each range line in the image, compressing data from 1/2 the resolution based swath before to 1/2 after. So you end up not getting image data at the ends of your full data set since the needed data window extends beyond it.
Cool thing: effective resolution is not dependent on distance to target either, just on your effective antenna width so this works great for satellites far from the imaged object.

(Wikipedia's intro is either misleading or wrong...)

 

[Lessmog]

They'll be missing (10,000 cars * $55k/car) $550 million in revenue, while still needing to account for the COGS for those cars in-transit. I think the lack of profitability is pretty clear.

What's not stated is that we can probably expect something close (lower ASP due to changing sales mix - maybe $500 million?) as profit for q2.

Did you account for the time shift? Unless I missed something crucial (always a safe bet) that loss of revenue is only contingent on the transit still occurring during the period of accounting, ie end of quarter day. I'm sure this has been beaten to death too here, but as long as all the ships have delivered their cargo to customers before that date, revenue is in the bank.

Sure, it will be a problem once deliveries do happen continuously regardless of quarterism and before steady-state is reached. But not until then. (Note caveat above.)

Also, high mix ASP is reset on virgin markets.

 

[MarcusMaximus]

I do not believe that Tesla will be accumulating the memory of recent road conditions.

I think you can't count on road to be visible or for another Tesla to provide you what it saw etc. ahead of time.

The assumption needs to be that it's been snowing all night and you're sending your driverless FSD car to pick up someone in the morning and the roads have not yet been cleared in places, so your car will be first to navigate them.

If it can't do it, then you can't satisfy your customer's taxi request.

We all speculate what will happen. There were some vids posted. My thought about those videos is that 90% success rate is far from 99.999%.
With AP the liability lies with the driver.
If Tesla assumes FSD liability and it has a 90% success rate, then they'll go bankrupt due to the 10%. It needs to be much much better. Pretty much no mistakes.

Somewhat side topic: I find the use of percentages in this discussion funny. 99.999% of what, exactly? Frames? Seconds? Intersections?

I think we have trouble coming up with an actual good measure of performance for this, so we just use some subjectively really high percentage. Note that I’m not actually picking on you, specifically. It’s a common refrain, and even Elon did it on the Q4 ER.

 

[M3Rider]

Indeed - AFAIK, this would be brand new ground, if Tesla were to do it. But it would be extremely useful new ground, and is already in use in non-automotive fields (aka, no "revolutionary breakthroughs" required). A lot easier than a lot of the things people are trying to do make better LIDARs, IMHO.

@Fact Checking

Both of you guys, book a 2024 seat on the BFR going to Mars. Elon will provide 3D printers, so full freedom to terraform and do other "easy" things with practically no supervision

 

[Carl Raymond]

OT

Sorry for the offtopic, but I just wanted to note (since Americans spent so long talking about the Superb Owl in this thread): it's increasingly looking like Iceland is going to troll Eurovision again this year - perhaps not quite to the degree that we did with Silvía Night, but close. The band "Hatari" (Hater) handily won yesterday's semifinals, with the song Hatrið Mun Sigra ("Hatred Will Win"). Their stage performance is in full S&M gear, including a band member in a gimp outfit who gets abused by the lead singer (the band is known for approaching the stage by physically shoving the audience out of the way). While Eurovision songs are generally of the "Tolerate everyone!", "Love is awesome!" and "Can't we all just get along?" variety, the chorus of their song proudly declares "Hatred will win, Europe will collapse" and declares love dead.

IMHO, their music sucks (not my taste), but I'll be cracking up nonstop watching them up on stage in Tel Aviv (assuming that they win the final on 3 March). And besides, most Eurovision music sucks regardless. Oh, and they're already driving our social conservatives into fits, which wins anyone kudos in my book .

OT

It’s ok, hate won’t win. Australia is going to win this year. Kate Miller-Heidke is seriously talented.

This is not the song she will sing, but it’s my favourite. The lyrics will move anybody not made of stone.

Edit: actually loosely related, if we remember that Elon was bullied within an inch of his life.

 

[AlexS]

P&O Ferries Update : General Strike at Port of Zeebrugge, Wed 13/02 06.00 - Thurs 14/2 06.00

whats up in Belgium, why do they have general strike. Isn't it one of the happiest nation on the planet?

 

[avoigt]

Looks like official short interest is down to 24.7 million. I'm really surprised how much covering has happened while keeping the price down.

Indeed. The price action after the ER is odd. It looks almost like a patt situation where bulls and bears are kind of in balance. Hard for me to imagine who does sell other than shorts. There is not a lot to see on the technical part too other than $300 is the new floor for now.

Still with the FUD in decrease (will change), good ER report, deliveries starting in EU and China and an Elon Musk that does not give any surface to attack him I am surprise that we do not see any uptrend building.

Some say the shares Baillie Gifford sold has been used by shorts to cover which create the situation we are in but for me that does not add up.

Others might argue that the uncertainty about Q2 results may hinder people to go in maybe in speculation to find some good entry levels but that is not convincing to me as well.

Recession concerns are piling up but there is no better place to invest than Tesla if we would get one. Both Trump as well as May (UK) work hard to create a recession like situation with a ton of political uncertainty in front of us and uncertainty is poison for markets.

Either way Tesla is in a great position and as often in life to have patience is hard....

 

[doctoxics]

Almost: they guided for Model 3 production to outpace deliveries by about 10,000 units, due to filling up the international logistics pipeline.

The depth of the reduction in Q1 Model S/X deliveries is also unclear to me:

• There's a seasonal dip of ~3k units,
• plus a new production constraint of about 20k units per quarter (due to the removal of the 75 kWh pack and the standardization on the 100kWh battery packs),
• plus the removal of the S/X production night shift at Fremont,
• and a leak of unknown reliability suggests a 50% reduction in S/X parts supplies.

This puts Q1 S/X deliveries into a very broad range: 13k-22k deliveries are all possible in principle. S/X deliveries below 20k would obviously be taken badly by the market.

What information leads to the conclusion that S/X production will decrease from 100K to 80K per annum? My understanding is that increased efficiency of the S/X lines makes up for the cancelled shift. Also, demand remains high for both S and X.

 

[KarenRei]

In SAR, the range dimension is done like any chirped radar.

We're not talking about ranging, we're talking about return intensity.

FWIW in LEO, a satellite radar pointed at the earth moves about 100 m in the time it takes to receive a transmitted pulse. This means the antenna beam width needs to be wide enough to pick up what is sent, not a problem when you are 2000km from the target (6 degrees about does it).... (snipping long technical explanation)

I'm not sure what you're disagreeing with here. I merely pointed out that the virtual aperture is relative to the distance that the satellite moves during the time between transmission and reception of a photon. If you want to add a technical explanation on top of that, that's fine, but it's not a disagreement.

The key issue is that your beamwidth is the beamwidth factor times the wavelength over the aperture width. Whether you're dealing with a single antenna, a phased array, multiple correlated independent antennas, or SAR.

Cool thing: effective resolution is not dependent on distance to target either, just on your effective antenna width so this works great for satellites far from the imaged object.

Misleading. Bearing resolution is an angular resolution (degrees, radians, etc). So yes, said angular resolution is constant, but the further away you are, the larger of a physical size that corresponds with. You give the impression that you'd get the same map resolution from a satellite orbiting 100km over the moon as you would a satellite moving at the same 5000km from the moon. This simply is not the case.

 

[NicoV]

P&O Ferries Update : General Strike at Port of Zeebrugge, Wed 13/02 06.00 - Thurs 14/2 06.00

whats up in Belgium, why do they have general strike. Isn't it one of the happiest nation on the planet?

It’s just the unions that want to show they still exist. We got used to it and work from home in order not te be bothered by the strike. This workaround works for a lot of people.

 

[Lucky_Man]

Irresponsible. Gutsy, if it was only the M3 on the road but, IMO this driver put several lives in danger. Best I can tell is the AP managed as well as it did by the contrast between the black pavement and the white snow edge. Yet is still failed several times. Also, I've seen many cars [right in front of me] drift into the slush at the edges and get thrown into a nasty spin-out. No thanks. I'll probably use AP only on interstates, and only in clear road conditions.

 

[MP3Mike]

What information leads to the conclusion that S/X production will decrease from 100K to 80K per annum? My understanding is that increased efficiency of the S/X lines makes up for the cancelled shift. Also, demand remains high for both S and X.

Because they are no longer making the smaller battery packs, and to the best of our knowledge the supply of cells haven't increased. So they can't make as many.

 

[Fact Checking]

Did you account for the time shift? Unless I missed something crucial (always a safe bet) that loss of revenue is only contingent on the transit still occurring during the period of accounting, ie end of quarter day. I'm sure this has been beaten to death too here, but as long as all the ships have delivered their cargo to customers before that date, revenue is in the bank.

That was a nice theory I found probable too, but in the Q4 update letter Tesla guided for Q1 Model 3 production to outpace deliveries by approximately 10,000 units - and revenue depends on completed deliveries.

Since in Q4 there were ~1,800 more deliveries than production, 11,800 deliveries will be "missing" in Q1, compared to Q4.

Hence the lowered Q1 revenue, profit and cash flow.

 

[KarenRei]

What information leads to the conclusion that S/X production will decrease from 100K to 80K per annum? My understanding is that increased efficiency of the S/X lines makes up for the cancelled shift. Also, demand remains high for both S and X.

They'll only be making 100K/year if someone gives them a new source of 18650s Remember, they ditched 75Ds, so now all models are 100kWh. That takes more cells.

 

[AndyH]

I do not believe that Tesla will be accumulating the memory of recent road conditions.

I think you can't count on road to be visible or for another Tesla to provide you what it saw etc. ahead of time.

The assumption needs to be that it's been snowing all night and you're sending your driverless FSD car to pick up someone in the morning and the roads have not yet been cleared in places, so your car will be first to navigate them.

If it can't do it, then you can't satisfy your customer's taxi request.

We all speculate what will happen. There were some vids posted. My thought about those videos is that 90% success rate is far from 99.999%.
With AP the liability lies with the driver.
If Tesla assumes FSD liability and it has a 90% success rate, then they'll go bankrupt due to the 10%. It needs to be much much better. Pretty much no mistakes.

It won't share everything but if anything on the route deviates from what's been observed before it will share that with all cars. All current Tesla's are already doing this. If it's been snowing all night you probably need a monster truck.
But as you say it has to be perfect and could take some time.

 

[defc0n]

Has Moody’s still not upgraded Tesla? Any theories on when that should finally happen?

Seems a bit ridiculous at this point. Maybe they are watching to see how the next product rollouts are funded still even though it’s widely accepted the March convertibles are no longer an issue.

 

[tinm]

What in the flying F is this bozo driving with Autopilot activated on non-freeway roads and, worse, in those snowy conditions!? Holy crap. Why doesn't Autopilot take a glimpse at its GPS location, realize it's on a road with intersections and traffic lights, and tell the driver, "No, you can't use Autopilot on this road"? Tesla drivers keep cheating and using this technology where it's not supposed to be used.

 

[Fact Checking]

What information leads to the conclusion that S/X production will decrease from 100K to 80K per annum? My understanding is that increased efficiency of the S/X lines makes up for the cancelled shift. Also, demand remains high for both S and X.

80k is just my cautious guess. Here's a background comment with all the details.

The real S/X supply ceiling could be anywhere between 80k-90k/year, but I think unlikely to be beyond 85k/year, with Panasonic as the supplier.

But that's OK: Tesla increased S/X prices and is positioning the S/X farther up the luxury brand spectrum. The Model 3 Performance is taking the pricing space of the 75D, at significantly better margins, and without long term supply constraints.

So overall cash generated would stay constant, or might even increase.

 

[TradingInvest]

Looks like official short interest is down to 24.7 million. I'm really surprised how much covering has happened while keeping the price down.

Maybe a long fund used short position to hedge their long position, and closed both near year end.