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

[jhm]

The Rocklin SC is located 25mi east of Sacramento and just a few miles from Sierra College. 5 days before the end of Q2 I noticed about 25 new Teslas parked in the Sierra College overflow parking lot across the street from the main entrance. I came to the conclusion there must not have been enough room at the SC so they likely contracted with the college to use the lot to temporarily store cars slated for end of quarter deliveries.

Today, I drove by to check out the overflow lot and they moved it to a larger covered lot closer to the main campus. I counted more than 50 new Teslas there this afternoon waiting for end of Q3 delivery.

Tesla is definitely a bustling growth story.

View attachment 714352View attachment 714353

OT, I happen to be a former Sierra College student. Got my first two years of college there before transferring to UCSC.

 

[DragonWatch]

Forward Observing

Beta FSD vs 10

listened to one UTube guy about Tesla Insurance ~ I might be willing to move from USAA, member over forty years, and accept Tesla.

Remember a half a year or so back; someone charted out self driving software companies. Tesla was in the middle somewhere. Tesla, Friday night, became the leader ~ in case U missed.

As someone did some napkin math, no governmental agency or private agency can dispute the driven truth. Unless the data is skewed, which I doubt!

Who is about to be driving Miss Dais? Xena II

 

[phantasms]

Crazy thought

I loathe conspiracy theories, more than ever now, but I’ve got to wonder…

I’m listening to yesterday’s Tesla Daily while doing the dishes. Tesla is inviting press/media to Shanghai to witness the 300k car rolling off the line on Wednesday the 29th. First I was like yeah sure that make sense. But, does it really? 300k cars, so what?!?

Who’s thinking what I’m thinking?

Spoiler

Small/cheap Tesla introduction

 

[Silent Ludicrosy]

I think Plaid is about to break some more Nurburgring records. Modified this time. These are the same guys that modified the Pikes Peak Mdel S Plaid.

 

[FireMedic]

OT
@Jackl1956
Leilani Maaja Münter race car driver who drives a Tesla

Leilani Munter - Wikipedia

en.wikipedia.org

Weekend OT.

Wow. I have a new favorite athlete and I didn’t even know it. Thanks for sharing this. Her Wiki page is pretty impressive.

 

[22522]

A forward collision warning with 4 car lengths maintained through the whole event on 30 mph city streets produced a score of 75. So maintaining safe distance by applying the brakes when warned seems to be punished.


The algorithm needs to look at cushion maintained especially with FCW, which can be flakey. So it can be a double whammy as seen here.

 

[DragonWatch]

Forward Observing

Disney recently came to mind. Mom and Dad would go to the drive in to watch a movie back in the fifties. Later I got excited over the Mickey Mouse club, and Annette. Thank god, I saved myself for Sandy. Then there was Disneyland. Now there is Disney World, Epcot Park and Disney+.

WtF. Think Tesla, not just a car. Not interested in Knots Berry farm. Went there for a fifth or sixth grade picnic and worked for architect that transformed them into a theme park ~ but, not Disney.

Look to the future, or check out the rearview mirror ~ balance the two.

 

[StealthP3D]

1) Different jurisdictions or nations will likely require testing to be done within their road system. They may even require separate NN fitting and had coding. So if 400M to 800M are required per jurisdiction, you could easily need on order of 6B on some 10 or so different jurisdictions. I doubt that Musk would be construing this as one giant test in which Tesla must demonstrate zero fatalities over the course of 6B miles. Indeed as your own calculation has show at 0.12 deaths per 100M miles (implying 7.2 fatalities), Tesla would have less than 0.1% chance of being about to go 6B miles with 0 fatalities. In practical terms, this is an nigh impossible test, doomed to failure. Maybe it would help to follow the implications setting 7.2 fatalities as your proposed null hypothesis. The test statistic, the actually number of fatalities over 6B miles, is Poisson distributed with mean 7.2 under the null hypothesis. It also has a standard deviation 2.68 = sqrt(7.2). With probability about 95%, the test statistic will be between 3 and 13 fatalities. So I think the test you are really try to set up is that your reject this null hypothesis if there are 2 or fewer deaths or 14 or more. In this case, we are talking about a two-sided test. So it is not really clear why Tesla would ever need to show that its fatality rate is so much lower than 0.12 per 100M miles that they'd be able to reject this 0.12 rate. If Tesla merely want to demonstrate they are safer than a human driver, it would be better just to estimate the rate and provide a confidence interval. The hypothesis testing framework is not really the best framing for that public communications objective. Rather, hypothesis testing is relevant when getting approval from a governing body; moreover, that body is likely to state what rate to test against. That is, the regulator sets the null hypothesis, while the Tesla needs to supply sufficient information to reject that regulatory hurdle.

2) The weighting and stratification I was writing about was in reference to testing for regulatory approval. Collecting data for training FSD is actually a much more complex task to do well. Certainly weighting and stratification can play a role in training a model, but that was not my point. Indeed, for training you want to be sure that you have broad data on the full scope of driving conditions, but you also want to oversample on certain data where critical and rare events happen. For example, you like will want to oversample on collisions, especially on collisions involving injuries and fatalities. Tesla is even using simulations of critical collisions to be able to augment data and revisit the scenario under varying conditions. Basically, Tesla want to learn as much as possible from each critical event so that FSD will never make mistakes again in such scenarios. So these are the sorts of consideration for curating training data.

Regulators will likely be interested in how Tesla curates training data and can do analysis on how representative the coverage is, the quality of the data, and many other issues. This is data review, but it is not road testing. For road testing, the regulator will likely want data on where and when the miles were driven and much more detailed information on critical events: collisions, injuries, fatalities, etc. They will analyze the data for representativeness and consider any weighting methodology deployed. The test day may well be segmented. For example, Tesla may require certain number of beta testers from each state. State or city segmentation helps assure representativeness. Time of day, day of year, weather conditions are other factors that may call for weighting or segmentation. So the regulator will need to be persuaded that the test exposure miles are sufficiently representative and have adequate coverage. They will also want to analyze the critical event data with special attention to any factors that may reveal weakness or flaw in the driving system.

But after all that work, the regulators are confronted with the final counts of types of critical events. The regulators will want data to show that the true frequencies of certain outcomes are far enough below a critical threshold that random error can be ruled out. This is where hypothesis testing comes in. The regulator may say to Tesla, you need to demonstrate that your fatality rate is below 1.2 per 100M miles. That's the null hypothesis. And Tesla might know that it is likely functioning is at or below 0.12 per 100M. So 0.2 is the alternative hypothesis they want to optimize around. The regulator doesn't care what the alternative hypothesis. But for Tesla it gives them the basis for planning how many test miles of exposure they will want to accumulate before they present there result to the regulators. The power of the test is extremely important to Tesla as they want a high probability that they will be able to submit enough data to pass the test, assuming their alternative hypothesis. But it is the regulator who cares about the significance of the test as they want there to be a low likelihood that they are fooled by statistical error. Type I error is an error that the regulator wants to avoid, while Type II is the error that the regulated entity wants to avoid.

Now suppose that Tesla makes significant advances in training its FSD neural nets, enough to convince them that their fatality rate is at or below 0.06 or bellow. In this situation, Tesla could chose 0.06 as there alternative hypothesis. What happens here is that for the same amount of test exposure, the power of their test goes up, they have a higher chance of passing the regulators test. Or put another way, they could proceed with less data and still have sufficient power. The implication here is that the choice of alternative hypothesis for Tesla drives how much exposure data they need. This is why I put the emphasis on Tesla engineering a better FSD system. The better it truly is, the less exposure data will be required to show that it can reject the null hypotheses posed by the regulator. This means Tesla can get through regulatory testing faster and cheaper. So how does Tesla engineer a safer FSD system, primarily it must do a damn careful job of curating training data. Any lack of vital experience in the training data exposes Tesla to incremental risk in beta testing (and full public release).

One other issue, suppose the regulators will be testing multiple outcomes. Say they require demonstration that the fatality rate is significantly below 1.2 per 100M miles, bicycle collisions are significantly below 10 cases per 100M miles, and pedestrian collisions are significantly below 6 cases per 100M miles. Now the power calculations become much more difficult. You need to have enough miles of exposure to have a very good chance of passing all three tests. So this multi-test situation could push Tesla to do more test miles than the fatality test alone would call for. This also could help explain question 1. Some jurisdictions might require multiple outcomes to be tested, and that could drive up the required sample size. Indeed I looked like the NHTSA was going on a fishing expedition, just looking for any outcome that might be higher than average. If a regulator aggressively pursues any finding any fault, they will likely succeed. And no amount of data would have an adequate chance of passing all the tests. But this is veering off into a hostile political situation, not good statistical or regulatory practice. At any rate, the point here is that, if regulators will be testing many end points, that can drive up the amount of exposure miles needed for regulatory approval. But again, even if that is what the regulators are demanding, the best strategy for Tesla is simply to work on improving the FSD along every conceivable test dimension, and to do that Tesla will need to curate substantial training data on every conceivable misstep a driver could make.

Just to get through regulatory approval, to pass all the stated and unstated tests, Tesla may well need FSD to be 10 times better that human drivers.

Just like to point out that your statistics appear to based on deaths per mile travelled and that due to the energy absorption of Tesla's vehicles with large crumple zones, strong passenger compartments, etc., avoiding deaths is largely a function of following the speed limit and avoiding head-on collisions. Even the few battery fires we have seen in recent years were almost always the result of high-speed driving.

The bottom line is that the deaths/miles travelled should be a fairly easily metric for Tesla to master in the near future. I'm not saying FSD is easy, merely that it's pretty hard to die in a Tesla. From a objective perspective it would seem the number of injury crashes would be a more telling metric than number of deaths.

 

[Singuy]

Crazy thought

I loathe conspiracy theories, more than ever now, but I’ve got to wonder…

I’m listening to yesterday’s Tesla Daily while doing the dishes. Tesla is inviting press/media to Shanghai to witness the 300k car rolling off the line on Wednesday the 29th. First I was like yeah sure that make sense. But, does it really? 300k cars, so what?!?

Who’s thinking what I’m thinking?

Spoiler

Small/cheap Tesla introduction

Or it could just be the press wanting a story about Shanghai giga and Tesla thought 300k car is a fun bonus so a good date to allow them in.

I don't understand all this hope for the cheaper Tesla. Not to mention how they are battery constrained before 4680s are in the hundreds of gwh production levels, but the Chinese are not a fan of having the brand being diluted with cheaper cars. It has always been a status symbol in China to own expensive cars more so than the west. It make more sense to introduce the cheaper car in the US first but will not happen till 2024 is my guess.

If you ever visit China, you'll be surprised that half the cars on the street are 7 series, S classes, and A6s. This is really why BBA are so popular there. If you want to see people driving low end cars, look at the streets of the US.

 

[InTheShadows]

I raced a lambo in my Model X while towing a trailer (I won).

What’s the scoring hit for ludicrous launch while in trailer mode?

View attachment 714147

Broken dishes, glasses, and mugs.

 

[generalenthu]

Semi OT, but Elon is back at the top of the heap edging past Besos.

 

[Paul_SF]

Or it could just be the press wanting a story about Shanghai giga and Tesla thought 300k car is a fun bonus so a good date to allow them in.

I don't understand all this hope for the cheaper Tesla. Not to mention how they are battery constrained before 4680s are in the hundreds of gwh production levels, but the Chinese are not a fan of having the brand being diluted with cheaper cars. It has always been a status symbol in China to own expensive cars more so than the west. It make more sense to introduce the cheaper car in the US first but will not happen till 2024 is my guess.

If you ever visit China, you'll be surprised that half the cars on the street are 7 series, S classes, and A6s. This is really why BBA are so popular there. If you want to see people driving low end cars, look at the streets of the US.

And you have your answer to your question in your own reply.

The people talking about the 25k Tesla are mainly from US, the “low end car” area.

The thing is, if you make a 25k product that’s way better than the high end ICE products, it accelerates the transition.

Only idiots will buy the expensive lesser product. And nobody is that stupid, anywhere.

 

[SOULPEDL]

The good doctor offers a compelling perspective about Tesla's disruptive future...

What if competitors simply offer to match any legit Tesla offer since they already know their relative risk just by the price to match? Unless that's not profitable for them. Maybe as a sacrifice to get the home etc, but I bet this drives down insurance costs overall, maybe some into the ground I think was his point.

We can hypothisize that Tesla drivers are also less prone to home fires (no oily rags, gasoline around)? I can't wait to see what Allstate offers me when Tesla arrives in Az.

 

[phantasms]

Or it could just be the press wanting a story about Shanghai giga and Tesla thought 300k car is a fun bonus so a good date to allow them in.

I don't understand all this hope for the cheaper Tesla. Not to mention how they are battery constrained before 4680s are in the hundreds of gwh production levels, but the Chinese are not a fan of having the brand being diluted with cheaper cars. It has always been a status symbol in China to own expensive cars more so than the west. It make more sense to introduce the cheaper car in the US first but will not happen till 2024 is my guess.

If you ever visit China, you'll be surprised that half the cars on the street are 7 series, S classes, and A6s. This is really why BBA are so popular there. If you want to see people driving low end cars, look at the streets of the US.

I’ve been to China. There’s a lot of money but there’s a lot of mopeds as well.

A LOT of people want a Tesla but simply can’t make the payments. The waitlist for 3 and Y grows by the day. 4680 is just around the corner and, probably, they won’t use that for the small one right away. Iron all the way. The Tesla Earnings Reports have indicated for a while this car was being worked on. 3 and Y are essentially fully ramped at Shanghai. Seems as good a time as any to introduce it. Also, much like the Model 3 wasn’t just a cheap Model S, this new thing will have a lower price but still be incredible. We’ll see.

 

[Nolimits]

Tesla drivers can now request Full Self-Driving Beta with the press of a button, despite safety concerns

Tesla rolls out 'FSD Beta button' after NTSB Chair Jennifer Homendy voiced safety concerns

www.cnbc.com

That didn’t take long..

No press is bad press?

 

[Singuy]

I’ve been to China. There’s a lot of money but there’s a lot of mopeds as well.

A LOT of people want a Tesla but simply can’t make the payments. The waitlist for 3 and Y grows by the day. 4680 is just around the corner and, probably, they won’t use that for the small one right away. Iron all the way. The Tesla Earnings Reports have indicated for a while this car was being worked on. 3 and Y are essentially fully ramped at Shanghai. Seems as good a time as any to introduce it. Also, much like the Model 3 wasn’t just a cheap Model S, this new thing will have a lower price but still be incredible. We’ll see.

The Chinese roads are not met for everyone to own a car. That's why plates are way harder to come by than the cash you need to buy a car. People generally don't waste a good plate on a crappy car, especially the wealthy who are willing to spend more money on a plate than a typical person would spend on a car.

 

[OrthoSurg]

I’ve been to China. There’s a lot of money but there’s a lot of mopeds as well.

A LOT of people want a Tesla but simply can’t make the payments. The waitlist for 3 and Y grows by the day. 4680 is just around the corner and, probably, they won’t use that for the small one right away. Iron all the way. The Tesla Earnings Reports have indicated for a while this car was being worked on. 3 and Y are essentially fully ramped at Shanghai. Seems as good a time as any to introduce it. Also, much like the Model 3 wasn’t just a cheap Model S, this new thing will have a lower price but still be incredible. We’ll see.

I'm starting to believe the waiting list for Teslas will grow to 2 years sooner than we think

 

[Singuy]

I'm starting to believe the waiting list for Teslas will grow to 2 years sooner than we think

It's kind of crazy that the supply for NA almost doubled and yet wait list keeps increasing.

 

[SOULPEDL]

Tesla drivers can now request Full Self-Driving Beta with the press of a button, despite safety concerns

Tesla rolls out 'FSD Beta button' after NTSB Chair Jennifer Homendy voiced safety concerns

www.cnbc.com

That didn’t take long..

No press is bad press?

NOOoooo not the PETS!!! That's pretty low.

"FSD Beta gives drivers access to an “autosteer on city streets” feature, which has yet to be perfected and enables drivers to automatically navigate around urban environments alongside other vehicles, pedestrians, bicyclists and pets without moving the steering wheel with their own hands"​

 

[lascavarian]

What if competitors simply offer to match any legit Tesla offer since they already know their relative risk just by the price to match? Unless that's not profitable for them. Maybe as a sacrifice to get the home etc, but I bet this drives down insurance costs overall, maybe some into the ground I think was his point.

We can hypothisize that Tesla drivers are also less prone to home fires (no oily rags, gasoline around)? I can't wait to see what Allstate offers me when Tesla arrives in Az.

They could match but they don’t have access to the driving data like Tesla and they can’t dynamically adjust prices based on rolling avg use of FSD.

When it comes to house coverage, there is the potential to utilize the sensors and cameras on the vehicle to have a far superior picture of residential fire risk Dynamically. The car has a horn that can alert as well as the ability to text fire concerns.

Similar to the auto industry the insurance industry (to me) has seemed resistant to really innovating in any way other than advertising.

I see this Safety score as a leveraging of NN learning that goes way beyond actuarial tables and it is a bit out of the blue. It is a peek into an interesting future.