☕ Signal or noise, smart investors know which is which
Oct 22, 2026Howdy! 👋
“Tesla on deck,” I said as I signed off yesterday.
You know who certainly stepped up to the plate last night.
If all you read this morning are the headlines, you’d think he struck out. Auto revenue down. Energy revenue down. Profits down. Stock down ~4%.
Shame because the folks writing those headlines are reading the front page.
They’re reacting to three things: comments from Nvidia CEO Jensen Huang about AI compute demand and capital scarcity, Meta’s observation that “large” amounts of funding will be needed, and renewed kinetic risk in Iran.
Keith’s Rule of the Back Page applies.
Not a one of these things impacts the long-term viability of what’s happening nor does it disturb the business case for investing. In fact, today’s selloff is yet another gift for smart investors.
I won’t be the least bit surprised if the markets finish green today.
Please take a long, measured and very deliberate, careful look at this chart.
Let it sink in.
Then… DO… SOMETHING… ABOUT… IT!
I know I sound like a broken record when I say that the markets are the only place on earth where people fear a sale.
Here’s the hard truth.
If the numbers didn’t show that most investors can’t control their emotions, I wouldn’t have to bring that up yet again.
Short term noise makes long term profit potential cheaper and the sooner you internalize that the better.
Tesla’s results are a textbook example.
Let’s talk about ‘em.
Here’s my playbook.
1 – Tesla beat. Again. You just won’t hear much about it.
Tesla earned 48 cents a share on an adjusted basis versus 43 cents expected. Revenue came in at $28.05 billion versus $27.9 billion expected. (Read)
A double, meaning it beat analyst expectations on both the top and bottom line.
Predictably, the spreadsheet gang is fixated on everything that went down. Auto revenue slipped 3%. Net income fell 43% from a year ago.
Terrible right?
Wrong.
Smart investors will look at why.
Research and development spending jumped 69% to $2.76 billion, most of it on AI, robots, and the infrastructure behind them. Stuff that will show up on the sheets a few quarters from now exactly as it did with SpaceX which transitioned from a $1.05 billion loss last year to a $4.4 billion profit this quarter.
I’ve talked about how modern accounting rules can’t adequately capture what’s happening, so I won’t repeat all that today.
As plainly and respectfully as I can put it, Tesla is buying its future with its own money.
Gross margin improved to 19.6%, up from 18% a year ago and 16.8% last quarter. This matters because gross margin is what’s left from every sales dollar after paying to build the product.
Rising gross margin while spending like crazy on the future is no easy feat.
2 – Demand isn’t the problem. Batteries are.
This is the part I wish more people would read twice… no scratch that, as many times as it takes.
Tesla ended the quarter with the largest order backlog for its passenger vehicles in four years. A backlog is orders customers have placed that Tesla hasn’t filled yet.
Why?
Because battery supply is still the limiting factor.
Read that again.
Customers are lining up to buy Tesla’s cars and Tesla can’t build ‘em fast enough because there aren’t enough batteries. This is a world class problem that Ford, GM and virtually every automaker on the planet would give their socket wrenches to have, btw.
Not for nothing but Tesla had record deliveries in Canada, Australia, Japan, and Taiwan. Gigafactory Berlin had record production. Tesla also built its 10 millionth vehicle.
Musk says Tesla is aiming for 5 million vehicles a year by 2030… three times what it delivered over the past 12 months. Again, nearly every Western automaker is struggling to sell… anything.
Here’s a detail almost nobody will mention.
Tesla used to earn a nice chunk of change selling “regulatory credits” to other automakers who couldn’t meet emissions rules on their own. That revenue fell to $113 million from $417 million a year ago but auto margins excluding those credits rose to 18.1% from 15.4%.
In other words, Tesla is making more money on more cars, not by selling hall passes to the competition.
Btw, the fix for the battery bottleneck is already underway: pack capacity in Berlin, lithium refining and cathode materials in Texas, and LFP cells in Nevada. I’ve been banging the drum on that refinery for a while now.
And yep, also exactly as I suggested would be the case, the Tesla Semi is now in volume production, too. The top version goes 500 miles fully loaded at 82,000 pounds. It uses about a third of the energy a comparable dinosaur juice-guzzling diesel truck burns for the same load.
I want one but my bride simply smiles.
Go figure??!! 😜
3 – FSD is the tell
There’s only one thing to think about if you want to know where Tesla’s headed – pun absolutely intended.
Active Full Self-Driving subscriptions hit 1.74 million, up 67% from a year ago.
In North America, more than 75% of new Tesla buyers now take FSD with their car. Analysts call this the “attach rate” – meaning the share of buyers who add FSD, and it’s up nearly 20 percentage points in a single quarter.
Every new Tesla already comes with a free one-month trial but the thing that stands out to me is that buyers are signing up to pay anyway.
Musk also said more and more people are walking into Tesla showrooms to experience FSD, not to shop for a particular model.
Think about that.
When’s the last time anybody walked into a car dealership to try the software?
I’m not surprised.
FSD is that good.
We’ve logged thousands of miles on FSD without a single glitch… in crowded cities, on rural roads… the interstate… in all kinds of weather ranging from sunny days to snow and the dark of night.
Our friends were stunned when they asked after about 30 minutes in the car if we could “try FSD” only to learn it’d been on the entire time. The look on their faces was priceless.
FSD, btw, is now approved in nine European countries too. Musk said he’d be surprised if Europe-wide approval takes longer than the first quarter of next year, though that’s up to the regulators.
Then there’s Robotaxi.
As I said back in July 2025, “Robotaxi is an iPhone moment, right? People don’t understand this. They think it’s just about taxis and they’re constantly comparing it to Waymo.” (Watch)
Tesla started giving rides in the Cybercab in Austin on September 3. Cumulative driverless miles grew eightfold during the quarter to nearly 2 million. Service now runs from 6 a.m. to midnight.
Musk says the chart “looks like a rocket going to orbit.” Fitting, considering what we talked about yesterday. 😄
Next up is Dallas in November, with more cities to follow. Tesla’s goal is 24/7 service by the end of this year.
Here’s the part I’ve been waiting for. Tesla expects some customer-owned cars to join the Robotaxi network around the second quarter of next year. Earlier this year I said, “Every Tesla owner can potentially send his or her car out to work for the day when they’re not using it.” (See #8)
That’s starting to happen.
You may remember me saying that Tesla’s real advantage has never been the car. It’s the data engine behind it. Waymo buys more hardware to grow. Tesla hits a button and potentially makes billions.
To be fair, Musk was clear that Robotaxi won’t “move the needle” financially this quarter. He expects it to become noticeable in the second quarter of next year and a big deal by late 2027.
I agree.
Call me crazy but I’d rather own the company running late on the future than the ones still arguing about whether it’ll arrive.
4 – The cash question… and the cash machine
Capital spending this quarter more than doubled to $4.7 billion. Capex is the money spent building factories, chips, and equipment.
And yet…
Free cash flow came in positive at $1.5 billion, a big swing from negative last quarter. Free cash flow is the cash left over after paying the bills and building the future. Cash and investments rose $1.9 billion to $45.4 billion.
Heading into the report, many analysts expected another quarter of negative free cash flow. (Read)
And yet...
Tesla delivered $1.5 billion of positive free cash flow instead.
Convenient how the worrywarts are leaving that out of their doom-laden missives.
Energy had a softer quarter. Revenue fell 16%, partly because of tariff-related customer refunds. But storage deployments still grew 10%, the order rate and backlog hit all-time highs, and Megafactory Texas just started production with planned capacity of 50 GWh. AI data centers are thirsty for power, and Megapacks are a fast way to supply it.
Even so and despite that, energy gross margin came in around 33%, up from about 31% a year ago.
And Musk says Tesla’s Megapack production capacity for 2027 is nearly sold out. Management expects higher prices and better margins from here.
Sold out… a year ahead of time.
I’ve said for years that energy is Tesla’s cash machine. (See #1)
Then there’s SpaceX. Tesla bought a stake earlier this year, and SpaceX shares jumped this week. That’s a nice little kicker for next quarter.
5 – What’s next? Robots, chips and a 2027 to circle on your calendar
Most companies buy AI.
Tesla is building the entire dang stack – meaning every component from top to bottom.
Optimus, Tesla’s humanoid robot, is now in production on high-volume lines in Fremont. Tesla says these are the first production-quality bots it’s built.
Musk was refreshingly candid about the road ahead, too. “The growth of Optimus is an S curve, but the initial curve is a shallow slope,” he said. He also said there’s “still quite a lot of work to do,” and he expects the remaining issues to be solved around the end of next year.
Candor is underrated in corner offices these days and I’ve always appreciated his even if I didn’t agree with it at times.
Most of the equipment for a line that can build about 1 million robots a year is already in Fremont. Musk expects it’ll take at least another year, possibly longer, to get to that pace. That’s roughly 20,000 robots a week.
He also teased a new Optimus demo video in the next month or two, showing genuinely AI-driven tasks rather than preprogrammed moves.
Grab the popcorn.
He thinks Optimus will be “the biggest product on Earth.”
Again, I agree for reasons we have talked about many times.
On the compute side, Tesla expects to reach 400 megawatts of AI training capacity by year-end. Think of that as the engine room that teaches both the cars and the robots how to think. Tesla also expects to start installing equipment at its Austin chip fab in 2027. Musk hopes to have the first experimental chips around May or June and useful chips at volume around the third quarter of 2028. He called it “the most valuable building in the world.”
I told you a while back that Optimus would be so successful people would forget Tesla makes cars. (See #2)
This earnings report makes it clear that we’re one step closer.
As for the merger everyone was buzzing about yesterday, Musk kept it short: “I think I said everything that should be said yesterday.”
Fair enough.
The building blocks are already in place, though. Tesla and SpaceX are already working together on a chip factory in Grimes County, Texas, alongside Intel.
To reiterate something I mentioned yesterday.
If you don’t like Musk or have been sitting on the sidelines for whatever reason, you’re not alone. Plenty of smart people have for plenty of reasons.
I’m not here to change your opinion nor stand in judgement. In fact, I respect your conviction 100%.
But if you haven’t invested because you doubt whether he’ll be able to get the job done or simply cannot reconcile the results with the investing potential, that’s another matter entirely.
At some point, the numbers really do add up… the cars are getting built, as are the robots, the trucks, the energy, the AI and more.
Tesla has created more wealth faster for investors than almost any other company on record. In fact, the stock has returned roughly 34,000% since its June 29, 2010 IPO through last night's close. For context, that’s enough to turn every $1,000 invested at the baseline IPO pricing into $341,000 as of last night’s close. The S&P 500, by comparison, has returned about 645%.
Keith’s Investing Tip: Profit potential is directly correlated to the ability to change your mind when presented with accurate information that contradicts your perception and your beliefs.
Bottom Line
A while back I said Tesla could be worth $20 trillion. Musk’s response was that it’s possible with “extreme execution.” And, I submit we’re seeing that play out in plain sight.
Profits are always found on the edge of absurdity.
You got this – I promise!
Now and as always, let’s MAKE it a great day. 💯
Keith 😀
