The 47% earnings headline has an asterisk… the sector actually carrying the load… and the guy who keeps building things nobody saw coming
The second-quarter earnings season has turned into one of the strongest in years.
The S&P 500 is on pace for blended earnings growth of 47.4% for Q2. If that holds, it’ll be the best growth rate the index has posted since Q2 2021.
Just one week ago, that estimate sat at 38.0%. Two weeks before that, it was 23.2%. In other words, this number has nearly doubled in a month.
So, what’s driving it?
Mostly two companies – but not for the reason you’d think.
Real strength? Or the illusion of strength?
Alphabet, Inc. (GOOG) and Amazon.com, Inc. (AMZN) both posted eye-popping earnings surprises this quarter. But dig into the numbers, and you’ll find the “surprise” wasn’t really about search ads or Prime subscriptions.
Here’s FactSet on Amazon:
It is important to note that the (GAAP) EPS actual for Amazon.com for Q2 2026 included non-operating, pre-tax other income of $53.4 billion, primarily from investments in Anthropic.
However, as previously stated, the vast majority of analysts providing EPS estimates to FactSet used the (GAAP) actual EPS of $5.75 including the other income as the comparable number to their estimates.
Alphabet’s beat has a nearly identical story. Its own earnings release attributes $99.0 billion of its profit to unrealized and realized gains on equity securities – stakes in companies like Anthropic and SpaceX – which by itself boosted diluted EPS by $6.26. Strip that gain out, and Alphabet’s “beat” mostly disappears.
So, add Amazon’s $53.4 billion and Alphabet’s $99.0 billion together, and you get $152.4 billion in paper gains from the two companies’ investment portfolios – padding this quarter’s headline profit numbers before either company sold an extra ad or shipped an extra package.
But here’s the thing – these two companies are so large, and their paper gains so massive, that they’re not just inflating their own numbers – they’re pulling the entire S&P 500’s headline growth rate up right along with them.
Two stocks, out of 500, are a major driver of the index’s “best since 2021” story. And what happens if you remove Alphabet and Amazon entirely?
The blended earnings growth for the S&P 500 would fall from 47.4% to 28.8%.
That’s still a very good quarter. It’s just not a record-breaking one.
So, let’s dig deeper to see where the gains have been concentrating.
This sector is doing the heavy lifting
That gap – between the flashy 47.4% headline and the more grounded 28.8% reality – is exactly why it’s worth looking past the index-level number and asking a sharper question…
Beyond Big Tech, where has the strength been coming from?
Energy.
It’s posting the strongest year-over-year earnings growth of any sector in the index, at 135.3%. And these gains aren’t riding on one or two names.
Four of the sector’s five sub-industries are showing double-digit growth, led by Oil & Gas Refining & ÃÛÌÒ´«Ã½ing (277%) and Integrated Oil & Gas (172%).
But is now still a good time to be in oil? Aren’t prices in danger of crashing given the growing chatter around a new potential ceasefire deal with Iran?
On that note, this morning, Treasury Secretary Scott Bessent said that the U.S. and Iran could reach a deal to open the Strait of Hormuz today or tomorrow.
While the oil markets would certainly breathe a sigh of relief if that happens, it might matter less than many investors realize. You see, the real bottleneck right now isn’t crude – it’s refining.
Where the money is likely to keep flowing within the oil patch
Between disruptions in the Strait of Hormuz, Ukrainian strikes on Russian refineries, and China’s export ban, roughly 10% of global refining capacity is effectively offline. This is keeping fuel prices and refining margins elevated even as crude wobbles.
Here’s Bloomberg from last Friday, quoting ExxonMobil’s CFO:
“The constraint pain point in the energy system is refining,” ExxonMobil Chief Financial Officer Neil Hansen said in an interview.
It’s “something that perhaps the market isn’t fully focused on.”
Translation: even a ceasefire that cools crude prices may not bring fuel prices – or the refiners’ margins tied to them – down with it. Or at least not as fast as the market expects. So, this trade could have more life in it than the “oil is falling” headlines would have you believe.
This is part of why legendary investor Louis Navellier has put his subscribers into refiners.
From Louis’ latest issue:
Global oil refining margins recently breached another record high, as global refining capacity remains pinched due in part to restrictions on Russian diesel, limited shipping traffic in the Strait of Hormuz and Ukraine’s relentless attacks on Russian refineries.
As a result, many American refineries need to meet the ongoing demand for refined petroleum products like gasoline and jet fuel.
One refiner that both Louis and trading veteran Jonathan Rose of like is HF Sinclair Corporation (DINO).
While Jonathan has played DINO due to the crack spread, Louis has been attracted by the earnings strength.
On that note, the company’s refining segment alone earned $877 million, up from $166 million in the second quarter of 2025. HF Sinclair’s refinery gross margin also surged to $25.95 per produced barrel sold, representing a 57% year-over-year increase.
Louis recommended DINO at the end of June, and the official position is already up more than 30%. As I write, it trades under Louis’ “Buy Below” price of $100.
Another name that shows what “real” earnings growth looks like
DINO isn’t the only place Louis is finding real earnings strength – the kind that’s backed by an actual, growing business, not unrealized gains on an investment.
Across his Growth Investor Buy List, Louis has been watching the same pattern repeat all season. Here’s Louis:
Our Growth Investor stocks are characterized by 53.7% average annual sales growth and 112.4% average annual earnings growth.
Our stocks are also on track to post wave after wave of positive earnings surprises, as the analyst community has revised earnings estimates 16.8% higher in the past three months.
So far this earnings season, we’ve had 21 companies report results, and 18 have exceeded analysts’ earnings estimates. .
Eighteen of 21. That’s a beat rate right in line with the record pace the broader market is putting up – except these beats are showing up in companies actually building the AI boom, not just holding stakes in it.
Take Celestica, Inc. (CLS), another one of Louis’ Growth Investor holdings.
It manufactures the servers, networking gear, and data center hardware that hyperscalers like Amazon, Microsoft, and Google need to actually run their AI workloads – the physical plumbing behind the AI boom, not the AI itself.
Louis just noted that last week, Celestica reported second-quarter revenue of $4.70 billion, up 62% year-over-year, with adjusted earnings per share of $2.54, up 83% from a year ago. Management also raised full-year guidance to $20.5 billion in revenue and $11.30 in adjusted earnings per share.
Once again, Celestica’s earnings weren’t goosed by an equity stake in a venture capital company. They came from the company selling more hardware, at better margins, because demand for AI infrastructure keeps outrunning supply.
Louis’ Growth Investor subscribers are up 84% in CLS as I write. But here again, you have room to get in. Louis’ “Buy Below” price is $409.
For all of Louis’ picks in Growth Investor, .
Now, I just mentioned “an equity stake in a venture capital company” like it’s a bad thing. Of course, it’s not if you’re the one who owns the stake, and not if you got in early with the right person.
Which brings us to our next story…
The guy who keeps building things nobody saw coming – twice
Palmer Luckey built his first virtual reality headset out of spare parts and duct tape in his parents’ trailer, while homeschooled and still a teenager in Long Beach, California.
By 21, he’d sold that company, Oculus VR, to Facebook (now Meta Platforms (META)) for $2 billion.
Most people would call it a day and end up on a white sandy beach somewhere.
Instead, within a year of leaving Facebook, Luckey founded a defense company technology startup called Anduril, building autonomous drones and AI-piloted weapons systems for the Pentagon.
Today, private backers value Anduril at somewhere north of $60 billion.
One breakout company, built by a kid tinkering in a trailer, might be a matter of luck. Two breakout companies, in two completely unrelated industries, is skill.
But Luckey isn’t done…
He’s now working on a small nuclear energy startup called Valar Atomics. It builds compact reactors designed to sit directly behind data centers and supply them with power around the clock, a real bottleneck as AI’s electricity appetite outruns the grid.
In March, private investors valued Valar at $2 billion. As of two weeks ago, it’s already back in the market, talking to Sequoia about a new round – at a $6 billion valuation.
That’s a triple in about four months.
With this track record, would you be willing to bet on Luckey’s next company?
Of course. And it points toward a truth that every venture investor will tell you…
A great founder with a decent idea beats a great idea saddled with a mediocre founder, almost every time.
Why?
Simple – ideas get copied. ÃÛÌÒ´«Ã½s shift overnight. Competitors show up from nowhere. The one variable that adapts to all of that in real time is the person running the company.
That’s exactly why the first letter in Luke Lango’s PPT framework – People, Product, Timing – isn’t an afterthought. It’s the filter he runs before he looks at anything else.
Here’s Luke:
Getting through the door is not the same as making money.
The fact that a company is raising money does not make it a good investment. It makes it an opportunity to evaluate…
PPT starts with: Are the founders the kind of people who figure things out when everything goes wrong?
This question is exactly why a name like Luckey moves the needle before a single dollar of revenue shows up on a spreadsheet.
It’s also exactly the kind of filter that matters more in AI than almost anywhere else right now, given how many companies are raising money simply because they can, not because they should.
Luke dove deeper into this question and his broader PPT framework in last Thursday’s . He walked through how he applies People, Product, and Timing to the private AI opportunities he’s tracking today – including one opportunity he believes could be a portfolio-maker.
As I covered in yesterday’s Digest, Luke gave it away during last week’s presentation. It’s a robotics startup offering a one-of-a-kind “robot school” platform that Luke considers the “ChatGPT moment” for artificially intelligent robots.
.
As for Luckey, he’s not done with Valar Atomics either. He’s also founded Erebor Bank, a national digital bank tailored specifically for deep tech, defense, crypto, and hard science startups.
Keep your eye out for opportunities with people like Luckey. History shows that builders who win once rarely stop at once – and .
We’ll keep you updated on all these stories here in the Digest.
Have a good evening,
Jeff Remsburg
