The past decade has seen the rise of several new trading platforms. Robinhood, founded in 2013 used commission-free trading and a Covid-trading boom to get to 27.4 million funded customers. More recently, Kalshi rode a friendly regulatory environment to $265M in fee revenue and a $22 billion valuation.

An even newer story is Hyperliquid. Hyperliquid was launched in November 2024. Run by a team of ~15 people in Singapore, Hyperliquid in the last 30 days has carried out over 100 million trades and $178B in total trading volume.

Hyperliquid has brought in over $800M of fees last year.

All of this is an incredible growth curve and raises some questions: such as 1. how did Hyperliquid take off so quickly, 2. what’s being traded on there, 3. where do all the fees go? And relatedly: how can one get exposure to Hyperliquid’s growth? Stock investors, don’t worry- there’s a Nasdaq listed proxy.

Decentralized Trading, Perps, and Leverage

Hyperliquid falls under a category known as DEX or decentralized exchange. Decentralized exchanges use smart contracts, algorithmically tying your cash to the trade, effectively bypassing the need for brokers and time-consuming settlement.

A DEX can support a 24/7 cash settled future in almost any index or commodity. Even in illiquid markets. How?

The exchange relies on an oracle price, a benchmark the future on Hyperliquid is supposed to track. For BTC this could be the latest settlement price from Binance. When the price drifts from the underlying, i.e BTC is at $100K on Binance but $101K on Hyperliquid, a funding mechanism kicks in.

The longs will pay a yield for shorts to enter the market, set hourly until the $1K is resolved. The result: every exchange on hyperliquid has an equal amount of longs and shorts no matter how lopsided market sentiment might be.

This sets up the next trick. Since longs and shorts are in balance, Hyperliquid can match counterparties for absurd amounts of leverage without needing a balance sheet. If trader A has $1K and wants to be 30x long bitcoin, Hyperliquid doesn’t need to lend him $29K. It just matches his trade to a $30K short position. If bitcoin goes down 2% and his long position is liquidated, a profitable short position is closed (gaining $1K).

While Hyperliquid isn’t the only exchange that does this (Perpetual Contracts), it is the market leader. In April, Hyperliquid processed $181B of contract volume, 3x Aster, the #2 player.

Perpetual volumes as a whole are down since 10/10 (along with crypto volumes generally), but Hyperliquid has been gaining share.

Why? On 10/10, a threat of a 100% tariff on China led to $19B in Bitcoin positions were liquidated in a single day. This was a stress test for all leveraged crypto protocols and not every exchange passed, As Crypto Blog The Defiant reported:

”Lighter and other smaller perps venues went fully offline for minutes at a time. Meanwhile, Hyperliquid stayed online and profited.”

Critics argue Hyperliquid’s leverage mechanism accelerated the crash, but for traders, an early liquidation is preferable to a platform being offline during heightened volatility. Hyperliquid has consolidated its position as the leader.

There are also returns to scale. As Hyperliquid has the most active volume, it has the lowest slippage on new BTC trades, which helps fortify its lead over time.

Volume to Fees

Whenever a trade occurs on Hyperliquid, a taker fee and a maker fee is charged. While there are discounts for volume, the taker fee starts at 0.045% of the trade value and the maker fee 0.015%. As hundreds of billions are traded on the platform, Hyperliquid captures a small component. So on ~$181B perp volume in April → $60M platform fees, or 0.03% on average. So the more volume Hyperliquid does, the more fees it generates.

Where do all those fees go? Hyperliquid has a crypto coin called HYPE. Something on the order of 98%-99% of all fees go directly to buybacks (in crypto mechanics, they “burn the float”, as in move to an inaccessible wallet)

For analysis terms, this means the HYPE coin is more or less equivalent to the equity of Hyperliquid the company, given fees are the main revenue stream and 98% of fee revenue goes directly to buybacks, and the coin represents the main financial interest of the founders.

Quick Valuation Note

So we have a company founded in Nov. 2024 that made $814M last year with 98%+ operating margins and all profits going to buybacks.

So what’s a fair valuation? The HYPE coin is currently at a fully diluted market cap of $40B. So 50x trailing earnings. If you extrapolate the YTD pace that would be $590M or 67x earnings. None of that seems crazy for an asset this unique and high-margin.

The relevant question is was October last year a long-term top in fees? Or can Hyperliquid regain fee momentum and accelerate growth into the 2H of the year? For this, there are 3 relevant catalysts worth highlighting.

Three Catalysts

1. Hyperliquid has been expanding from crypto markets

In October, Hyperliquid was only a crypto trading platform but that has changed dramatically. The HIP-3 upgrade allowed developers to create markets on top of the Hyperliquid framework.

When the Iran war started on a weekend, traders turned to Hyperliquid’s Oil future market, hitting $1.7B in peak volume. TradeXYZ built equity trading on top of Hyperliquid, including the official S&P 500 index. Almost ~10% of HYPE fees are now being generated from non-crypto activity.

This diversification will continue with HIP-4, which brings prediction market functionality to Hyperliquid. HIP-4 launched officially on May 2nd, meaning this is a very live catalyst as Hyperliquid’s prediction market ecosystem builds out.

HIP-4 has a distributed structure where anyone with 1M HYPE (~$40M) can deploy a prediction market, with their stake at risk if they manipulate outcomes. Over time, this could lead to more variety vs. Polymarket/Kalshi which curate markets.

Beyond prediction trading, HIP-4 can be hedging tools within the HYPE ecosystem (hold a stock while buying a ‘no’ on its next earnings beat). The more HYPE can activate non-crypto activity, the greater the argument for its outperformance vs. crypto as a whole.

2. Potential Crypto Recovery

A YTD performance chart of BTC, HYPE, and the S&P 500, reveals two things:

1. HYPE has been vastly outperforming BTC (64% vs. -9%)

2. BTC was at one point down 25% YTD vs. the S&P but has narrowed that gap considerably despite macro turmoil

Together this suggests a potential recovery from bitcoin with pessimism having peaked at some point in February.

One reason for optimism? On May 1st, Senators Alsobrooks and Tillis released a new compromise draft of the CLARITY Act after months of negotiations between banks and crypto firms. The act which clarifies CFTC over SEC regulation over most crypto use cases, is widely seen as a positive for the crypto industry but has been on hold due to questions on rewards programs as a loophole for paying interest on stablecoins.

The compromise bill delineates between interest and activity-based rewards and is now likely head to a markup and vote over the summer. If the act catalyzes an improvement in crypto as a whole, HYPE will be a beneficiary.

Crypto trading volumes as a whole are down 60%+ from October. A recovery in volume with Hyperliquid’s improved market position would lead to meaningful fee growth.

3. Possible US Market Entry

Hyperliquid is currently banned for US users (some use VPN) but they are making a legitimate lobbying push. In February, Hyperliquid launched the “Hyperliquid Policy Center,” with $29M in initial funding.

Jake Chervinsky, previously Chief Legal Officer at Variant is serving as CEO of the center and is making the argument that “the regulatory picture for crypto derivatives in the United States is uncertain”

While a tail outcome, US approval would lead to a derivative adoption by one of the largest crypto markets by trading volume. It may be possible for Hyperliquid to find a way into the CLARITY act before it passes.

Price Target:

Retaining a 50x multiple on Hyperliquid's fees, the price can move to the $52-$88 range should volume recover. The bet is that between HIP-3 continued growth, HIP-4 traction, and any crypto recovery, HYPE drives its fee run-rate above last year's peak. $70 as a 6-month PT for HYPE is achievable without multiple expansion.

How to Buy It

Three Options for Buying Hyperliquid

1. Buy HYPE direct on a crypto exchange (Robinhood, Coinbase): as the 10th largest crypto by MC, most exchanges now offer it

2. PURR (NASDAQ) - Hyperliquid strategies: this is a Hyperliquid treasury company: it sells stock to buy HYPE tokens. Currently it is trading at 1.05x NAV or 1.15x including its tax liability (if it were to sell all HYPE). MSTR which popularized the category, traded above NAV before compressing after an ETF launch.

3. TXXH (NASDAQ) — 2x daily leveraged ETF, launched last week. Note a 2x leverage ETF will have slippage: if HYPE was down 5% one day and up 5% the next, the underlying would be down 0.25% but the 2x ETF would be down 1% (as went down 10% before up 10%)

Why not just wait for the spot ETF?

• While there will be a spot ETF eventually (even XRP got one), Bitwise just filed its S-1, so actual ETF approval is likely 3+ months away

• Tokens front-run ETF approval. BTC rallied almost 50% from October 2023 to ETF approval in January 2024

• The most meaningful catalyst window, HIP-4 traction + clarity act is within the next 3 months

• In short: the 1.15x premium is the price of being early.

For newsletter tracking, PURR will be the pick. While 1.15x premium is not ideal, that is the price of an accessible equity wrapper. Assuming mNAV compression from 1.15x to 1.05x during the trade window, this translates to a $9.78 PURR price target (+56%).

Risks

• HIP-4 Fails to Gain Traction (want to see similar to HIP-3 a trajectory to 10% of volume)

• CLARITY Act fails to pass (stalls in Senate)

• Crypto rolls over again (HYPE would get hurt in a broader crypto risk off)

• mNAV compression/slippage (for equity instruments)

• Another exchange rises to the top (Hyperliquid is currently 3x its closest competitor)

Disclosure: Author is long PURR shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research. Past results are not indicative of future results.

Chaos in the Middle East. Crude oil is trading above $100 a barrel. While this has been a clear tailwind for oil stocks in general ( is up almost 40% YTD), there is a producer in Argentina that is adding drills into the crisis faster than almost any US-based E&P and trades at a fraction of their valuations.

Vista Energy was already riding a secular trend: the Permian boom of the last 15 years has peaked. The largest shale formation in the Americas outside of Texas is the Vaca Muerta. Combined with the most corporate-friendly political regime in years, Argentina was poised to experience its own version of the 2013 shale boom before the Iran war broke out. Vista Energy grew production 66% YoY and aims to double exports from 22 million barrels in 2025 to 44 million in 2028. They added over 10% production YTD… and they’re not done. Yet the market is valuing them at 11x last year’s earnings… on fears oil prices will revert to the pre-war levels in the $60s. Does that make sense? Vista energy is poised for further upside even on a quick resolution of the war.

Oil Prices are Not Returning to Pre-War Levels: $80 is the new $60

The most common retort from investors is “how can you invest in energy when the war could end tomorrow?” After all two weeks ago the war was “very much complete.” Marco Rubio told the G7 this past weekend the war was “2-4 weeks” from wrapping up. The pentagon is preparing for “weeks of ground of operations.” Whether the war is ending in days or months, investors have been overly focused on the timeline of the Iran operation and not the reality that the post-war environment will have structurally higher oil prices. Why?

1. The existing closure has already significantly set back production
With the Strait of Hormuz closed, Gulf Countries have shut in production as their storage facilities have filled. To quote the IEA, “Gulf countries have cut total oil production by at least 10 mb/d. In the absence of a rapid resumption of shipping flows, supply losses are set to increase.” As we’re a month into the operation, this means there are ~300M barrels that never came out of the ground.

Additionally, this production that was shut-in does not get re-opened overnight. As Rystad Energy’s Aditya Saraswat noted, “It could take two to three weeks to restore full production at a small field that’s been halted, and four or five weeks for a larger one.” So even if the war ends tomorrow, there will be a lag to ramping up to full production.

2. The strategic reserves that have been drained need to be refilled
Countries have responded to this significant drop in production with releases from oil reserves: “IEA member countries agreed on 11 March to make available an unprecedented 400 million barrels of oil from their emergency reserves available to the market to mitigate the negative impact on economies from the supply disruptions.”

The issue is the current drain follows the 2022 reserve IEA reserve drain of 182 million barrels, which was also “unprecedented”. Many countries did not fully refill their reserves to pre-2022 levels. The US is projected to end the war with the lowest levels in the strategic reserve since 1982.

While the US may delay the refill until after midterms, there will be a critical need to refill the SPR from these low levels. China, which also maintains a sizable strategic reserves for strategic/commercial reasons, is likely to refill any reserves drained during the war.

3. Neither OPEC or the Permian Can Afford Sub-80’s

The damage to Middle East infrastructure has been extensive. Energy infrastructure has been targeted throughout the war. Even without further escalation, over 40 Middle East energy assets have been “severely damaged” per the IEA, in addition to lost tourism revenue, airports targeted among other collateral damage.

While Saudi Arabia and other gulf nations have sub-$40 barrel breakevens, their budgets assume significantly higher oil prices. Saudi Arabia’s Fiscal Breakeven before the war (to be in a government surplus) was $94/barrel. Now with the added burden of repairing war damage/replacing missile defenses, Saudi Arabia (and other Gulf States) have a massive fiscal hole.

While OPEC in the mid-2010s facilitated “price dives” to take out US Shale players, it does not make sense in the post-war environment. US shale players don’t want $60’ oil either. After ramping significantly over the last decade, the permian production is considered to have peaked.

Like any oil boom, the easy sites were tapped first. The sites being drilled in the Permian today require significantly higher breakevens. Execs surveyed by the Dallas Fed in March stated they need at least $60-$70 oil prices to profitability drill a new well.

They were also asked by the Dallas Fed if they planned to increase their drilling in light of the recent events. 50% of E&P’s said they would not change their plans, even with the current oil surge. Only 21% of E&P’s said they would increase drilling significantly.

At the CERAWeek energy conference in ‌Houston held on March 23rd, ConocoPhillips’ Nick Olds said they were “not currently increasing production (in the lower 48 states)” adding ConocoPhillips would need to see “sustained higher prices.” SLB exec Steve Gassen stated “the cycle from the time you begin to when you make a decision that you're going to add rigs to then ultimately drilling and producing and getting to market, that can be a year-long process, even in the ​U.S., which is a short-cycle market. Nine months would be the very best-case scenario.”

The US which was “the swing producer” is not adding rigs into this crisis. So who is?

Vista Energy: A Rapidly Scaling Pure Play Argentina Shale Producer

The Vaca Muerta is the “second largest shale gas reserve and fourth largest shale oil reserve on the planet.” Compared the Permian, the Vaca Muerta is relatively untapped. Wells are 20% more productive than the Delaware basin and 100% more productive than the Eagle Ford on a barrel/lateral foot basis.

Vista Energy’s CEO/Founder Miguel Galuccio was formerly the CEO of Argentina’s State Oil Company YPF, where he was branded "the Architect of Vaca Muerta.” While YPF had much more resources it is also 51% government owned. Galuccio took Vista Energy public in 2019 as a more “nimble” independent pure play. It has worked out spectacularly for shareholders, with the stock returning almost 800% since inception

Today, Vista energy owns 229,000 acres with 1,300 ready to drill locations.

Unlike most shale producers, they are operating at only 1.5x leverage. 64% of their barrels produced are exported outside of Argentina into the global market.

While the operating performance and track record are already better than most independent oil producers, a more recent move speaks to the quality of management.

On February 2nd, just weeks before the start of the war, they bought Equinor’s full onshore Argentina assets. The deal added ~22,000 barrels/day to Vista’s production. Management bought this asset for US$712 million for a highly accretive 3x 2025 EBITDA (at a $70.2 average oil price). The deal also includes an annual contingent payment linked to oil prices but no payment is due below $65 Brent and above $80, Vista keeps 100% of the incremental upside on every barrel produced from the acquired assets.

Shale is already the fastest type of oil asset to scale production. Yet Vista has a head start on every shale producer as it was actively acquiring and drilling ahead of the crisis, instead of trying to catch up on a 6-9 month timeline.

Vista By the Numbers

Vista energy had already guided to a great year on its Feb 26th earnings call. This guidance included growing production 22% and showing operating leverage with lifting costs only going up 7% and FCF doubling.

The fun part is the guidance assumed $65 oil for the rest of 26. Here’s the current futures curve for Brent oil.

Analysts are pricing this is as a massive jump in profits in Q2, followed by declines in Q3/Q4. Even with this assumption, Vista generates $2.8B in EBIDTA and $1.6B in operating income.

The company as a whole is valued at $10.1B EV even after the 50% run-up YTD, putting it at a 3.6x 2026 EV/EBIDTA. Unlike most oil companies, Vista is growing production fast enough to outweigh the expected price decline in the 2H of the year (see the 2% QoQ growth in Q4).

On the FY25 Q4 earnings call (February 26), CEO Galuccio called out that the company had added 10 wells in January with “very good productivity,” leading to a production rate of 132,000 barrels of oil per day in February. He added that March was a “very good surprise” estimated to surpass 140,000 bbl/d, the full-year guidance number. And that’s before Vista gets the 22,000 barrels from Equinor (closes in Q2).

A recent UBS note on February production confirms the ramp is accelerating, with Vista’s production “up by +10% MoM” with 4 new wells “in production early stages, implying potential additional production ramp up ahead.”

As Galuccio mentioned in a March 12th interview with Bloomberg: “We have a lot of elasticity: We drill a pad of four wells in four months […] so if we push a button today because we have more capex because the oil price is higher, in four months we’ll be delivering more to the world.” Vista is likely already trending well above its 140K barrel full year guidance.

And the production growth story has no reason to stop in 2026. A recent Goldman Sachs note on South American oil producers argues “Vista and YPF would be more likely to speed up production growth" in a sustained oil disruption compared to Brazil’s Petrobras or Colombia’s Ecopetrol due to less leverage on the balance sheet.

Goldman Sachs estimates: “fracking projects under development in Argentina and Saudi Arabia alone will meet roughly one-fifth of expected global oil demand growth through 2030.” Vista is the best pure play on the Argentina side of the equation.

Price Target: $100 in the next 3 months (4.5x 26 EV/EBIDTA). This does not require a significant re-rating but merely maintaining close to the current multiple in a period of significant production growth.

Risks:

1. Any headline that confirms a ceasefire or Strait of Hormuz opening will lead to an immediate 10-15% drop on sentiment alone. For the reasons outlined above, a drop like this should be viewed as a buying opportunity.

2. A large part of the discount is explained by political risk in Argentina. While Milei won the midterm election, his approval rating has fallen in recent months. YPF re-nationalized in 2012 and investors have painful memories of the situation. This will be more significant in 2027, the election year.

3. Investors believe a sustained period of $100+ oil would be bad for oil producers due to demand destruction. While this was a price level associated with demand declines historically, the bar is higher in 2026. The world is 56% less dependent on Oil for GDP growth since 1973 (Columbia SIPA). The 2008 oil high was $147 or $223 adjusted for inflation. Demand destruction is still a risk but requires higher prices than $110. The global economy still grew 3.4% in 2022 despite WTI averaging $100.

4. Vista is likely to reinvest back into CapEx as opposed to significant buybacks/returns to investors. With an ROIC of 19.7% (as of the end of 2025) and the immediate return on the Equinor acquisition, Vista’s management has demonstrated themselves to be capable capital allocators.

5. Vista energy is still limited in part by the ability of oil services to get to Argentina and build pipelines/exporting infrastructure. As Galuccio noted in the Bloomberg interview: “to accelerate, we need more capital, we need more oil service providers, more product providers, more competition.”

Disclosure: Author is long VIST shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research. Past results are not indicative of future results.

Today we’re making the argument that Tesla is a short, not just on valuation but with actionable catalysts within the next 6 months.

Tesla is one of the most successful stocks of the last decade and at every turn, Tesla has made a fool of shorts.

In 2017, Jim Chanos, who famously shorted Enron, said Tesla’s equity was “worth zero”.

In 2020, Tesla short sellers lost $40B betting against Tesla.

Yet Tesla has marched on to be the tenth most valuable company in the world, with a $1.5T market cap.

So why on earth, would I bother to bet against Tesla? Hasn’t enough brain power already been spent on this?

Every Tesla bear case has been a fundamentals argument… but Tesla has never traded on fundamentals. Here’s the catalyst that changes the narrative for Tesla.

SpaceX is aiming to IPO at $1.75T MC, bigger than Tesla today. Sources have told Reuters that “an IPO was likely in June.”

The SpaceX IPO will create an alternative way to bet on Elon at a company more centered on his long-time ambitions of Space and AI. The IPO will draw funds away from Tesla at a time Tesla is facing a revenue decline with a robotaxi program crashing 4x more often than human drivers.

Part 1: The Core Auto Business is Shrinking

Tesla is an outlier. It trades at one of the highest forward PEs of any company in the S&P 500. It’s priced for growth… only it’s not growing.

From last quarter’s earnings: revenue for 2025 was $94.8B vs $97.7B in 2024
(-3 YoY%), Q4 deliveries were down -16% YoY.

Tesla is struggling to grow revenue despite attempts by Tesla to discount. Tesla which once had an operating margin of 19% has seen that shrink to 5.7%.

The evidence is since earnings Tesla’s auto sales trend has gotten worse, not better.

China, domestic retail sales are down 45% YoY the lowest monthly figure since November 2022. China, Tesla faces increased competition in a highly sophisticated EV landscape. Xiaomi’s YU7 outsold Model Y 2:1 in January at the same $35,000 (USD) starting price.

In the US, January registrations are down 17% YoY. Tesla had pulled forward demand with the $7,500 EV tax credit last year.

In Europe, registrations are down 17% YoY, the 13th consecutive month of sales declines amid Musk’s entrance into politics and increased competition from BYD whose registrations were up 165% YoY in January. BYD is doubling its sales outlets in Europe. BYD’s Atto 2 Boost plug-in hybrid is now selling in Germany at a discounted price of €16,000.

This isn’t a cyclical trough. Tesla is losing on product in China (YU7), price in US (no tax credit), and brand in Europe (13 consecutive months of EU declines). Instead of reinvigorating its lineup, Tesla is retreating, ending Model S and X production.

Part 2: Robotaxi at Risk as Waymo Accelerates

One of the Tesla bull cases has always been that Tesla will be the leader in self-driving cars. Tesla launched its Robotaxi service in Austin in June 2025. Tesla currently has an active fleet of 45 vehicles. Elon claimed last year Tesla would be at 500. What’s the delay?

Tesla’s Robotaxis have already been in 14 car crashes! This is in a best case scenario: small fleet, in Tesla’s HQ, most of those miles with a safety driver in the car.

From Fortune’s coverage on this:

“Tesla states that the average American driver (of a Tesla) is involved in a minor collision every 229,000 miles, meaning in the estimated 800,000 miles Tesla’s vehicles drove in Austin, the average human would have been involved in roughly four crashes, compared to the 14 involving Tesla robotaxis.”

Tesla claims it will launch in 7 additional cities in the 1H of 2026. One of those cities is San Francisco. Yet in California, Tesla hasn’t applied for a permit. The permit requires Tesla logging 50,000 miles of autonomous data. If Tesla was confident in its Robotaxi at this point, that shouldn’t be an issue.

While Tesla might scale robotaxi in 1H 2026, Waymo is pulling ahead in reality. Waymo is operating a fleet of 3,000 vehicles and had 127 million miles without a driver in 2025.

Waymo just raised $16B of funding in February and is expanding to 20+ additional cities in 2026. Waymo is only going to get more prominent in the next six months while Tesla continues to work through its Robotaxi struggles.

Part 3: Tesla Needs More Capex for its Ambitions

Robots have similar dynamics to electric vehicles- Tesla will once again have to fend off Chinese competition, only this time with no head start. China is going “all-in” on Robotics with over 140 startups and more than $26B state funding per the WSJ. Robots are already in a price war, with Unitree selling commercial humanoids for $29,000: less than a car.

Like Elon’s other timelines, Optimus keeps slipping. Elon has already set this up on the Q4 earnings call: “We are replacing the S/X line in Fremont with a 1 million unit per year line of Optimus […] and there's 10,000 things that need to go right, it's, you know, it only takes one to be slow to lag that.”

While Tesla has a long-term ambition to scale to 1,000,000 Robotaxis and 1,000,000 Optimus robots (and built into Elon’s incentives), scaling production is also expensive. Tesla is doubling capex to $20B in 2026. Tesla is set to go FCF negative this year, for the first time since 2017.

The one bright spot in Tesla’s operating picture has been energy ($13B revenue up 26% YoY). This was not enough to offset the sales decline in auto.

While utility scale solar (Megapack) is a true opportunity tied to AI capex, the overall outlook for the energy segment is clouded by the end of the tax credit for residential solar (Powerwall). Tesla on Q4 earnings said it expects “margin compression from the increased low-cost competition, impacts to market from policy uncertainty, and the cost of tariffs.”

Part 4: The Elon Proxy Unwind

To recap so far:

1. Tesla’s valuation is inflated

2. Tesla’s fundamentals (revenue, margins) have been deteriorating

3. Tesla’s Robotaxi/Optimus behind schedule

None of this is new. Tesla bears have been making these arguments for years.

So what will be the catalyst that gets Tesla to close the gap? What makes the next six months a window of opportunity?

The Upcoming SpaceX IPO is a Negative Catalyst for Tesla Stock

SpaceX is likely to IPO this year. SpaceX entered a quiet period in December of last year. An internal company memo noted “it’s preparing for a possible public offering in 2026 that would be aimed at funding an “insane flight rate” for its developmental Starship rocket, artificial intelligence data centers in space and a base on the moon.”

Bloomberg reported last week “The Starbase, Texas-based firm expects to submit its draft IPO registration to the US Securities and Exchange Commission in March, the people said. Such a move would keep it on track for a June listing.”

Why is this bearish for Tesla? Let’s start with a simple question: if Elon Musk was not the CEO of Tesla, what market cap would it trade at tomorrow?

Tesla earned ~$4B in net income last year. Even at 125x trailing earnings, a high multiple, Tesla would be worth $500B. A significant portion of its Market Cap is tied to Elon’s time, talent, and even his broader business empire helping Tesla. Let’s call this gap- Elon’s “X-Factor”- as $1T of Tesla’s valuation .

SpaceX wants to IPO at $1.75T, bigger than Tesla. The multiple is lower, but SpaceX earned only ~$8B in profit last year. SpaceX’s valuation also requires Elon’s time, talent, and broader efforts to justify it. There’s only one Elon. Where will he place his focus?

What will happen when SpaceX IPOs? A simple mental model: some of the Elon premium shifts out of Tesla into SpaceX. Tesla’s Market Cap drops from $1.5T to $1.25T (-20%) while SpaceX goes from $1.5T to $1.75T.

Where will Elon spend his time? This was part of the rationale for Elon’s new pay package: “Mr. Musk’s other ventures, including SpaceX, xAI, Neuralink and The Boring Company, which may start to seem more attractive as these other companies continue to drive innovation within their respective industries, including AI.”

The higher SpaceX’s valuation at IPO, the more of Elon’s “X-factor” it absorbs. After all if Elon said he would only spend 10% of his time at SpaceX, it would be hard for SpaceX to justify an IPO valuation at $1.75T.

A direct link is index flows. The Nasdaq is considering letting in mega IPOs (such as SpaceX) into the index, only 15 days after IPO.

Tesla has benefitted from being a top 5 weight in the QQQ from a flow perspective.

Under the fast IPO rules, SpaceX could IPO at $1.5T, trade to $2.5T after 15 days, and get a 5% share of QQQ. The 5% has to come from somewhere. While this would impact all components in QQQ, the other dynamics present, Tesla will be the most affected.

Elon’s pay package will give him an extra 12% of Tesla stock ($1T) if Tesla goes to $8.5T, for a total of 25% ownership and 25% voting control. Sounds like plenty of motivation- except Elon owns 42% of SpaceX today and 79% of voting control! $1T is great but if SpaceX went to $8.5T instead, he’d make $3T!

This also creates dynamics for retail investors to shift flows from Tesla. A retail investor could be more bullish on Elon than Tesla (the car company). Retail investors love space, they love names with defense exposure, and they love AI. The logical answer is at the margin, a significant portion of Tesla’s retail base will switch names if SpaceX and Tesla have the same MC.

Over time, Elon is financially incentivized to prioritize the combined SpaceX/xAI entity, even if at Tesla’s expense. Take Optimus production. Elon has told xAI investors, xAI can “eventually power humanoid robots like Optimus.” If Tesla is selling the robotic body one-time for at/near-cost (5% margins) and xAI is receiving the monthly subscription revenue for the brain, that is not a great scenario for Tesla shareholders.

This is not an outlandish scenario. Elon in the past has moved employees between his various companies and done intra-company deals (Tesla sold $430M in Megapacks) to xAI. This is only about to get worse as SpaceX IPOs and Elon is incentivized to support SpaceX’s price through any lockup period.

Part 5: Price Target + Risks

Tesla has outperformed the Nasdaq by 15% over the last 6 months. There is not a good reason for this. Tesla ran up into good Q3 earnings (+12% rev YoY), but Q4 (-3% YoY) made it clear that Q3 involved significant demand pull-forward.

While Analysts are projecting 8.5% (consensus) revenue growth in 2026, this requires some unlikely assumptions, including ~3% auto sales growth, against last year, which included the expiring tax credit and Model S/X still in the lineup.

If auto sales fall 2% while keeping other analyst assumptions, Tesla’s revenue grows only 6% vs. 8.5% consensus.

Combined with the SpaceX catalyst, the base case is roughly 30% downside to $300 a share over the next 6 months. One would expect the repricing to begin when SpaceX files in March, ahead of the actual IPO date.

This is in line with where Tesla traded in the 1H of last year. It is becoming clear that the acceleration in the 2H of last year was tied to a tax credit pull forward and the stock price has yet to reflect the deterioration in fundamentals and narrative.

Risks and considerations:

• SpaceX IPO moved to 2027: removes immediate downside catalyst

• SpaceX IPO does not play out as expected: broader Musk sentiment improves into the IPO (aiding Tesla)

• Musk hits a timeline (either on Robotaxi scaling, Optimus production, etc.)

• Tesla surprises to the upside on energy

• How to short: Tesla’s cost to borrow is low as a highly liquid megacap. Puts are not recommended as Tesla has a high implied volatility. Neither is inverse ETFs (TSLQ). Inverse ETFs are trading instruments and are not designed to be held over time (decay is built-in).

Disclosure: Author is short TSLA shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

Today we’re covering Nokia (NYSE: $NOK). You may know Nokia as the Finnish tech giant that at one point made the world’s best selling phone.

Nokia today does not make any phones, having sold that business over a decade ago. Nokia instead operates in the telco infrastructure space, selling the radios, antennas, and signal processing that makes cell networks work. Nokia has thousands of patents for cell infrastructure, making it one of the most important companies worldwide for 5G infrastructure.

Nokia sells equipment to cell networks including T-Mobile (US), Telefonica (Europe), Softbank (Japan). The current industry expectation is that cell networks will slow spend on new equipment. Their 5G networks are mostly built out and 6G is years away with 6G standards not expected to be finalized until 2030.

Two near-term catalysts could disrupt the consensus narrative. An upcoming EU mandate will phase out existing Huawei equipment, boosting demand for 5G equipment in the short-term. Meanwhile, AI applications have the potential to dramatically increase data traffic on cellular networks, requiring 6G to be deployed ahead of schedule.

Nokia is well-positioned to be a leader in the 6G cycle with Nvidia backing ($1B investment), positioning towards US leadership, and a refocused business strategy under a fresh CEO.

Unlike other AI plays, Nokia is still an afterthought that is largely under-appreciated by the market and has plenty of room for earnings inflection.

Why is Nokia Relevant?

To compete in the cell infra space, there are significant barriers to entry. One has to have the patents for the radio frequencies and processing for effective cell equipment, in addition to network relationships and manufacturing.

Per congress,“To improve networks, operators purchase equipment from network equipment manufacturers. There are four major global network equipment makers: Chinese firm Huawei; Finnish firm Nokia; Swedish firm Ericsson; and Zhong Xing Telecommunication Equipment (ZTE), a Chinese firm that is partially state-owned.”

Huawei, Ericsson, and Nokia are near-parity for network solutions per Gartner.

While Huawei is the largest infrastructure player globally, Western countries are aligned on avoiding Huawei equipment due to past incidents of Chinese spying through cell networks. Due to this dynamic, Ericsson and Nokia are now an effective duopoly for western telco infrastructure, with the two companies roughly equal in scale and patents. This also creates significant pricing power during telco cycles.

Has Telco Capex Peaked?

There have been 3 major investment cycles for Telco infrastructure over the past 20 years, corresponding to the 3G, 4G, and 5G standards.

How do these standards come about? The International Telecommunications Union, a UN agency outlines targets for the next generation. 3GPP, an international consortium of industry representatives turns the targets into formal technical requirements. For 5G, commercial standards were agreed upon by 2020, kicking off a telco infra cycle.

In each cycle, network operators have spent more than the previous one.

Network operators have built out most of their 5G coverage. 5G is now seen as “past the peak, but it is not falling off a cliff,” per the Dell ‘Oro group. Cell networks including Verizon are now boasting “continued CapEx efficiency”

Nokia is highly sensitive to these cycle dynamics.

In the previous cycle, Nokia had peak TTM revenue of €25B and €0.8 EPS/share. Since then, revenue has declined to €19B with EPS in the €0.2 range.

This is also important from a valuation perspective. Nokia currently trades at 35x trailing earnings, which appears expensive until you realize this is through earnings. As each cell cycle is more capex intensive, Nokia can likely earn a peak €1 EPS, putting the stock at under 6x peak earnings potential.

So when does Capex inflect? Is it 5 years away? 2 years away? Deloitte is expecting telecoms through 2030 “to work to keep cutting costs, keep capital expenditures under control, monetize their past investments” as 5G matures. The consensus view is bearish on near-term capex.

Catalyst 1: EU Mandating Phase-Out of Chinese Equipment

On January 19th, the EU Commission proposed a new cybersecurity act (EU Cybersecurity Act 2.0). The initial proposal draft includes a full phase out of Chinese equipment.

While the EU initially recommended the removal of Huawei equipment in 2020, “only 13 of 27 [member countries] have so far acted on it, commission sources said. This marks the first time Brussels has attempted to make their removal mandatory.”

Under the proposal, EU countries would have 36 months to phase out their use of components provided by high-risk suppliers (China).

How big of a deal of this? Per Strand Consult, some of the largest EU countries were still buying Chinese 5G equipment in 2024. Countries like Germany, Italy, and Spain were buying 59%, 35%, and 32% of their 5G equipment from Chinese providers.

Now those same countries will have to tear out the existing Chinese equipment and replace it with Nokia or Ericsson.

This development is significant enough to disrupt the trajectory of the typical telco capex cycle and provide a material short-term boost to Nokia earnings.

Catalyst 2: AI Increasing Data Demand and Accelerating 6G Timeline

While the EU mandate provides near-term revenue improvements, AI is fundamentally changing the timeline and potential scale of the 6G investment cycle.

The Data Growth Debate

Cell network operators are currently expecting data growth to slow from the historical 30 percent to “10 percent across both fixed and mobile networks in the near future” (McKinsey). If data growth slows to 10%, there isn’t a huge incentive to launch a capex cycle before 2030.

Cell equipment providers believe the 30% growth continues: Ciena CEO Gary Smith stated “Traffic growth has been between 20% and 40% per year very consistently for the last two decades” and AI investments ultimately translate to data usage: "It has to come out of the data center and onto the network.” Consumers spent 48 billion hours in AI apps in 2025 per SensorTower data.

Why AI Will Drive Cellular Traffic
Some analysts claim AI, despite clear usage trends, won’t drive capex growth: “many gen AI use cases for both consumer and enterprise markets are not exactly bandwidth hogs: In 2025, they tend to be text-based (so small file sizes) and users may expect answers in seconds rather than milliseconds”

But AI use cases are rapidly evolving past simple text. Low-latency voice mode, AI-generated video, and file uploads are becoming more popular and are far more intensive on cell networks.

More importantly, AI is expanding beyond phones. As Nvidia’s Ronnie Vasishta testified to congress: “Traditionally, telecommunications networks delivered voice, data, and video. In the future, mobile networks will also be called upon to support a new kind of traffic—AI traffic. AI traffic will include the delivery of AI services to the edge, or inferencing at the edge. Mobile networks will support applications such as autonomous vehicles, smart glasses, generative AI services on phones or devices, holographic communication services, collaborative robots, and many more applications that we haven’t thought about yet”

Why 6G is required

While it would be easy to dismiss Vasishta’s statement as hype, Nvidia is legitimately worried about a potential cellular bottleneck in AI adoption. Last October, Nvidia invested $1B in Nokia at $6.01 USD/share to develop “the AI platform for 6G”

What is 6G ? At higher frequencies, it is possible to transmit data faster. As 5G moved up the frequency spectrum through massive radio antenna arrays, 6G will move even further up the spectrum with even larger antenna arrays.

At this scale the amount of data that has to be processed likely requires a GPU locally in the tower itself. Nvidia’s own data specs propose 6G will be a 20-100x improvement in data rate for the end user with speeds of up to 1 GB/s.

Nokia’s 6G Positioning and Accelerated Timeline

Nvidia has chosen Nokia as its core partner in 6G infrastructure before the 3GPP finalizes standards. Nvidia noted 6G standards are “quite far in the standardization process.” T-Mobile is already working with Nokia and Nvidia on a 6G trial implementation: “T-Mobile is one carrier that has recognized that and is working with Nokia and us on an AI-native implementation for 6G.”

It’s not just T-Mobile, with Softbank having extended an official partnership with Nokia.

With trials underway in 2026, and Nvidia pushing for U.S. 6G leadership with Nokia as its core network infra partner, we could see a 6G cycle develop in advance of official 6G standards, moving the capex cycle inflection from 2030-2031 to 2027-2028, pulling forward Nokia’s earnings inflection 2-3 years before consensus.

Improving Business Strategy and Focus

Beyond the two catalysts, new CEO Justin Hotard has been positioning the company to capitalize on the 6G cycle and AI while reducing non-core assets. Hotard, who became CEO in April 2025, brings a deep understanding of AI from his time as an executive VP at Hewlett Packard Enterprise where he delivered the “world’s first exascale supercomputer” for their HPC & AI lab.

Streamlining Operations

Hotard has reorganized Nokia under two core business units: Mobile Infrastructure (radio, software, and licensing) and Network Infrastructure (optical and broadband).

As part of this refocus, Nokia is divesting several non-core assets including Fixed Wireless Access CPE, Site Implementation and Outside Plant, Enterprise Campus Edge (currently in Cloud and Network Services), Microwave Radio (currently in Mobile Networks). These units generated EUR 0.9 billion in net sales with an operating loss of EUR 0.1 billion. Exiting these businesses will reduce operating expenses by €200 million by 2028.

US-Aligned Positioning

Perhaps most strategically, Hotard is positioning Nokia as the American-aligned infrastructure partner for the 6G era. Nokia is investing $500M in manufacturing and R&D across Texas, New Jersey, and Pennsylvania in addition to its already active broadband manufacturing in Wisconsin. The February 2025 acquisition of Infinera, an optical networking player with US manufacturing facilities, further solidifies this positioning and has already prompted analyst upgrades. Nokia also powers US military operations with private 5G infrastructure.

Combined with the Nvidia partnership, this US focus likely gives Nokia a competitive edge over Ericsson in winning American telecom business as the 6G cycle accelerates. As Hotard stated at Nokia’s Capital Markets Day: “As the trusted western provider of secure and advanced connectivity, our technology is powering the AI supercycle.” Hotard is uniquely positioned as an American CEO of a European company to win business on both sides of the Atlantic.

Financial Targets

Nokia is targeting €2.7-3.2 billion in comparable operating profit by 2028, up from €2.0 billion currently. The company expects 6-8% revenue growth in Network Infrastructure (with 10-12% growth in Optical and IP Networks specifically) and aims to expand Mobile Infrastructure gross margins to 48-50% by 2028. These targets reflect both the streamlined cost structure and Nokia’s positioning to capture growth in AI-driven infrastructure and the coming 6G cycle under Hotard’s technically focused leadership.

Price Target/Summary:

Nokia (NYSE: NOK) closed at $6.77 on Friday, January 25th. After initially running to $8+ following the Nvidia partnership announcement, the stock has pulled back and consolidated, creating an attractive entry point.

With peak earnings potential of €1+ per share, Nokia trades at under 6x normalized earnings power, a significant discount that reflects market skepticism about the telecom capex cycle. Nokia trades at 1.1x sales versus peer Ericsson at 1.3x, despite comparable positioning in the Western infrastructure duopoly and improving business momentum (Ericsson reported strong results last week).

The investment case rests on two distinct catalysts with different timeframes:

Near-Term (Next 3-6 months): The EU Cybersecurity Act 2.0, proposed January 19th, will move through the legislative process in 2026 and mandate removal of Chinese equipment from European networks. This will provide near-term revenue visibility and earnings support on forced replacement spending during what was expected to be a capex trough.

-Long-term (2-3 years): The Nvidia partnership and AI-driven data growth could accelerate 6G deployment from 2030-2031 to 2027-2028, pulling forward Nokia’s next major earnings cycle. With trials already underway and Nokia positioned as the US-aligned infrastructure partner, this represents the significant upside case.

Nokia reports Q4 earnings on January 29th. These catalysts support a price target of $9 (+30%) over the next 6-12 months as EU mandate clarity emerges, with further upside to $12 (+80%) as 6G deployment accelerates in 2027-2028.

Key risks:

• Execution: While Hotard has proven his ability to strike partnerships, cutting out $200M in non-core expenses is a different skillset

• Competitive dynamics with Ericsson: Ericsson is a formidable competitor. It’s not obvious who has the patent lead for 6G or if a third entrant is able to break through in the 6G environment (Samsung)

• Geopolitical factors: The EU could re-welcome Huawei/ZTE under Beijing pressure as they’ve done for BYD subsidies. However, it is in the EU’s best interest to support Nokia and Ericsson as tech leaders

• AI data use: AI could remain predominately a text-driven application, causing data use to grow below historical norms

Disclosure: Author is long NOK shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

Booz Allen is also in a 50% drawdown, its worst as a public company

What happened? The Department of Government Efficiency

98% of Booz Allen revenue is “from contracts where the end client was an agency or department of the U.S. government” (Form 10K). For most of Booz Allen’s history, this was the safest customer in the world: recurring contracts and a customer that rarely cuts budget or pulls back with the business cycle.

Booz Allen with just this one customer has grown revenue 5% and EPS almost 25% annually over the last 15 years, becoming a Fortune 500 in the process.

So why can it recover? And why now?

1. Sentiment lagging DOGE developments

2. Well positioned for expanded military tech spending

3. Last quarter guidance was intentionally sandbagged
   

   This Substack is reader-supported. To receive new posts and support my work, consider becoming a free or paid subscriber.

The DOGE Effect

DOGE came into the white house with a mandate to cut costs in the federal government. As a pure government contractor, Booz Allen was logical target for unnecessary contracts, Booz Allen revenue is down 8% YOY, and FCF down 29%

The damage is concentrated in Booz Allen’s civil portfolio. While Booz Allen’s speciality is national security/intelligence consulting, Booz Allen had built a side business of consulting for civil agencies.

In FY 2025, “Civil customers represented approximately 35% of our revenue” with clients including Departments of Veterans Affairs, Health and Human Services, Treasury, Labor, Homeland Security, Justice, Energy, Commerce, and Transportation”

Booz Allen anticipates “civil business revenue will decline in the low 20% range for the year.” Additionally, the civil contracts were higher margin than the military contracts, “putting downward pressure on margins in the near to medium term” (Q2 2026 earnings).

Now is the damage done?

DOGE as an entity no longer functionally exists.

Elon has returned his focus to Tesla/SpaceX and “U.S. President Donald Trump’s Department of Government Efficiency has disbanded with eight months left to its mandate.”

The end of DOGE is more than symbolic, it corresponds to a real shift in the administration’s priorities. Scott Kupor, director of the Office of Personnel told Reuters: “There is no target around reductions anymore.”

BAH confirmed the shift in the latest earnings call, stating “our civil business has stabilized and we have not experienced any contract actions [in the latest quarter].”

Unique National Security Moat

Now what about other 65% of the business? It’s classified.

While set up as a traditional management consultancy structure, Booz Allen accumulates talent with military experience and security clearances.

Obtaining a security clearance is a lengthy process, taking 140 days on average and 240 days for top-secret.

Booz Allen has a unique employee population. 72% of Booz Allen employees hold security clearances and 29% have military experience (Form 10K). The contracts Booz Allen bids for often require a security clearance with unique expertise such as cybersecurity, programming, etc. Booz Allen serves as an outlet for individuals with these unique backgrounds to serve the government outside of the federal payscale.

When Booz Allen is recruiting, they have very specific experience in mind.

When there is a new solicitation for an IT contract, Booz Allen has already typically accumulated the most relevant talent with a security clearance.

Booz Allen wins “56% of new contracts” it bids on. Once Booz Allen has won a contract, its positioning becomes even stickier, winning “92% of re-competed contracts” (previous contracts up for renewal) (Form 10K)

A bid requires “ a wide variety of skills, so companies pursuing prime contracts for those projects commonly assemble a group of subcontracting firms serving as teammates with specialized capabilities” (DOJ).

Because of the difficulty in assembling talent and submitting a bid, Booz Allen is often competing with one or zero qualified bids for a contract.

The talent pool means even when Booz Allen does not have the best tech, the best tech often partners with Booz Allen, whether it’s Palantir, Nvidia, or AWS. Booz Allen has the relationships and the know-how to navigate procurement.

Booz Allen is the #1 AI provider and a leading cybersecurity contractor for the federal govt.

Booz Allen is also deeply integrated into the functions of the DOD and other military agencies, with relationships dating back 80+ years.

As the CEO Rozanski put it: “Asking why Washington needs Booz Allen is like asking why the government doesn’t build its own tanks, missiles or satellites.”

Emerging Environment for Military Tech

Rozanksi said on the last call: “this is the most bifurcated environment I have seen in my decades with Booz Allen. Our civil and national security portfolios are experiencing completely different dynamics.” Last year, defense/intel grew 5%, not enough to offset civil declines. But the setup going forward looks different…

With civil stabilized, defense doesn’t need to grow dramatically to drive overall growth and the defense funding environment is improving. Additionally, Booz Allen can move resources (technical talent) towards defense bids if new civil opportunities do not emerge.

The military has learned from the Ukraine war a need for next-gen emerging war tech including drones, cybersecurity, AI, and space.

This is not just administration talk. The One Big Beautiful Bill Act added $150B in military spending over the next decade. This will require several new contracts in areas where Booz Allen has the right to win- and has won in the past.

Defense stocks are having a great year from RTX (+60%) to defense consulting peer CACI (+30%). Booz Allen’s civil struggles have prevented it from joining the rally.

The defense spending may go beyond the OBBBA- the White House’s Genesis Mission, Golden Dome, and battleships proposals could motivate a follow-up military funding bill.

While Booz Allen is likely to receive major contracts from this environment, it is not factored into current guidance.

Weak Guide Setting Up Trade Opportunity

Booz Allen gave a weak guide for the second half of 2025:

The guide assumes “current funding and procurement trends persist through fiscal year end and therefore on-contract and new award growth relative to bookings will remain below relative to years past “ (Q2 Earnings) Despite the last quarter having several major contract wins, including a $1.5B award for countering weapons of mass destruction, the guidance “is not really based on any significant new wins [...] there’s not any material things that need to happen for us to land in this range”

So nothing could go right and Booz Allen could still hit the guide. Rozanksi took advantage of his own weak guide to buy $2M of shares at ~$85/share following the last earnings call.

Two additional factors have been weighing on the stock since earnings, despite the improving regulatory environment.

1. EOY tax loss selling. In a year where the markets were up double-digits, Booz Allen has been easy sell for investors looking for tax savings. This will no longer be a headwind moving forward.

2. The CFO is leaving in February to be the CFO of the mobility spinoff from S&P global. While this raises investor worries, S&P Global is a reputable company and the spinoff is expected to be worth $12B+. A new CFO might want to reset expectations, but in this case they’ve already been reset.

If Booz Allen returns to growth mid next-year, with midterms potentially shifting the political environment for civil spending, investors may feel comfortable getting back in. By that time may have already regained much of this year’s losses.

Booz Allen is trading at 11X EV/Ebidta, 30% below its typical multiple in the last decade.

Or one can get in at the same price that the CEO bought at, with Booz Allen currently trading below $85/share.

Disclosure: Author is long BAH shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

TeraWulf announced $9.5 billion in contracted revenue last week and the stock dropped 20%. Oracle gave back its entire OpenAI rally. OpenAI’s struggles have doomed the entire data center complex- or so markets think. With Nvidia earnings Wednesday, investors are asking: is the AI trade over?

Today we’re going to answer three questions: 1. Is OpenAI in trouble? 2. Does the DC Complex need OpenAI to survive? 3. What’s the best way to “buy the dip”?

1. Is OpenAI in trouble?

OpenAI spent most of September making blockbuster deals. 10GW from Nvidia. 6GW from AMD. 4.5GW from Oracle. An incremental $250B in Azure credits from Microsoft.

These deals led to a question on Brad Gerstner’s podcast:
“How can a company with $13 billion in revenues make $1.4 trillion of spend commitments?” to which Altman fumbled “We could sell your shares or anybody else’s”

OpenAI’s CFO in a WSJ interview asked for a “government backstop.” To make things worse, Microsoft made a financial disclosure in its 10Q, that it “lost $4.1B in other income (expense) respectively of net losses from investments in OpenAI.”

As Altman disclosed, OpenAI “expects to end this year above $20 billion in annualized revenue run rate” which implies they had less than $5B in revenue last quarter. And since OpenAI has a 20% rev share with Microsoft, in Q3 they had $5B+ in compute costs at Microsoft against $4B in revenue (that’s negative gross margins?)

These spend commitments are coming at the same time, OpenAI is giving up traffic share to other models including Gemini.

Since OpenAI is losing market share and is operating at negative GM’s, this raises legitimate questions for how sustainable OpenAI’s business model is.

Answer: YES, OpenAI is in trouble

2. Does the Data Center Complex need OpenAI to survive?

The market has decided that OpenAI’s health is important to the health of many stocks involved in the data center.

$CRWV is down 50% in the past few weeks. $CRWV 5-year credit default swaps are getting bid to ridiculous levels:

Crypto turned neocloud $IREN is down 35%+

Former Yandex subsidy turned AI cloud provider $NBIS is down 40%

The market has decided if OpenAI is in trouble, the revenue projections from AI clouds have to be in question. There was a lot of revenue projected:

But is that simple? OpenAI could go bust in two years… and AI cloud revenues could move higher. The Netscape IPO famously marked the start of the dot-com bubble.

What’s often left out is Netscape underperformed after its IPO, eventually being sold to AOL in 1998. What happened to NetScape? Despite reaching 80% market share in early 1995, it bled users to competition including Microsoft’s Internet Explorer.

Did NetScape’s failure mean the internet was a fad? Would shorting the dot-com bubble based on NetScape’s 1997 economics have been a good idea?

It’s not obvious if OpenAI has the ideal business model for AI. OpenAI is optimizing for Meta-esque metrics from “time spent” to “weekly active users” and thinks ads will be an eventual path to profitability.

Anthropic believes corporate adoption can allow for higher gross margins. Satya Nadella believes that “grounding data” is the differentiator and will allow Microsoft to charge more for AI services. Zuckerberg believes it will simply come down to cash flow, and the company with the most cash to support AI will win.

When looking at OpenAI’s compute needs, it’s worth remembering that Sam Altman is buying compute under the assumption OpenAI will be a monopoly (80% of AI traffic share). What if that isn’t the case? In a competition scenario, the total pie for compute can be larger, even if the largest compute consumer is smaller.

AI models are continuing to get better with scaling laws. As detailed before Gemini 3 is likely to be a step change over current models. Grok 5, which will be the first model powered by a 1GW data center is likely to be as good as Gemini 3.

Other 1GW data centers including Meta’s Prometheus, Amazon’s New Carlisle , and Microsoft’s Fairwater will all come online next year, powering further model advancements.

Another angle to Altman’s commitments was to convince competitors that OpenAI would spend so much it doesn’t make sense to compete.

OpenAI’s struggles are an invitation to competitors. Maybe OpenAI only is spending $300B over the next 8 years, not $1.4T. That means it’s possible to catch up to OpenAI in the near-term on compute.

In the past weeks, we’ve continued to see more, not less AI data center announcements. Anthropic announced $50B in AI infrastructure. Google $40B.

We are also finding more constraints on the data center buildout from memory, to hard drives, to power (gas turbines). This means completed/existing data centers are likely to be more valuable assets. Any data centers fully constructed by the end of 2026 should not go empty, whether OpenAI succeeds or fails.

Answer: NO, data centers are likely to remain in a shortage whether OpenAI succeeds or not

3. What’s the best way to buy the dip?

Given the market is making the misconception that OpenAI’s struggles mean that the entire economics for data centers are in question, how can one play this opportunity?

$ORCL: Oracle is one of the most straightforward ideas. Oracle went from $230 to $340 on OpenAI driven-RPOs, only to give it all back.

The market has not only written off the entire OpenAI contract (in which years 1-2 should still be funded) but assumes Oracle will never be able to lease the compute they’re building. One can take advantage of this pessimism and go long Oracle.

$WULF is one of the crypto miners that have done the AI pivot. TeraWulf is trading down to $11 from $17 only a few weeks ago.

Both the Google $40B and the Anthropic $50B announcements involve TeraWulf to some degree. Google’s is $40B focused in Texas. Anthropic’s $50B data center spend is through “Fluidstack in Texas and New York.” TeraWulf owns 51% of the Abernathy campus in Texas, which is one of the mega-projects tied to both announcements.

This is a 25 year commitment backstopped by Google. TeraWulf’s 51% share represents $4.8B in contracted revenue when $WULF has a $4.6B MC. Yet WULF is down 20% because the market conflated OpenAI’s struggles with the entire AI buildout. If these announcements came out a month ago, Terawulf would have been up 30%.

These are just a few of the clearest tech opportunities. There are likely more dislocations in the market in the data center space.

Good luck on holiday shopping- these discounts might not last!

Addendum: NetPower Update
We had previously highlighted NetPower as a play on Texas natural gas demand, with its unique carbon capture technology and Project Permian buildout. NetPower in its Q3 earnings made a significant pivot, replacing its developmental Oxy-combustion technology with Entropy Inc.’s proven carbon capture system. While this changes the original thesis, the strategic rationale is sound:

• Timeline: 2028 first power (vs 2030-2031 for oxy-combustion)

• Capital efficiency: $75-90M equity from NPWR (vs $1.7B for original plan)

• Execution risk: Entropy tech (operating since 2022) vs first-of-a-kind

• Project financing: 50-70% debt vs 100% equity required

Danny Rice: “We can either keep our heads down and continue investing 100% of our capital to advance our oxy-combustion technology... or we can slow down that spending in order to free up some of our resources for near-term accretive opportunities. We strive to allocate our capital in a responsible manner that maximizes shareholder value.”

While “SPAC abandons breakthrough tech” is not an ideal headline, NetPower is trading at ~$250M EV net of cash, you’re paying almost nothing for two premium carbon capture sites, each with 300MW interconnects. Oxy has committed 30 MW + 100% CO2 offtake to the Permian phase 1 site with a second offtaker in advanced discussions for the remaining 30MW. The Midwest site is targeted at hyperscalers.

Disclosure: Author is long ORCL/WULF calls, and NPWR shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

After Mag 7 earnings season, one thing is clear: AI’s energy demand is accelerating faster than anyone projected. While nuclear names have rallied on this thesis, last week Google quietly announced something more immediately actionable: their first carbon capture and storage project using natural gas. This shift opens the door for $NPWR, a sub-$500M EV carbon capture pure play most investors have overlooked with asymmetric upside potential.

TL;DR: AI needs power ASAP. Nuclear takes too long. Solar is intermittent. Natural gas with carbon capture offers the goldilocks solution and $NPWR is the pure play.

AI’s Energy Demand: The Key Limiting Factor
Various components go into a data center: servers, racks, cooling, land, power, and of course GPUs. Many of these components are in shortage, but the most crucial is the power. In general, servers and data components are used to rising demand and secular growth. U.S. power generation is not. The US did not add power to the grid for most of the past 15 years.

Coal-fired plants have been replaced with less polluting natural gas, but net energy demand had remained stable. That’s about to change. Data centers are expected to add at least 325-580 terawatt hours to the US power grid through 2030.

Data centers are projected to go from 2% of US total power consumption (historical) to ~10%. This requires net new power plants. Mag 7 CEOs keep reaffirming their need for power:

Andy Jassy said on the Amazon Q3 earning’s call: “we’ve added 3.8 GW of power in the past 12 months. To put that in perspective, we’re now double the power capacity that AWS was in 2022, and we’re on track to double again by 2027”

CFO Susan Li said on the Meta Q3 earnings call: “our compute needs continue to expand meaningfully, including versus our own expectations last quarter […] CapEx dollar growth will be notably larger in 2026 than 2025”

Satya Nadella said on the B2G pod: “The biggest issue we have is not compute, it’s the power. It’s the ability to get the builds done fast enough close to power.”

Why There’s No Perfect Power Source
The quickest and cheapest way to spin up a new power plant is natural gas. Gas plants are roughly on par cost wise with renewables and solars (as IRA credits have been removed) and far cheaper compared to nuclear.

The LCOE on solar/wind is also often estimated too low, as when solar/wind conditions are not ideal, more of the energy grid needs to pick up the slack. For data centers which ideally are running 24/7, they prefer a steady state source of energy. The energy trilemma is visualized in this triangle:

Solar is green and fast to build, but intermittent. Nuclear is green and provides steady power, but takes a decade+ to construct. Natural gas hits the sweet spot, it’s fast to deploy and provides steady baseload power. The only drawback? Carbon emissions

Many of the big tech players including Microsoft, Google have made some pledges to be net zero by 2030. For Amazon, it’s 2040. These pledges are not that important in a Trump administration. But many of these power plants will come online in 2027 or later, with the bulk of their operating lifetime during the next administration.

Generally, it’s not good optics to be laying off tens of thousands of employees and polluting the environment. If AI continues to disrupt the labor market, a not so “out there” idea is a tax on AI carbon emissions, especially in a future democratic administration. But with natural gas being this cheap, hyperscalers can’t afford not to take advantage.

What if there was a way to still use natural gas but be effectively carbon-free? Enter carbon capture.

Google Adopts Carbon Capture
With AI data center announcements happening almost daily, an easy one to miss would have been Google last week announcing “their first carbon capture and storage project.” Google announced they “identified natural gas with carbon capture and storage (CCS) as a critical source of clean firm power”

“The Broadwing project is located at an industrial facility run by Archer Daniels Midland (ADM), which has nearly a decade of experience safely storing CO2 from ethanol production. A new power plant with over 400 MW of generating capacity will be built on site, and the CO2 it generates will be permanently stored in ADM’s adjacent EPA-approved Class VI sequestration facilities more than a mile underground.”

Google also attached this a helpful video illustrating how it will work:

CCS captures the emissions from natural gas combustion, condenses them, and stores them deep underground, turning a “dirty” fuel into 90%+ carbon-free power. Gas turbines with carbon capture are likely cheaper than wind or solar + battery storage while delivering comparable emissions reductions.

So why is carbon capture suddenly viable for hyperscalers? The “big beautiful bill” increased subsidies for carbon capture from $60 to $85 a ton, a 42% jump, at a time when most green technologies are seeing subsidies cut. As it turns out, the easiest way to win Republican support for green energy is to say “clean natural gas.” Combined with improved bonus depreciation on land, natural gas with carbon capture might have just become economically superior to renewables.

How do you play this? Google is using Archer Daniels Midland, which is too large of a company for this to move the needle. Luckily, NetPower ($NPWR) is a pure play natural gas carbon capture name, trading at sub-$500M EV.

NetPower: A Credible Carbon Capture Play
When you hear “zero-revenue energy SPAC” your brain auto-correlates to fraud (zinc battery storage didn’t work?) This isn’t your typical spac disaster.

NetPower is 40% owned by Occidental Energy, not the type of investor to throw millions on fake science. The Rice Brothers, who built Rice Energy, took Archaea Energy public (acquired at $26/share) are the driving force. Danny with his brother Tony Rice have been power players in the natural gas space. Danny Rice is on the board and Tony Rice is the CEO of EQT, the largest US natural gas producer. Baker Hughes is managing equipment and construction. The core technology, the Allam Cycle, developed by MIT and tested successfully in a 50MW proof of concept.

Even the conservative Joel Greenblatt was long at one point:

The idea of the Allam Cycle is to use as much of the CO2 back into the combustion process as possible, reducing the amount that needs to be stored underground.

While NetPower has almost revenue today, their core project is Project Permian, a 370MW carbon-neutral plant, located in Odessa, Texas.

What went wrong that sent the SPAC tumbling down to $2-$3 a share? Since initial projections, the cost for the plant rose to “between $1.7 and $2 billion, up from an initial cost range of $750 to $950 million and then $1 billion when it was announced in November 2023.” The demand for natural gas plants nationwide has caused a shortage in components, along with various issues one faces when scaling an innovative technology for the first time. While NetPower had started with a healthy $650M cash on hand, the project now is likely to use all the cash on the balance sheet. The sell-off was rational and that creates the opportunity.

Here’s what changed. Danny Rice has taken a more hands-on role at the management of the company, after having fired the COO and the CFO in April. NetPower has mitigated the delays with a plan to generate revenue before the cycle is finished:“in August released an updated strategy that would bring simple cycle natural gas turbines to the project site sooner - potentially in 2027 or 2028 and subsequently integrate those assets with the Net Power Cycle turbine, all on the same schedule announced in March.“We said why can’t we sequence to deploy gas turbines now and when the Net Power Cycle is ready, those same turbines can provide auxiliary power to the plant,” Rice said.

The other key catalyst is the increase in CC subsidies in the “big beautiful bill” along with improved bonus deprecation for land projects.

This helps to ensure the project’s economics will make sense once complete.

There is a final key catalyst remaining, a PPA (power purchase agreement) from a hyperscaler for the plant. Danny Rice believes this could happen soon “Most important, and unique to West Texas, data centers and hyperscalers are wondering how to get as much power as they can as quickly as they can. We’ve seen the last couple of years a shift in the market to as much natural gas generation as possible. It’s the only thing that can scale to the amount needed and offers affordability.”

A PPA would fill the funding gap needed to finish the plant without dilution and allow NetPower to raise funding in the credit markets. The hyperscalers are paying a lot for clean power. Terrawulf’s 200MW NY plant signed a 10Y PPA for $3.7B in contracted revenue. While NPWR, is not as close to a dense metro, a PPA for NPWR’s 370MW of carbon-free power could deliver $3-4B+ in contracted revenues against the current $430M enterprise value. The stock would likely double on a PPA announcement and this is not just wishful thinking.

Texas is becoming AI infrastructure ground zero, with 395 data centers operational or in development, only 2nd to Virginia as hyperscalers look for power close to the natural gas abundant in the permian.

West Texas in particular is seeing an a pickup in DataCenter development. NRG and LandBridge are constructing a 1.1GW Data center in Reeves county, within 100 miles from Odessa. Australian NeoCloud SharonAI is building a 400MW data center in Odessa. Microsoft recently contracted out to nScale a 240MW data center in Texas, to be constructed at an unknown location. OpenAI is building the StarGate complex in nearby Abeline Texas. 58 data centers are currently in development in the Greater W. Texas area.

Why West Texas? The region offers the cheapest natural gas in America (possibly, the world), cheap available land, and fiber connectivity. Perhaps even more important is what it lacks: state taxes, excessive regulations, or local govts who care about noise pollution. As Odessa Oil producer Kirk Edwards put it: “ “No other region can compete dollar for dollar with West Texas when it comes to fueling energy-intensive data centers with abundant, low-cost, clean-burning natural gas.” With all these data centers coming online in Odessa, the potential to contract 370MW of a carbon-free plant already in production starts to sound attractive, at least as a component of a larger project.

A PPA for Project Permian would accomplish three things: 1. solve then financing gap without dilution, 2. validate the Allam Cycle technology, and 3. open the door for NetPower to pursue additional projects in net new natural gas build outs or through retrofits of existing natural gas plants.

We see NetPower reaching $6-$8 on a PPA announcement, with potential tail outcomes to $10-$15 driven by commercial adoption. This is similar to outcomes that recently played out in the AI-adjacent energy space including Bloom Energy and Fluence.

Risks:
- While the project appears to be moving in the right direction timeline wise, a delay is always possible (as is the case with any first-of-its kind infrastructure project
-8Rivers, who exclusively licensed the core tech to NetPower has made insider sells. 8Rivers is structured as an LLC with investments across various clear energy technologies so this may be portfolio management instead of a lack of conviction
- the AI capex “bubble”/data center build out could crash before Project Permian comes live (for now current Mag 7 commentary suggests capex is accelerating, not slowing)

Next catalyst to watch: PPA announcement, likely within the next 2 quarters as West Texas data centers progress toward operation.

Disclosure: Author is long NPWR shares. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

This month, there’s a time-sensitive catalyst on a more familiar name, Google is expected to release Gemini 3 by the end of October. Based on early indications and rumors in the AI community, Gemini 3 may not only cement Google’s lead in the LLM wars but could be a significant step change in AI capabilities. Additional hype around Gemini 3 could send Google past the $300 mark by December.

From AI loser to AI winner

Reviewing the chart of Google over the past two years, one may notice a clear “character change.” From Nov 2024 to April 2025, Google was flat, despite growing revenue 10%+ and earnings 30%+, the same performance that investors are used to from this long-time compounder.

Investors were initially worried about Google’s profitable search business transitioning to the AI era. Major websites have seen visits fall YoY including Forbes (-55%), Business Insider (-38%), and the Washington Post (-33%). Apple’s Eddie Cue said in May Google’s safari searches had fallen YoY.

A few notable events have changed the narrative:

1. Google’s earnings in July showed search revenue up 12% YoY

2. Waymo started expanding outside of San Francisco. Waymo does 250K+ trips a week, in 15+ cities, and has started proving that Waymo’s are safer than human drivers

3. Google not only avoided a breakup in the antitrust case, but was allowed to keep most of its existing distribution deals. On the antitrust case, Google was actually saved by the potential AI threat to its business.

Since Nov 2024, Google has became the leader in the AI race and Gemini reached 400M+ monthly active users.

Google’s initial attempt to compete with ChatGPT- “Bard” failed to compel investors or gain traction. Google falling behind OpenAI in the model race always seemed odd. It’s the same kind of odd when Arch Manning was bad at football. You know Arch couldn’t possibly be as bad as he was playing earlier this season. There’s been some reporting on why Bard was so bad. Google’s AI talent was split in two (Brain/Deepmind). Google’s management might have been afraid of releasing something too good (innovators dilemma).

Whatever the reason for the initial struggle, Google has all of the advantages. Near-infinite cash flow, ($134B OCF), one of the deepest benches of AI researchers (even without Ilya), a custom ASIC (TPUs), tons of training (YouTube), distribution (Chrome/Search), hyperscaler know-how (Google Cloud).

Since Bard, Google combined all its AI teams into one, released the Gemini series of models. Even Sergey returned in pursuit of building a “god” model.

It’s been showing in the data. On the LLM arena, which uses blind tests of real users, Gemini 2.5 models are #1 in text, #1 vision, and #4 in WebDev. Keep in mind Gemini 2.5 pro was released in March and is beating out OpenAI and Anthropic models from Aug/Sep respectively. So Google has a 4-5 month lead here for Gemini 3.

The market is unsurprisingly giving Google a 73% chance of finishing the year with the best model Yet investors may still be surprised by just how good Gemini 3 is.

Gemini 3: The Next Leap Forward

Gemini 3 is expected to release in the next few weeks. An intern supposedly leaked that Oct 22nd is the target launch.

In the last week, Google has rolled out beta testing for Gemini 3. While we don’t have benchmarks, there are limited samples of Gemini 3 output. The mechanism is within AI studio, one out of every x prompts is being fed to an “advanced model.”

Some of the things users have built so far:

A virtual windows environment including a working snake and calculator game.

An SVG of an Xbox controller:

SVGs are a bit trickier than generating images as they require distinct shapes. Here’s Anthropic’s flagship model 4.5 sonnet trying the same thing.

or here’s a rough emulator of a Nintendo Switch, made in under a minute.

or even a simulation of a blackhole.

These are also mostly “one-shots” as one would need to wait 1/x prompts for Gemini 3 to appear again on the a/b test. Which means either (a) Gemini 3 is legitimately incredible, or (b) someone with access is really good at cherry-picking demos. My money’s on (a)

Google already had a ~4 month lead on the competition. With the release of Gemini 3, Google has at least kept its lead or even stretched its lead to 6 months +. This is terrifying for the competition. xAI had to go $12B in debt for the 1GW Colossus 2 data center. OpenAI has been busy doing financial gymnastics to fill its own funding gaps. Even Meta is going $26B into debt for a data center.

Unlike some of its competitors, Google can fund its model development directly from its $134B in operating cash flow, ahead of all other companies outside Microsoft. Unlike Meta or Microsoft, Google doesn’t have to give up 70% gross margins to Nvidia on chips. Unlike OpenAI, Google doesn’t need to promise some margin to Oracle on data centers. Google is able to get better ROI on its AI spend due its vertically integrated stack. While OpenAI will have its own custom chips eventually, that’s still two years away and OpenAI will be on a first-gen product which has its own issues.

Valuation and Risks

Google is trading at a forward PE of 25, below the 27 Forward PE of the Nasdaq 100 as a whole and within Google’s historical range of 20-30 times earnings. This means investors are on average neither bullish or bearish Google compared to the market. If Gemini 3 establishes clear AI leadership, multiple expansion should be in play.

If AI really is a 10 trillion market and Google is 1st, great! If AI falls apart and Anthropic/xAI go bust, search will be fine. One potential scenario, is 2-3 AI players reach profitability at scale and share the AI market, similar to today’s cloud market. The pure margins of Windows/Office licenses are likely better for Microsoft than Azure, but the sheer market size of cloud has made up for it and more. The same could be true when thinking about Google search vs. AI mix.

The bottom line is AI is the most important story in the market, Google is already leading today and is likely to lead even more after Gemini 3. If Google gets too far ahead, the “vibes” at Meta might not be so good.

Investors are likely to recognize if Google’s AI lead is strong enough, the next logical question is why shouldn’t Google be the most valuable company in the world?

Disclosure: Long GOOGL Nov 21 255 Calls. Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

The 2025 One Big Beautiful Bill Act included $24B in new funding for “next-gen integrated missile and defense systems”

This funding is part of the “Golden Dome” project, a gigantic iniativie to protect the US from missile threats. The $24B is just the initial funding. Trump’s 2026 budget proposal includes another $45B towards the project.

A solicitation has gone out for Golden Dome proposals. The proposals are due on October 10th under the Missile Defense Agency (MDA) SHIELD Multiple Award IDIQ Contract. To quote from the contract: “The anticipated stated maximum value is $151 Billion.” You heard that right: a $151B DoD contract is up for grabs.

BlackSky is positioned to be involved in at least one of the major proposals on Oct 10th and the stock could re-rate accordingly.

What even is a Golden Dome?

Unlike most projects with a $150B+ price tag, the idea for the Golden Dome is not from the pentagon, a think tank, or even the National Security Advisor. This was an original idea from Donald Trump.

During Trump’s 2024 campaign, he stated “In my next term, we will build a great Iron Dome very much like Israel has, but even better. I’m saying, why don’t we have that? We should have that too. We have a lot of hostile people out there. We have a lot of bad actors out there. We’re going to build the greatest dome of them all.”

He followed this up in January with an executive order calling for an “Iron Dome for America.”

Yes, Israel has the Iron Dome, which uses patriot missiles to intercept incoming rockets and has an 85-90% success rate.

Several factors make the Iron Dome hard to replicate for the US.

1. Israel is only 8,600 square miles. The United States is 3.5 million square miles. While the Iron Dome is able to work with only 10 Patriot Battery systems, the same level of coverage for the US would need ~4,000.

2. The threats from Gaza or Iran are typically improvised from PVC pipes or utilizing short-range ballistic missiles . A Golden Dome would presumably be geared to protect America against a first strike from China. China has hypersonic missiles, which would tear through an Iron Dome.

However implausible the idea is, there is now $24B in funding (and more to come) for proposals that can realize the “Golden Dome” concept.

The initial sketch is the “The Trump administration's flagship Golden Dome missile defense system will include four layers -- one satellite-based and three on land -- with 11 short-range batteries located across the continental U.S., Alaska and Hawaii.”

That “satellite-based component” of the Golden Dome is where it gets interesting for a certain US Satellite Operator: $BKSY.

BlackSky: Current Operating Business

Today, BlackSky competes with Maxar and Planet Labs in the satellite imagery market. These companies operate constellations of commercial satellites and provide real-time imagery for government and commercial clients.

Satellite imagery has improved and continues to reach higher-resolution. In terms of the next-gen products, Maxar’s Worldview 3 has a resolution of 31 cm. Planet Labs has a resolution of 50 cm. BlackSky’s Gen 3 has a resolution of 35 cm. What this means is through Maxar, BlackSky, or Planet Labs, one can see objects larger than 1-1.5 SqFt in real-time from space in real-time.

What would anyone need that level of detail for? It’s hard to imagine, but a “long-term strategic partner in the international defense of sector” gave BlackSky a $100M total contract through 2032.

The core business isn’t bad but it’s not great. BlackSky’ revenue is up 5% YoY at about $100M run-rate.

The general problem with the business is the core expenses (satellite launching) is upfront and the cash flows from imaging ( ~ 75% gross margin) are delayed. BlackSky took advantage of a move to $28 in July to issue strengthen its liquidity position.

This liquidity should cover operating expenses through the Golden Dome project timeline. At 775M EV, BlackSky trading at 7.4 EV/Sales or 39x operating cash flow.

Competitor Planet labs is at a pricier 12 EV/Sales or 41x operating FCF, but this premium is justified as Planet Labs has achieved better economics of scale and is FCF positive.

There are legitimate questions if BlackSky can continue to keep up in the race towards higher-resolution satellite imagery (requires Capex) as the #3 player in the market by revenue.

BlackSky revenue is expected to grow 34% in 2026, as they compete their 3.0 constellation rollout. The core business alone appears to already pricing in growth, trading at 7x sales when defense contractors more commonly trade in at 2.5x sales. This would not be a compelling opportunity if were not for the Golden Dome potential.

BlackSky: Positioning for the Golden Dome

Last November, BlackSky took full ownership in a company called LeoStella, previously a joint-venture with the French Thales Group. LeoStella is a satellite manufacturer and last fall it was selected for the “Hybrid Acquisition for Proliferated Low Earth Orbit, or HALO”, a select group of non-traditional defense contractors tasked with creating a new constellation of satellites to provide “scalability toward necessary performance for beyond line of sight targeting and advanced missile detection and tracking.”

While not directly tied to the Golden Dome project, the companies that the DoD selected are representative of the companies the DoD trusts with satellites at scale. This pool includes SpaceX and other space stocks including $ASTS, $FLY

BlackSky has historically had an exclusive focus on government revenue, helping to strengthen its ties to the DoD. Commercial is only 3% of BlackSky’s revenues, lower than competitors.

The expected requirements for a golden-dome system is an additional 36 LEO satellites and 30 MEO satellites for missile warning/tracking

AEI describes the function of the satellite constellations and their economics here:

Why are satellites essential for the Golden Dome? A North Korea hypersonic missile could reach the US in ~30 min. Detecting changes in North Korea’s missile posture using satellites and disarming before launch or shortly after is less risky than having to shoot down the missile when it’s already in air. The satellite constellation is the necessary first-layer of defense against a surprise launch.

So we know BlackSky is one of the 3-major satellite imagery operators and has ties to the DoD. What has the CEO Bran O’Toole said about potential Godlen Dome involvement?

As stated in the Q2 earnings on August 7th: “New commercial acquisition strategies in the spaceforce and emerging opportunities for programs like Golden Dome, offer long-term opportunities for BlackSky, which we believe we are well-positioned to capitalize on with our portfolio of Gen-3 satellites and manufacturing, AI, and real-time software capabilities.”

BlackSky is working on a large-satellite program called Aros complimentary to its Gen 3 satellite infrastructure. Aros is not so-secretely part of BlackSky’s Golden Dome bid: “BlackSky’s future, integrated constellation will be a first of its kind platform that combines very high-resolution broad area search with site monitoring to provide dynamic tipping and cueing for advanced maritime and Golden Dome type applications.”

While Aros will not be ready to launch until 2027, that lines up with the planned 2028 target for the first test of the Golden Dome system. As the Golden Dome $151B contract is multi-award, BlackSky could submit an independent bid. However, it could also join a prime contractor’s bid, such as supporting the space layer of Lockheed Martin’s bid.

Another option could be a partnership with Palantir, who has close ties to BlackSky and invested in BlackSky in 2021.

Given the SHIELD Multiple Award contract’s scale, even a 5 satellite contract could materially impact BlackSky’s valuation ($120M x 5 for procurement + $100M annual revenue for operations). If BlackSky is connected to a credible bid on the Golden Dome, the valuation is likely to catch up to Planet Labs’ current 12x Ev/sales.

Conclusion: BlackSky trades at a reasonable valuation for a defense contractor in growth sector, but has potential exposure to a $151B contract that could transform the company if they capture even a small portion. With the October 10th proposal deadline creating a near-term catalyst and management explicitly discussing Golden Dome opportunities, the setup offers asymmetric upside potential

Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

At the same time, Wix quietly made one of 2025’s best acquisitions. It bought Base44, an AI website builder for $80M in June. Base44 is already on track for $50M ARR and catching up to lovable in search trends. Wix has not seen revenue deceleration and is a rule of 40 company trading below 4x NTM sales.

The “Vibe-Coding” threat

The reason Wix is down 40% YTD is investors believe vibe-coding platforms such as Lovable are disrupting Wix’s business model. These platforms allow individuals to create a website in minutes just by entering a prompt.

As a demo, Lovable is impressive. I asked Lovable to design a website for a gym called “elite gym” and Lovable created a professional website in under a minute..

Founded only 8 months ago, Lovable is already valued at $1.8B, a quarter of Wix’s MC.

As put bluntly by an analyst on the Q2 earning’s call: “I guess, if you look at the share performance, the investors would seem to view that vibe coding replaces Wix, not create a new market and complementary. And for example, our agencies using this with Wix Studios or replace Wix studio. I just want to understand how you envision the ecosystem evolving.”

Wix’s CEO and founder, Avishai Abrahami responded: “Well, I think it's complementary, right? […] for websites, it's very hard just with a text interface to move things around and design them the way you want them. We can see already the tools that do vibe coding already starting to add a very weak visual editing elements. So obviously, the solution in the future will be a combination. I think the vibe coding has tremendous potential when it comes to building applications. So that way I think it's very interesting because a lot of the business logic is extremely hard, and that's where vibe coding shines […] vibe coding allows you to start very quickly and switch between designs […] and to build the logic of the applications. For website it’s a bit different. It’s very hard to prompt a full blown e-commerce package.”

While my Elite Gym site looks nice, it’s more of a shell. For getting the website to full functionality, including adding content libraries, payments and scheduling, I need to connect to a platform called Supabase, which uses PostegreSQL for API connections. This is all a bit too complicated for amateurs, but it is still a threat to the Wix studio product.

Wix’s response: the Base44 acquisition

In June, Wix acquired a AI website-building platform called Base44 for $80 million. While Base 44 only had $3M revenue at the time of acquisition, it has since gained considerable search momentum against Lovable.

In fact, Base44 is on track to hit $50M in ARR by the end of 2025. That means Wix got it for less than 2x ARR which is a bargain in today’s AI startup landscape.

As Base44 uses the same APIs, it’s not a surprise that the output is similar to Lovable.

Neither of these platforms have a huge technical moat as they are mainly AI wrappers. They will structure your prompt and send it to OpenAI or Anthropic. Once again, the website is a shell that needs to be connected to other services for payments, scheduling and so on.

However, Wix has a distribution advantage through its millions of existing customers along with a technical advantage as Wix has already built many of the other required services in-house.

These types of add-ons have been a key driver of Wix’s economics in the past few years. Much of Wix’s real margin comes not from website creation itself but with other services relating to business needs. This is a lane where they compete with Shopify.

You can set up an e-commerce business on Wix and Wix will handle payment, SEO, shipping etc. Or if you are a gym or makeup consultancy, you can get a website template along with scheduling clients and handling recurring payments. These mini-SMB platforms have helped drive Wix’s increasing revenue at the cohort level.

These services continue to remain in demand as websites have become complimentary with social media. Customers may discover a business through Instagram but they still want to pay on a website. Businesses don’t want to accept payments through instagram as a. it’s not professional and b. there’s quickbooks integration and other things you need to run a business competently. Wix continues to expand its functionality here, including a unified experience with PayPal and checking products.

It’s easy to imagine how in the vibe-coding model, Wix could leverage native connections to its own services including Wix pay and sell-them as add-ons to make the sites usable, continuing to drive high-margin revenue.

Wix is cheap, relatively speaking

If the 40% drop was not enough evidence Wix is on sale, a comparison to other SAAS peers makes it clear. Wix trades at below 4x NTM sales with 13% revenue growth expected next year. Unlike some of the other peers in its neighborhood, WIX has positive FCF and makes GAAP profit:

When narrowed to the set of SaaS companies with positive FCF, you can see Wix trades at one of the lowest FCF multiples of any SaaS company:

The underlying business remains strong

One may assume based on increased AI competition, that revenue growth is decelerating. That’s not the case…

What about earnings, maybe Wix keeps missing earnings? Nope. They’ve beaten top-line and bottom line four of the last 5 quarters, with the latest quarter being a 30% EPS beat.

In fact, Wix is trending above its rule of 40 guidance given three years ago:

In case you were wondering, this is not a “fake” rule of 40 metric from management. By Altimeter’s scoring, Wix.com had a 29% LTM FCF margin + 13% rev growth for a rule of 42.

Why Wix’s moat survives the AI era

Investors may be conceiving Wix’s moat is in drag-and-drop website building but that conception is a decade behind.

Yes, Wix and Squarespace’s initial promise was to be an out-of-the box tool for an individuals to create a websites without developers, but their real advantage is in the ongoing maintenance.

When one drags and drops an element onto a Wix page, Wix turns it to HTML on the back-end. Given the website-builder needed to be simple enough for a small-business owner with no coding experience to use, Wix intentionally limited the amount of features available within its HTML sandbox.

What Wix discovered by surprise was several SMB owners started hiring agencies to a build a website within Wix instead using HTML. Initially Wix was confused for the behavior: why would a business pay someone to make a website using Wix?

As it turns out, business owners hated having to go back to developers every time they wanted to tweak the structure or change the design of the website. When built on Wix instead, business owners could make updates on their own. Wix realized this was a sizable market and released Wix Studio in 2023, targeted for website agencies.

While being similar the same Wix platform in the front-end, Wix opened up the sandbox for developers to add custom visual, HTML, and metadata elements.
This agency segment which Wix calls “partners” is growing faster than Wix as a whole, with Q2 revenue up 24% YoY.

Like with developer created HTML sites, tweaking a vibe-coded site can also be a nightmare for SMBs. Integration with Wix’s existing tools is essential to make these AI sites malleable.

The truth is Wix has never had full product market fit: their products are too simple (regular Wix) or too complex (Wix Studio). Wix has never had a website creator that’s fully functionality and easy to use. Could platforms like base44 fill this gap?

While it’s still possible for Wix to be disrupted, Wix is adapting to the vibe-coding game faster than Lovable is adapting to the problems Wix has already solved. Wix already has payments, billing, e-commerce, and scheduling. Vibe coding technically is the easiest piece, requiring only basic API calls.

Wix is valued on a NTM 3.1x sales multiple compared to the median SaaS’s 4.7x and 11x FCF compared to the median SaaS 27x. If Base44 continues to show momentum and drive incremental revenue growth, Wix could re-rate to $200 a share driven by a fast-growing AI asset and decreased AI disruption risk.

Without getting political why does this bill matter?

1. Presidents typically get only one signature bill during their term. Why is that?

   1. Bills take at least six months to put together- the legislative process isn’t known for being quick

   2. Historically presidents are most popular during the first year of their term, their political capital decreases after that

   3. The president’s party on average loses 28 house seats and 4 senate seats during their term. Republicans only have an 8 seat advantage in the house. It’s hard to keep a trifecta after midterms

2. Markets are bad at pricing signature legislation

   1. Did you understand the implications of Medicare Advantage in the affordable care act? Congrats! You outperformed the market for a decade

      

   2. Or what about buying the largest construction equipment company after the largest infrastructure bill in 50 years was passed. That worked?

Why is the market bad at pricing this stuff in? Most market participants don’t have the time to read a 800 page bill. Most also don’t grasp how big a trillion dollars is (that’s 1,000 billion if you’re not sure).

These things also generally take a couple years to kick in, so it doesn’t go immediately to the bottom line.

3. This is a lot of money being spent

The OBBBA will add more to the debt than the all of Biden’s legislation combined

However, the White House argues if you consider the TJCA policies as baseline, the OBBBA will actually reduce the deficit by $395B over the next 10 years

Even if you accept this logic, the OBBBA will be a stimulus of at least $100B every year within Trump’s term, with the bulk of the cuts kicking in 2030 or later.

(N.B. Most cuts are medicaid-related, the most significant provision being a clause requiring in a month 80 hrs of work, 80 hours of community service, or a half-time education program to qualify for Medicaid)

This also means the current budget deficit is expected to go from 6% to 7.5%

First takeaway: You cannot be short the market heading into a 7.5% non-recessionary peacetime deficit

• Job numbers, tariffs, whatever, you are fighting gravity if you are going against the market in the next couple years

• Some of these tax breaks directly increase disposable income (standard deduction increase, deductions for seniors, no tax on tips, no tax on overtimes)

You can short bonds if you are feeling a bearish impulse, but do not short equities.

In general, this bill makes a higher for longer scenario more likely as you have the deficit contributions. The border spending should also lower the supply side of employment, decreasing the unemployment rate.

_______________________________________________

Now diving into the unique implications

1. Time to long defense stocks ($HII, $AVAV)

The big beautiful bill contains a boost to defense spending (beyond what was already in the budget)

Shipbuilding gets the largest boost with items including:

• $2,100,000,000 for development, procure-

  ment, and integration of purpose-built medium un-

  manned surface vessels;

• $492,000,000 for next-generation ship-

  building techniques;

• $1,470,000,000 for the implementation of a

  multi-ship amphibious warship contract;

• $1,300,000,000 for expansion of unmanned

  underwater vehicle production

There’s really only one shipbuilder of relevance, that’s $HII. HII is up 30% YTD but is flat since 2018 and its 18 PE is within its normal trading range for the past decade, so it goes in the bucket.

We also get a decent amount of “new tech defense funding.” This includes

• $1,000,000,000 for the expansion of the one- way attack unmanned aerial systems industrial base

• $250,000,000 for development and testing of directed energy capabilities by the Under Secretary for Research and Engineering

• $350,000,000 for development, production, and integration of non-kinetic counter-unmanned aerial systems programs;

We’ll include $AVAV, a drone stock that recently bought BlueHalo, a directed energy military contractor. It’s trading at a hefty 100+ PE, but is in all the right lanes.

2. No tax on tips ($PZZA)

The OBBBA introduces a deduction on tips- under this section for any taxable year up to $25,000. Specifically this is deduction is for:

A) The providing, delivering, or serving of food or beverages for consumption, if the tipping of employees delivering or serving food or beverages by customers is customary.

B) The providing of any of the following services to a customer or client if the tipping of employees providing such services is customary: (i) Barbering and hair care, (ii) Nail care, (iii) Esthetics, (IV) Body and spa treatments.

The long for this one would be $DASH. The gig economy is something someone can do part-time and now not be taxed on. DASH is already up 40% YTD. and trading at 300+ PE. So we’re going to go more artisanal with Papa Johns $PZZA at a 5 PE. Pizza delivery is already reliant on tips and now that can be tax free. There was also an acquisition offer from Apollo at $60 a month ago, so this could get bought out and you make 20% or you can wait for Todd Penegor (Wendy’s CEO from 2016 to 2024) to turn it around.

3. Welcome to the Trump Account ($HOOD)

The law creates a new type of investment account, the TRUMP ACCOUNT

It functions like an IRA where for kids: “no distribution will be allowed before the first day of the calendar year in which the account beneficiary attains age 18.”

The account can only invest in non-leveraged mutual funds or ETFs that charge no more than 0.1% fees and is composed of the S&P 500 index or any other index that is primarily US companies and is traded on futures (so tied to S&P, Dow or Nasdaq).

Any child born between Dec 31, 2024 and Jan 1, 2029 will be eligible to receive a $1,000 contribution from the Federal Government in a Trump account.

If you check who was pushing this policy, the Invest America roundtable with Trump included “Micheal Dell, Brad Gerstner of Altimeter Capital, Rene Haas of Arm Holdings, Parker Harris of Slack and Salesforce, William McDermott of ServiceNow, Dara Khosrowshahi of Uber, David Solomon of Goldman Sachs, Vladimir Tenev of Robinhood”

Unfortunately that means you have to long HOOD on this one as Tenev had the insider track and likely has a product already in development to be your Trump account brokerage. Every kid gets a Robinhood account with $1,000! Yay!

4. Taking a hammer to gambling ($IGT):

The bill includes the following language, in a last-minute addition:
"For purposes of losses from wagering transactions, the amount allowed as a deduction for any taxable year— shall be equal to 90 percent of the amount of such losses during such taxable year”

Imagine I go to the Casino twice this year and the first session, I win $5,000 in Poker and the second session I lose $5,000. I would normally owe no tax, but under this bill I would owe a tax on $500 ($4,500-$5,000)

The initial gut instinct would be to short $DKNG, but most DKNG customers are not doing large enough bets, to where they wouldn’t be covered by the standard deduction. This is the same audience that is doing 5-leg parlays on the regular. They aren’t the best at math.

The next idea would be going after the mega casinos (MGM, WYNN), but they have solid international exposure.

One potential long is $IGT. IGT recently sold its gambling operations to Apollo, making it a pure-play lottery operator (~60% US, 40% Italy). The new company is to pay a $3 special dividend on July 14th and has authorized $500M buyback. The lottery business is a better one than one would think and any headwinds to state gambling revenue may encourage a larger focus on lotteries.

5. Spaceports are airports ($RKLB)

Here’s a fun one: “SPACEPORTS ARE TREATED LIKE AIRPORTS. UNDER EXEMPT FACILITY BOND RULES.
(C) SPECIAL RULE FOR SPACEPORT GROUND LEASES.—Spaceport property located on land leased by a governmental unit from the United States shall not fail to be treated as owned by a governmental unit

(3) PUBLIC USE REQUIREMENT.—A facility shall not be required to be available for use by the general public to be treated as a spaceport for purposes of this section.”

This means Spaceports will be able to issue municipal bonds even while not being available for the general public. That’s cheap tax-free financing.

That’s good news for RKLB who operates a US Space Port (Launch Complex 2) and is currently constructing Launch Complex 3.

6. Orphan Drugs Get a Win ($ROIV)

Vivek who spent 10 days in DOGE was able to get this nice rule change in
Section 1192(e) of the Social Security Act by striking ''only one rare disease or condition'' and inserting ''one or more rare diseases or conditions''; and (B) by striking ''such disease or condition'' and inserting ''one or more such rare diseases or conditions''.

There is also a clause that moves the date of Medicare price negotiation eligibility from the original approval date to "the first day after the date of such licensure for which such biological product is not a biological product described in (A)".

This means for Orphan Drug companies, they can use the same drugs for multiple rare conditions without losing the Orphan drug status and they also get to keep their. Medicare pricing protection until they completely lose orphan designation - which could be indefinitely if they maintain rare disease indications.

This makes ROIV’s portfolio of orphan drugs much more valuable.

7. A Ton of Border Spending ($GEO, $CXW)

The OBBBA significantly boosts border spending including

• $47,550,000,000 for the border infrastructure/wall

• ~$6B for Border Personnel

• $10B/year for generally safeguarding the borders of the United States

The most impactful clause is “$45,000,000,000, for single adult alien detention capacity and family residential center capacity.”

That’s $45 BILLION, specifically for constructing new border detention facilities.

Private prison stocks GEO and CXW have been Trump trading sardines for a decade now, so it’s easy to forget they are real businesses. While GEO and CXW are up 75% and 60% since the election respectively, I don’t think the bulls in their wildest dreams had a $45B budget for detention facilities (to be spent by September 30, 2029).

The MCs of GEO and CXW are $3.5B and $2.3B respectively. So while it would be easy to say “priced in, I don’t think that’s remotely the case.

There’s more things in the bill, but those were just a few of the themes that stuck out
Here’s how I would position a “OBBBA” portfolio:

Currently I own one of these names $AVAV, but will be looking for key spots to add the others.

Not investment advice - just ideas that seemed brilliant at 2am. Do your own research.

In fact, the financials are better than the entire software landscape. AppLovin scores a rule of 66 - even with share-based compensation factored in. That’s higher than PLTR, PANW, and MNDY. This puts APP in the highest caliber of growth stocks.

At the same time APP is trading on more reasonable multiples than its high-growth peers: for reference APP trades at 25x 2027 PE.

Compare this to PLTR, which despite similar expected growth #’s is at a 135x 2027 PE

Also did you spot that revenue acceleration in 2024 for APP (17% → 43%)?
They did it with less employees than the year before!

The Applovin Debate

The key points of the story here:

1. Applovin was the primary advertising platform for games to advertise to other games

2. Applovin recently expanded into e-commerce ads within games

3. Somehow Applovin performance in those ads is ~90% as effective as Meta targeting

Even bears will admit to Applovin’s dominance on mobile game advertising. This is a market that Applovin won effectively from Unity (who were mismanaged). AppLovin was able to be early to the real-time bidding model for ads.

Bears are happy to admit Applovin’s dominance here as the mobile ad landscape isn’t glamorous. This starter pack is a good overview of the ads that are popular:

The “interactive elements” and small X’s serve to drive clicks which send users to the App Store. Much of the ads have more in common with the pop-ups that you would find on an illegal streaming site instead. An aggregator of these ads takes different skills than typical ad platforms.

After winning dominance in the mobile game-mobile game space, AppLovin set its sights on the e-commerce and general advertising space, surprising the industry.

At the core of Applovin’s reinvention and the e-commerce pivot is Applovin’s “AXON 2.0 Engine” released in early 2024. Advertisers using Axon 2.0 have reported performance almost as effective as their Meta campaigns. The weird part? No one really knows why the algorithm is so effective. It’s AI?

The weird part is Applovin is a 3rd party advertising platform. They don't have a captive audience like Google or Meta or a unique source of spending attribution like Amazon. No 3rd party advertising platform should be putting up numbers like this. So short-sellers are rightfully skeptical.

Axon 2.0: Did They Steal Meta’s Homework?

A theory from the short-sellers for the performance is AppLovin used Meta data to kickstart their e-commerce/general advertising business.

When rolling out the E-Commerce product, AppLovin required advertisers to spend $600k a month on Meta alongside AppLovin campaigns. As AppLovin has various integrations with Facebook audience networks (Facebook ad-targeting outside the apps), AppLovin could see the prices Meta was paying for ad space on its network. Combined with the $600K verified spend, AppLovin could attempt to reverse-engineer the Meta ad-targeting.

Short sellers allege this was the reason for initial effectiveness. Applovin could have used this Meta data as a base to add on third party data (cookies, website interactions) to build customer profiles and reduce their reliance on Meta over time.

This would be a smart strategy on AppLovin’s part and not too uncommon for the fast-moving digital ad industry:

But it would be a mistake to assume this is the only reason Axon 2.0 succeeded.

Axon 2.0: Advertisers Just Gave Their Data?

AppLovin acquired a platform “Adjust” in 2021 which measures performance across different ad platforms. Adjust is the second largest advertising data analysis and attribution platform in the world, behind the Israeli AppFlyer.

When you are switching to Adjust from another ad measurement platform, in the onboarding process, you upload a CSV with historical data for all your ad campaigns including device_Ids, networks, time stamps and ad groups.

This is an insanely rich source of data for a company like AppLovin to train their models on. So if they were getting enough CSV files, that could explain out-performance.

The Reality: Axon 2.0 Just Works

AppLovin’s CEO posted a blog post at the end of May giving a bit more detail on what goes into the Axon 2.0 Black Box. First thing: Facebook and Snapchat may show up on their pixels, but they aren’t directly going off an instagram ID

Even if a competitor tried to the same strategy, they would struggle to be effective. You need enough ad space for Meta to bid on. At the same time, AppLovin is building a moat over time through repeated interactions (which they have more of):

So AppLovin has mastered the art of ad targeting, but that’s only the beginning. The industry is shifting from ad platforms performing targeting only ad to taking on the entire advertising process.

Let’s Zoom Out for A Second

Initially in advertising there was a complex work flow. You had to plan an ad campaign. Then target the ads. Then buy the ads. Then optimize the results. Then report performance. AI-first ad platforms like Google, Meta, and AppLovin have systematically taken on more of these steps.

With generative AI, the platforms may even take on the ad creation themselves. Another fundamental change is the shift to attribution models. Originally, in the Google search ads model, you would use the last click, the blue link, to attribute the sale. But what if you clicked on that link after seeing an ad on TV? How do you know what really drove the sale?

Modern advertisers now track multiple interactions along the customer journey and assign attribution scores based on engagement and deep learning models. The key is making sure you can track the same customer across all these touchpoints over time.

With AI and better probabilistic models, advertising platforms are finally solving the attribution problem. Meta, Google, and AppLovin are leading this shift.

By tying advertising directly to actual sales outcomes rather than just clicks, this increases what advertisers are willing to pay per ad. They'll pay premiums for ads that demonstrably drive revenue, for outcomes over clicks. This is where AppLovin's background gives them an advantage, mobile games were forced to solve this attribution problem years ago. You can't rely on last-click when users might see an ad in one game, then download your game hours later. Mobile ad platforms like AppLovin pioneered user-attribution models out of necessity, giving them a head start in the outcome-based advertising world that's now emerging across all channels.

The winner in all this will be someone who can coordinate data across the different platforms from open web to CTV to social media.

AppLovin's management understands this and has assembled a full-stack ad platform through strategic acquisitions over the past few years.

In doing so, AppLovin is in the process of of building a vertical ecosystem that can compete with Meta and Google's integrated approach.

The missing piece is a walled garden of their own. AppLovin has submitted a bid for TikTok's US business, but that's not their only option. With a market cap of $120B, AppLovin could pursue other social platforms like Pinterest with smaller market caps to complete their ecosystem. Adding a social network would create another crucial touchpoint to improve ad effectiveness across their entire platform.

To sum it up:

1. AppLovin has developed a leading AI ad algorithm (Axon 2.0) that ranks behind only Meta and Google in effectiveness

2. The data sources and technical capabilities that built this algorithm create significant barriers for other competitors to replicate

3. They're building vertical integration spanning ad creation through deployment to reporting. This becomes more valuable as generative AI advances

4. Strategic acquisitions enabled horizontal expansion across ad markets from mobile games to web pixels and CTV

5. The market still isn’t sure if any of this is even real, creating opportunity.

Once you’ve figured this out, $120B MC is too evidently small for where this is going. For a company positioned to be the third major player in digital advertising during an era where AI is expanding the value of digital advertising. The 20% revenue growth estimates from analysts may prove to be too conservative.

Disclosure: I am long 150 shares of APP

LiDAR is a system that uses lasers to detect objects. A laser will keep going in a straight line until it hits an object. When it hits the object, the laser will actually bounce back to the source (at a lower magnitude). You can use this to figure out an object’s distance from the source:

A Waymo maps its surroundings with a bunch of lasers rotating very quickly. This allows it to create a map that will look something like this:

In doing so, Waymo can quickly detect nearby pedestrians and vehicles. Waymo uses its own proprietary sensors, but let’s say you wanted to play LiDAR in the public markets.

There’s three sensor-plays that aren’t penny stocks. HSAI is the global leader in this space, but it’s Chinese. AEVA is your next option, which has been meme’d 250%+ YTD based on an unnamed “top-10 passenger OEM” partnership. OUST is operating with named partners, including a top-10 OEM of its own (Hyundai) and a larger-scale business including industrial partners such as John Deere and Komatsu.

OUST went public via SPAC in 2021 at roughly $2B, crashed 80%, but has been firming up technically, including a recent breakout. So what’s the story and is there actual fundamental value beyond meme-potential?

Ouster Background
Ouster was founded in 2015 by a group of Stanford mechanical engineers, including founder-CEO Angus Pacala. In 2017, they raised a $27M series A, originally a 64-channel sensor that was 85% cheaper than its main competitor. Ouster took the opportunity to go public during the SPAC-boom in 2021 and used the proceeds to acquire Sense Photonics which also had an automotive focus.

In 2022, Ouster merged with Velodyne LiDAR. Velodyne was one of the first companies in the LiDAR space, tracing its history to a 2005 DARPA challenge. Velodyne had also gone public via SPAC, but had fallen 67% from IPO at the time vs. Ouster which had fallen 50%.

Business Momentum
Ouster has begun to gain momentum with a series of partnerships:

Announced within 2025:

1. Ouster to power Komatsu mining equipment

2. Ouster to power John Deere autonomous tractors

3. Ouster to power Chatanooga’s Smart Traffic solution

An already existing partnership of note is Serve Robotics (SERV), which uses Ouster as its LiDAR provider.

Business Case
On paper, the economics do not look great today: however, there are starting to be signs of operating leverage. Revenue has grown from $83M to $118M from 2023 to 2024 while operating expenses went down from $187M to $148M. The business is operating at a healthy 40% gross margin. Ouster has pledged to maintain its 40% gross margin while keeping R&D and operating expense flat, allowing it to potentially grow its way out of its current loss structure.

What it’s really lacking is scale. If you think about 4,800 sensors on a quarterly basis, that’s not an industrial-scale business. Despite this, Ouster is still in an industry leadership position. As explained in a recent CEO interview: Ouster is the “largest provider of LiDAR technology in the Western world.” Ouster also has a moat in their custom silicon Chronos chip platform. However, to date Ouster has mainly provided sensors to R&D divisions and universities and has not been asked to deliver sensors at scale.

Even Waymo, the flagship LiDAR industry solution, isn’t operating at scale yet, with around 2000 vehicles worldwide. While not directly partnered with Waymo, Ouster does power the Google Maps street-view vehicles, which will be needed for continued Waymo expansion to new cities.

To value this, one could begin by penciling in growth in the individual business divisions: Komatsu/John Deere partnerships will drive incremental volume, a city larger than Chattanooga decides to go with a pedestrian-safe traffic grid (the state of Utah recently allocated a grant), SERV has done a few offerings to double its cash position to $200M so will likely deploy more robots. They have two robotaxi customers, Motional - a subsidiary of Hyundai and May Mobility- a smaller player with Toyota investments. You could have some upside from humanoid robotics, with platforms such as Unitree incorporating a small LiDAR sensor into their infrastructure.

All in all, that’s enough to keep Ouster growing at 30% moving forward. But you’re also dealing with high share dilution (~5% QoQ) and being years away from real profits. The real interesting investment case which the market hasn’t caught onto yet is defense.

The Next-Gen Defense Platform
In May, the army announced an “Army Transformation Initiative,” its new comprehensive strategy to create a “more lethal force.”
While still in its early days, the basic principles are
1. We live in a new age of warfare driven by autonomous warfare (as shown by the effectiveness of Drones in the Ukraine war)
2. The army should cancel current procurement programs for “outdated crewed attack aircraft such as the AH-64D, excess ground vehicles like the HMMWV”
3. The army should redeploy the budget from those canceled programs into “long-range missiles and modernized UAS into formations”

The implications of this strategy are still coming into view. Meta and Anduril recently announced a joint partnership project developing systems that provide “enhanced perception and intuitive control of autonomous platforms."

LiDAR is a core and undervalued component of next-gen warfare. It’s also one where the US is currently lagging behind China. As stated in a congressional report:

“U.S. and PRC forces are among the militaries using LiDAR to support autonomous navigation capabilities for uncrewed ground and aerial vehicles (UxV). LiDAR's precision and speed enable military-specific use. UxV equipped with LiDAR could be used to conduct battle damage assessment—an estimate of the physical and functional damage produced by an attack—removing the need for military personnel to be physically present and exposed on the battlefield. LiDAR-equipped UxV could be used to survey structural damage after natural disasters and to collect environmental data in remote or dangerous locations. The U.S. military has explored using LiDAR to identify and determine the depth of littoral sea mines and to conduct atmospheric monitoring to predict laser weapons' effectiveness. The U.S. Army says potential uses include "platform target identification, aim point selection, range instrumentation support, weapon defenses, and mapping."

There are several draft bills in congress to ban US defense departments from buying Chinese LiDAR solutions. Beyond obvious reasons, there was a specific incident where a “Chinese lidar manufacturer planned to transmit data culled from Estonian cars back to China.” If the US was to focus on a next-gen LiDAR strategy, it is unlikely they would turn to Chinese HESAI but instead pick the largest-US based LiDAR sensor manufacturer (Ouster).

While UAV drones can use infrared cameras for general strikes, LiDAR can come into play for other contexts. As explained by Northrop Grumman “A drone fitted with LIDAR technology could do all sorts of survey work, but it could also do something far more sensitive: fly into a dangerous urban environment, even a building interior inaccessible to GPS data, and give American troops a quick, real-time look at the layout, obstacles, and hazards.

In other words, the pivot to a Drone-First military will also become a pivot to a LiDAR-first military. There hasn’t yet been a provider selected at scale to fulfill this need. By scale alone, Ouster is likely on track to do this. Forterra, a $75M series B defense contractor partnered with Oshkosh, already uses Ouster sensors for autonomous military vehicles.

Ouster maintains a commitment to “not sell our products to adversaries of the United States or repressive regimes.” Ouster’s Gemini Software platform, a surveillance platform meant for “perimeter security, transportation safety systems, traffic signal control, retail and customer analytics, and logistics” also seems tailored-made to be a dual use software for border surveillance and military operations.

Bottom Line: At $500M enterprise value, you're getting the dominant Western LiDAR provider at the inflection point of multiple demand drivers. I believe LiDAR is likely to become more important in the market’s view than previously as the new defense strategy and “Physical AI” continue to gain relevance in the market narratives. The economics are not as bad as they seem with 40% gross margin on units sold, just operating at sub-scale. The upside allows for a speculative position.

What’s going on with the economy anyways?

Will we enter a recession? Will the consumer break? I am not an economist, but those are questions the 19 members of the FOMC study in detail. The median FOMC member does not expect the economy to enter recession, but sees an environment with lower GDP growth, slightly higher inflation, and slightly higher unemployment. And while it can be described as “mini-stagflation” scenario, it’s not the same as the 1970’s. In 1974-1975, we had 10%+ CPI and negative real GDP.

A 3% CPI and a 1.5% real GDP, that the more pessimistic members of the FOMC may assign, is not a recession or even necessarily a “bad economy”

As Jerome Powell put it “it's not that inflation is really high, it's not that unemployment is high. It's none of those things. We have, you know, 4.1 percent unemployment, we've got 2 percent growth, and you know, it's a pretty good economy. But, people are unhappy because of the price level. And I do, we completely understand and accept that.”

In other words, we could be entering a 2022-style “Vibecession” but not a true economic downturn. One reason: tariffs “apply only to wholesale costs at the border, and US goods imports make up around 12% of GDP.” In most industries, goods are sold retail at a “keystone markup” of 2x the wholesale costs. When you see a headline “20% tariff” you can usually divide that by 2 to get the actual price increase a consumer may face. At the margin, goods still get more expensive and consumers buy less. The FOMC is pricing this in by predicting a -.5% decline from current GDP Growth levels into the end of the year.

The other dynamic is government spending. While DOGE has made a lot of headlines about cutting billions in federal spending, real-time tracking by the Hamilton Project shows federal spending up 5% YoY (~$100B)

While spending on US AID is down ~$2B YoY, this is significant compared to things like Medicare prescription drugs (~up $11B YoY).

While congress could change this with major spending cuts, current proposals (tax cuts) and Trump’s general rhetoric (using proceeds from DOGE or tariffs) to not repay the debt, but to distribute as consumer stimulus) suggests the deficit will not be meaningfully cut in 2025.

Why consumer lending can go higher?

Having outlined the Fed’s expectations and examining the tariff and DOGE impacts, here is a plausible economic scenario for 2025:

Products get more expensive, growth slows but is still positive YoY. The Fed cuts 2-3 times. The consumer does not pull back significantly on spending, but feels pinched at the margin.

In this scenario, the consumer may turn to consumer lending to fill the gap. This could include credit cards, BNPL (Affirm, Klarna), or direct lending.

How can this work? Consumers are still less levered on their household balance sheets than they were before Covid. The stimulus checks helped to reset consumer’s debt levels. As interest rates decline, consumer debt becomes more affordable. So consumers, if pinched have room to take on more debt, instead of pulling back spending. There’s also some generational factors here: Gen Z has bad spending habits.

So there’s the thesis, consumer spending slows but doesn’t break as consumers have an available well of consumer credit they can tap. This will break eventually (2026? 2027?)- but what if there was a business that was profiting on this trend, without taking on credit risk? Introducing LendingTree.

What is LendingTree?

LendingTree ($TREE) is a recognizable business for most consumers, but as a public equity is under-covered. It’s not a new business, having first gone public during the dot-com days. Its market cap is under $1 billion and its stock is down 60%+ in the past 5 years. This is what creates the opportunity, as LendingTree has been through quite a turnaround in the past the two years.

At its core, LendingTree is a marketing platform, having more in common with Nerdwallet than a Rocket Mortgage. LendingTree offers a marketplace for consumers to compare providers across their three segments: home lending, insurance, and personal loans (inc. SMB). LendingTree has 430 “Network Partners” who provide the loans. Consumers will fill out their basic financial information and LendingTree will get paid for matching the consumer to a network partner.

How are these three segments doing? Home grew revenues 44% YoY and profit 44% YoY. Insurance grew 188% YoY and profit 90% YoY. Consumer is LendingTree’s most profitable segment (51% gross margins), but has lagged with segment profit down 2% Y/Y.

For 2025, LendingTree gave the following guidance: Our full-year 2025 outlook assumes double-digit revenue growth in both the Home and Consumer segments, with more modest Insurance segment growth following a record year.

Insurance

This higher 10-year environment has been gold for insurers and insurance companies have found they now have extra dollars for marketing spend that they didn’t have previously. Insurance premiums have also gone meaningfully higher, making consumers more apt to compare options.

We don’t need to speculate who these insurers are. Per the 10K: “For the year ended December 31, 2024, two Network Partners, Progressive Casualty Insurance and Allstate Insurance Company, accounted for 22% and 11%, respectively, of total consolidated revenue”

Management expects moderate growth in this segment in 2025 per their Q4 earnings:
"Almost all of our carriers are still in a growth position with us, and they've made that pretty clear to us [..] So definitely, we still expect insurance to be a really good story in '25. As the year goes on, I think you're going to see that growth moderating.”

Progressive and Allstate have presumably communicated higher spend, but this segment does have tough comps (as any segment with 188% YoY growth would.)

Home

LendingTree offers comparisons of mortgage loans and home equity lines of credit. The double-digit guide for 2025 does not take into account lower interest rates:
Per LendingTree’s CFO: “Jed, I'll just add on to that. As it relates to the guide for home, we're not contemplating any material reduction in rates. We're kind of assuming basically the rate environment that we are in today. What we're basically assuming in the guide is a continuation of the strength in home equity”

While mortgages are down, home equity lines of credit have been popular in this environment, offering later rates than credit cards or direct loans, but requiring valuable collateral (your home).

The COO added: “from a consumer standpoint, I think it's also the scenario of with very little buying and selling of homes and with the equity that the consumers are sitting on, the homeowners are sitting on specifically, just keep going up and up, they're going to want to to remodels. They're going to want to consolidate debt. They're going to want to go on vacations. And so it's like they're -- it's still the best interest rate product compared to all other lending products”

Consumer

On consumer, management expects double digit growth. By sub-segments, credit card is “rough,” SMB is on a “strong growth trajectory”, "Personal loans, that one has been growing well for us, grew 20% […] “that category is like -- the credit boxes are still pretty tight. But as long as you can find the consumers that meet those credit boxes, they're -- our clients are very actively telling us, "Bring them on."

The bottom line is: “interest rates are still a little high, but you've got a lot of consumers out there that are looking for lending options.” The guide again does not take into account lower rates, but if rates do go lower this is a segment that can pop.

Other factors

LendingTree has benefited from shifts in the insurance market, but why has their growth turned around in other segments? One reason was a shift in Google SEO:

“And on the SEO and organic traffic, call it, 15% to 20% of our traffic is organic. SEO is a piece of that. The rest would obviously be people hearing us and typing us in, et cetera […] in terms of the Google algorithm change, what I've noticed Google doing, and I, quite frankly, agree with it, is they are giving more credit to high-quality, unique content that really aids the consumer. So they did things like shutting down the"marketplaces" in -- of newspaper sites because they basically were leveraging their credibility on the news to go sell jackhammers and screws or whatever and mortgages. […] LendingTree has always been unabashedly in the paid marketing camp, and I've wanted our SEO side to work since 1999. I would say for the first time, it's really gripping.”

Google decided that LendingTree had higher credibility than Forbes Advisor, CNBC Select, and the other media sites that advertise loans, insurance etc.

Valuing LendingTree:

LendingTree grew revenue 12% YoY and is guiding to ~8-9% YoY growth this year, with this year’s cut not pricing in rate cuts. But is it profitable?

It wasn’t on a TTM basis, but this is misleading: Q3, was mostly a goodwill impairment towards LendingTree’s 2020 $80 million investment in Stash. LendingTree is not expecting any new investments in the near future. Another question is why revenue has doubled YoY but EPS has halved from Q423. This is also misleading:

The Q3 23 had a large gain associated with an Apollo debt. The latest Q4, with the $.55 GAAP EPS and few one-time adjustments, its a more accurate reflection of the earnings power moving forward.

As revenue increases 8+% next year and they continue getting more fit (notice the $1.5M reduction in SBC from 2024 to 2025), you can see a path to $0.9 - $1.0 GAAP EPS. That’s roughly where the analyst consensus is at:

You’re looking at NTM 15 PE, 9 EV/EBIDTA, and 1.12/Revenue, multiples generally below market averages. The current growth profile should argue for a higher valuation. Debt is high but at 5x Debt/Ebidta, but improving.

These multiples would make more sense if valuing a financials business. While this is exposed to similar cycles as financials, the business model itself does not involve consumer credit risk. The CEO (who is also the founder), has been through multiple cycles (dot-com crash, great financial crisis) and has been successfully de-levering the business in the past few years.

As LendingTree starts to screen better in the next year (debt metrics, GAAP earnings, revenue growth) and if the consumer segment turns around, this could see increased institutional adoption and rerate. With only a 10M share float, it doesn’t take much for this to move fast. You’re also starting to have some insider buys, most recently the COO buying 5000 shares in the $41-42 range.

In the last quarter, DraftKings handle (total $ bets on the platform) was $14.90B, up from $12.94B the year prior. This translated to $825M in Sportbook revenue with DraftKings “cut” being 5.5%

While DraftKings is not profitable, the most obvious reason is the sales/marketing expense, making 2/3rds of the cost of revenue.

While this would usually be too much marketing spend for a company, DraftKings and FanDuel have done a strategy of “blitzing” a new market with promotions and ads whenever a sports betting first becomes legal. Investors anticipate DraftKings to take down marketing spend as it once it obtains sufficient market share in most states.

Sports Betting Background

The Supreme Court struck down the Professional and Amateur Sports Protection Act in 2018, paving the way for Sportsbooks to become legal in the US states outside Nevada. States quickly began to legalize it- one reason was to offset declining revenues from gas taxes (higher fuel efficiency/EVs)

Advancements in technology meant bets could now be done on your phone, making it more accessible than ever.

Benefits from Imperfect Competition

While there were plenty of companies that wanted to get into the market, most states legalizing sports betting wanted this to be done by trusted casinos and vendors, not your local bookie. In practice, this meant most states limited their market to a # of Sportbooks, such as 20 in Ohio or 9 in New York.

For example: BetFred, a well-established British sports book doesn’t operate in many of the states DraftKings operates in. This also means that once a market is established and the vendors are selected, DraftKings and FanDuel only need to win against the initial competition, and not worry about new entrants, making the spend on marketing dollars attractive.

The business has started taken on Duopoly characteristics, with FanDuel/Draftkings taking share from the other competitors, who had often had worse digital experiences, were less aggressive with advertising, or were afraid of cannibalizing their casino business. Most investors imagine this emerging duopoly structure turning into significant profits. DraftKings currently has a $26B MC.

Enter the Prediction Markets

Prediction Markets are sites where users can wager on the outcomes of future events, such as who will visit the White House first: Dana White or Zelensky? These started to become popular in the last decade.

It was not clear from the text of the law if these sites constituted gambling (illegal in many states) or were more akin to a futures market (will Oil trade above $90 in August). The CFTC was given the power under Dodd-Frank, “the ability to view, and prohibit, event contracts that it determines are contrary to the public interest if, and only if, they involve specific activities, including ‘activity that is unlawful under any Federal or State law’ and “gaming.” The venture-backed Kalshi was able to obtain a conditional waiver from the CFTC to offer prediction markets in 2020. The CFTC preferred having oversight vs. the bets happening on its unlicensed crypto-based competition such as Polymarket.

The 2024 election boosted the popularity of prediction markets, as users bet on the outcome both of the election and the events leading up to it (who will be Trump’s VP?) (will Biden drop out?) This raised a relevant question for the CFTC: shouldn’t predicting election outcomes fall under “gaming” and be prohibited? The CFTC sent an order to Kalshi block election bets, arguing that gaming and gambling were synonymous and election predictions were a form of gambling:

Kalshi argued gaming required in its defintion playing a game - blackjack, poker, roulette. The court agreed:

This decision effectively stripped the CFTC of its ability to regulate prediction markets. Prediction markets still have to specify concrete outcomes under the current CFTC rules. For example, on this prediction market for the “best AI” the outcome has to be unambiguous (ranking on Chatbot Arena on Feb. 28th)

Sports are future events with unambiguous outcomes, so Kalshi and other platforms decided predicting sports was legal, akin to a futures contract. Various platforms such as Crypto.com and Robinhood began to offer bets on the Super Bowl:

The CFTC asked Robinhood to hit the brakes, noting concerns about sports “predictions” being offered alongside equity investments, but recent developments have made it much more likely that CFTC will embrace sports contracts, placing the margin profile of DraftKings in potential jeopardy.

The White House Embraces Kalshi/Prediction Markets

In one of the more likely displays of corruption in the first month of Trump’s presidency: Donald Trump Jr. joined Kalshi as a strategic advisor.

Following this move, Trump named as the head of the CFTC, Brian Quintenz, who while a former CFTC commissioner, was also recently on the board of Kalshi.

If Quintenz’ connections weren’t clear enough on his views, in 2021, he published a CFTC dissent relating to a company called ErisX which was offering NFL Moneyline futures. In his dissent, titled “Any Given Sunday in the Futures Market,” Quintenz’ was very clear that he believes sports betting is a typical commodity futures market:

Quintenz argued in the dissent that sports betting can be distinguished from gaming- pure games of chance such as Blackjack, because they are predictable to a point:

There are “sharps” in Vegas who have made a career out of a sports betting, just the public on average is very bad at it. The other question is whether allowing these contracts is in the public interest, Quintenz’ believed that “public interest” is too vague a term to make decisions on, but there is an argument that sports futures can be within the public interest, due to making sports betting cheaper.

This isn’t too hard of an argument to make.
When you play a bet, most sports book will take hold revenue. When a spread is set by a sports book, that is typically meant to be 50/50 odds. So for this all star game: there is assumed to be a 50/50 chance Team Chuck wins by 3.5 points

Now notice most books have -110 under the spread. This means you have to bet $110 to win $100. So even though Team Chuck has a 50% chance of winning by 3.5+, your actual odds on a bet on are closer to 45% as Draft Kings keeps an extra $10 if you win. Currently DraftKings’ average hold is 9.4%, but management believes it can get to 11% over time:

Prediction markets have the ability to run with much lower holds. Interactive Brokers’ ForecastEx is only charging 1 cent per contract.

With contracts being 100 cents total (100% of being right), this means the same 50% odds bet at before could have a hold of only 2% (1 cent). Kalshi is currently operating at a 7% hold (across the board), but I expect this to decrease given the competition.

Implications for DraftKings

So effectively, we’re in a period where sports “predictions” is a gray area under the law, with an incoming CFTC chair and the president’s son having a vested interest in prediction markets. It is also possible, given Quintenz’ ties to A16z Crypto, he may even try to open the market to crypto-based platforms such as Polymarket.

If the CFTC confirms a regulatory stance that sports prediction markets are legal nationwide this will create a number of headwinds for DraftKings. California/Texas would be open for business faster than expected, but it is already clear from the Super Bowl that new competitors will include retail brokerages such as Robinhood and crypto exchanges like Crypto.com. WeBull and other platforms may also enter this space. Instead of decreasing its marketing spend as anticipated in 2026/2027 DraftKings likely would have to increase it up to defend against these competitors.

Unlike previous competition (such as the ESPN Bet), these competitors should be structurally cheaper for sports betters. Beyond lower holds (the bar to be consistently profitable is winning 52% of the time instead of 55%), DraftKings has a practice of limiting or banning players who win consistently. Prediction markets work on a zero-sum brokerage model- if one player wins, another player loses. Winners in prediction markets can keep winning, and Kalshi or ForecastEx will take their cut. These markets are also likely exempt from current state taxes on gaming.

There are some differences- more niche markets such as college lacrosse will have trouble finding enough liquidity for bets. For those markets, DraftKings itself is taking the other side of the bet, effectively playing the role of a market maker. Still the big 4 sports and notable events such as March Madness should have liquidity enough for most bettors, given Robinhood has 24 million users. Illinois reports monthly data on wagers by sports. Football and Basketball in December were far more popular for betting than any other sport:

DraftKings in the face of the new competition may face pressure to reduce its hold percentage instead of working its way up to 11% as management anticipated.

On the last earnings call, DraftKings was asked by Merrill Lynch’s Shaun Kelly about prediction markets: “What's DraftKings' initial take here? Is this something you'd get into? Or what can kind of protect or differentiate you from that kind of offering?”

Management responded: "Yeah. I do think you're right. It's early. We are watching it very actively. It's certainly something that we have keen interest in seeing how it plays out. So I think there's some, in the next couple of months, 60 days or so, there's going to be a CFTC ruling and all sorts of other things, so I think we'll know a lot more over the next few months.”

It is possible the other FTC commissioners force Quintenz to adopt a compromise view, but given Donald Trump Jr.’s involvement, there could be substantial political pressure from the White House to get this done.

Conclusion

As the regulatory view comes into play over the next 60 days, it is quite likely DraftKings’ stock could suffer as DraftKings does not have room for error in its valuation. DraftKings is currently trading at 93 NTM PE, although analysts expect that to improve to 35 PE in 2026 as DraftKings pulls back on marketing spend while continuing to grow.

If DraftKings faces pressure on hold margins or is forced to increase marketing spend in 2026 to combat the new prediction market players, the turn to profitability could be delayed. Management’s initial non-answer suggests they do not believe this is bullish for DraftKings. This is likely a case where the market hasn’t caught up to a story that could develop quicker than expected.