InvestAnswers
InvestAnswersyesterday
Finance

🚀 Next 6x, Best Pair Trades, Best Pivots, Bitcoin Engine, AI Compute Plays🧠💥

38 min video7 key momentsWatch original
⚡
TL;DR

MicroStrategy's STRC preferred stock has collapsed below par value, breaking its designed 99-100 trading range, which threatens non-dilutive Bitcoin accumulation; meanwhile, AI inference compute is generating 20x more revenue per megawatt than Bitcoin mining, making Elon's empire better positioned for AI than crypto.

Key Insights

1

Losing $27k per coin — Bitcoin miners are losing $27,000 per coin mined when cost sits at $91k and price is $64k—the largest divergence in years.

2

20x revenue difference — AI cloud inference generates $1,600-$4,000 per megawatt-hour versus $80-150 for Bitcoin mining—a 20x gap explaining why miners pivot to AI.

3

Need 10x capacity — By 2030, AI inference needs to hit 93.3 gigawatts but hyperscalers have nowhere near enough capacity—they need to 10x in 5 years.

4

6x if hits trillion — Marvell could be a 6x from here if Jensen Wong's trillion-dollar thesis plays out, since it's the connectivity bottleneck leader.

5

Capex underrated — Tesla's capex spending of $25B looks cheap compared to Mag Seven peers building factories, refineries, and Optimus production at scale.

6

Practice one kick 1,000 times — For young investors with time, find one skill and practice it 1,000 times instead of spreading across 1,000 different approaches.

Want this for every new video InvestAnswers posts? Brevyd summarizes each upload automatically, the morning it drops.

Deep Dive

MicroStrategy's Bitcoin Engine Breaks Down

Doc's question hits a sore spot: InvestAnswers sold down STRC from 56% of portfolio because the preferred stock designed to trade 99-100 has collapsed to 86-87, breaking the whole non-dilutive Bitcoin accumulation machine. When STRC trades below par, new investor demand dries up, starving the cash MicroStrategy needs to buy Bitcoin without issuing common shares. The 12% semi-monthly coupon was meant to keep it stable, but the math no longer works—Bitcoin would need 2-3% annual gains just to cover that yield. If STRC stays underwater, MicroStrategy gets forced to dilute common shareholders, which is why InvestAnswers said they're 'cooked.' The ATR model shows it's tradable, and options exist on STRC, but the red flag is the 200-day moving average sitting 7 points higher. Recovery depends on Bitcoin moving, but right now the entire flywheel is stuck.

Why Bitcoin Mining Lost to AI Compute

Dex asks the theoretically smart question: shouldn't Elon space-mine Bitcoin? But the math is catastrophic. Average miners lose $27,000 per coin with cost-to-mine at $91k while Bitcoin trades $64k. That's the biggest divergence in years. Meanwhile, AI cloud inference—the ability to run models in real-time on edge devices and data centers—generates $1,600 to $4,000 per megawatt-hour. Bitcoin mining generates $80-150 per megawatt-hour. That's a 20x gap. BigTech hyperscalers like Amazon, Microsoft, and Google realize they don't have enough AI compute capacity, so they're leasing it from SpaceX and xAI. By 2030, inference workloads hit 93.3 gigawatts according to Hitachi, but the four hyperscalers combined have maybe a fraction of that. They need to 10x capacity in 5 years. InvestAnswers cites Jensen Huang saying 99% of future AI compute will be inference-based. So Elon's smarter play—and what he's already doing—is building 10+ gigawatts of inference capacity via Cortex, Colossus, megapods, and space-based data centers. Mining Bitcoin with that capacity would be leaving 20x money on the table.

Echo vs. SpaceX: The ARB Trade Nobody Wanted Yet

Finn asks if he should rotate 100 Echo shares into SpaceX three-year leaps. InvestAnswers runs the math: bare case values SpaceX at $81 per share via Echo, base case $140 makes it effectively $73, bull case $64. Echo's trading $88 now but should be $159 if the math holds. So there's an arb, but it's messy because Echo has subsidiary bankruptcies (Dish DBS, Dish Wireless prepackaged for June 2026), AT&T spectrum deals with contingent liabilities, and core business shrinkage. However, the pair chart shows Echo has actually outperformed SpaceX since June 17th, which is starting to show market appreciation of the discount. The problem: don't buy three-year SpaceX leaps right now because insider unlocks run till November and could dump millions of shares, tanking the stock and destroying option value. Better move is hold Echo for now, wait for the unlock calendar to clear, then consider leaps. The Starship landing on Friday was extremely bullish—if reusable, 200 launches could unlock a trillion in revenue and $20T valuation in 3-3.5 years. But timing matters more than thesis when leverage is involved.

Marvell: The 6x Nobody Sees Yet

Unruly Julie asks about swapping 10 Nvidia shares for Marvell since the pair trade suggests Marvell will outpace Nvidia. InvestAnswers confirms he has Marvell as a top-3 position (maybe second after Tesla and Nvidia), bought twice recently. The pair chart shows the optimal swap zone at 0.80-0.85 conversion. Marvell's $190, Nvidia's $207, so they're in the zone right now. The thesis: Marvell goes to $300 before Nvidia hits $330, giving 6x upside from current market cap of $170B to $1T. Jensen Wong said publicly that Marvell is the next trillion-dollar company, and he owns it, partners with it, and bundles it in Nvidia sales. Marvell controls one of three bottleneck keys: connectivity via photonics. The other bottlenecks are chip shortage and memory. Since the world needs all of these equally, Marvell's positioned perfectly. This won't happen overnight—it takes 2-3 years—but the capital flow from hyperscaler capex ($25B Amazon, $20B+ Microsoft, $13B+ Google) will flow to everything on the I13 list, and Marvell's a direct beneficiary. It's a patient trade with hard technical moats.

Why Kids' Portfolios Should Pick Winners, Not Indexes

Turfman worries about concentrating his kids' IRAs in I13 AI stocks plus Bitcoin, fearing a dotcom repeat where leaders vanished. InvestAnswers flips the concern: broad ETFs like VTI guarantee you miss the 0.3% of mega-winners and instead own mostly losers weighted by market cap. Kids have time, which is the weapon. The winners identified—Tesla, SpaceX, Nvidia, Marvell, ALAB, Micron, Saul—are no-brainers if you follow the capex flow. Yes, one might blow up (OpenAI's financials are horrific and could take Oracle with it), but 6x and 10x moves are coming. The hyperscalers will spend $25B+ annually on capex through 2028, all flowing to I13 companies. There's not enough connectivity, memory, or chips, so those receiving the spend don't even have to sell to win. The philosophy: don't buy-and-hold forever, rotate into new tech by 2028 (maybe Bitcoin, maybe space, maybe something new), use tax-free account flexibility to pair trade and sell covered calls, and follow the money ruthlessly. Confidence is warranted because the math is predictable even if prices are volatile.

Finding Your Rhythm: The Bruce Lee Rule for Trading

Axel Dirt, 32, financially set, wants to know which tool gives max ROI with 2-6 hours daily. InvestAnswers emphasizes: find your rhythm, like Bruce Lee said, 'I fear the man who practices one kick 1,000 times, not 1,000 kicks once.' Start with Lilo (layering model), learn in 1-2 hours. Then DCIS (DCA on steroids) to understand capital deployment. Then IDSS with confluence models, mean reversion, optimized trends. Finally ATR model for less frequent, higher-conviction trades. The magic is confluence: when two, three, or four indicators point the same direction, odds explode in your favor. InvestAnswers personally holds three 'rocks in pocket' trades per week max, sometimes waiting months for confluence to align. The discipline is ruthless: narrow focus on tools, assets, timeframes; track everything in a journal; identify why you won and why you lost; practice patience; expect 2-3 years of compounding, not overnight riches. Most new traders implode after 3-6 months of losses because they expected millionaire status instantly. Treating it like UFC training—every day, 2-6 hours, perfect the craft—is the only path to real success.

Takeaways

  • âś“If you own Bitcoin, deploy maximum 20% into MicroStrategy on evidence of bull market recovery, not before—its leverage works both ways and STRC's collapse shows the friction.
  • âś“Rotate Nvidia into Marvell at the 0.80-0.85 pair zone and hold 2-3 years for the trillion-dollar thesis; it's not tomorrow money but the moat is real.
  • âś“For kids' portfolios, forget broad ETFs and pick I13 winners (Tesla, Nvidia, Marvell, SpaceX) because hyperscaler capex through 2028 funds them directly with nowhere else for the money to go.
  • âś“Build a trading journal tracking every entry, exit, and why it worked or failed; practice one specific skill (one chart model, one asset class, one timeframe) until it's instinctive, not random.

Key moments

2:34STRC Trading Breakdown Explained

“if they don't get back to par, we're gonna get diluted to hell. And that's not good. That means cooked.”

6:46Bitcoin Mining Economics Collapse

“The average miner loses $27,000 to mine each Bitcoin, the cost to mine today is 90,993 bucks. And the price is 64K.”

7:41AI Compute Revenue Gap

“AI cloud inference yields anywhere between 1,600 bucks to $4,000 per megawatt hour compared to 80 bucks to 150 bucks for a typical modern Bitcoin mining fleet.”

10:28By 2030 Inference Dominates

“by 2030, inference will far surpass the dominant workload in AI heading to 93.3 gigawatt hours. AI training 62.2 gigawatt hours.”

21:00Marvell the Trillion-Dollar Thesis

“Jensen Wong said very vocally, Marvell is the next $1 trillion company. That could be a 5.88x. Call it a 6x.”

30:02Bruce Lee's Trading Wisdom

“I do not fear the man who practices 1,000 kicks one time each. I fear the man who practices one kick 1,000 times.”

35:06Tesla Capex Not a Disaster

“They're spending $25 billion in capex and that's a spit in the bucket compared to the other Mag Seven. They're spending way more.”

You just read one. Brevyd does this for every upload.

Follow InvestAnswers and every new video comes back as a summary like this, in your morning briefing. No watching required.