🎯 Stop Selling Winners: Portfolio Rotation, Position Sizing, and TAM Analysis Done Right

🧭 Introduction: The Challenge of Holding Winners

One of the most difficult skills in investing is knowing when to let winners run versus when to take profits and rotate capital. Analysis paralysis around selling often stems from FOMO — the fear that trimming a position will cap upside potential. However, strategic selling and rotation can amplify returns when done with discipline and a clear framework.

This discussion explores the mechanics of portfolio rotation, position sizing, and Total Addressable Market (TAM) analysis. It also examines key holdings across the tech and crypto sectors, from Bitcoin mining dynamics to the Tesla-SpaceX merger thesis, and provides concrete frameworks for trimming positions without killing compounding.

⛏️ Bitcoin Miner Capitulation: Does It Impact Scarcity?

Recent concerns about Bitcoin miner shutdowns have raised questions about network security and scarcity. The short answer: miner capitulation has zero impact on Bitcoin's 21 million supply cap.

Current data shows the Bitcoin mining cost-to-price ratio at approximately 1.2, meaning the cost to mine sits around $76,000 while Bitcoin trades near $64,000. Miners are operating at a loss unless they run ultra-efficient operations with cheap electricity and the latest rigs.

Key observations:

• Since early 2022, Bitcoin mining has been described as a "cutthroat business" with razor-thin margins
• Brief periods of profitability emerged in early-to-mid 2024, but difficulty adjustments quickly compressed margins again
• The 2021 China mining ban resulted in a 54% hash rate collapse in a matter of weeks, yet had zero long-term impact on Bitcoin's security or scarcity
• Bitcoin's difficulty adjustment mechanism ensures the network self-regulates as miners come and go
• Many miners now have optionality — pivoting to AI compute when Bitcoin mining becomes unprofitable

"Scarcity is fixed by code. Do not worry about miner shutdowns. As the price goes up, all the miners will come back in and FOMO back into mining."

The shift to AI compute creates a new revenue stream for miners, making them more valuable as businesses. Electricity is a bottleneck for AI, and miners control significant power infrastructure.

🚀 Tesla vs. SpaceX: Which to Stack?

With both Tesla and SpaceX offering compelling long-term theses, allocation decisions require careful analysis of market cap, TAM, and merger probabilities.

Current Market Caps:

• Tesla: $1.5 to $1.7 trillion range
• SpaceX: $2.1 to $2.6 trillion at peak

TAM Comparison:

SpaceX's S-1 filing listed a TAM of $28.5 trillion, with $22.7 trillion attributed to enterprise applications — essentially AI replacing enterprise software and certain employee functions.

Tesla's aggregated TAM across 20+ lines of business totals $76 trillion, with the largest components being:

• Humanoid robots (largest TAM)
• Robotaxi and autonomous software
• Energy storage (Megapack, Megapods)
• Compute infrastructure
• Full Self-Driving (FSD)
• Custom silicon (Tera)
• Lithium refining

Merger Arbitrage Model:

If the Tesla-SpaceX merger were to occur today, the arbitrage model suggests Tesla's stock price would move to $555 on the nose. Given Tesla currently trades around $457, this represents a premium for Tesla shareholders.

Strategic Recommendation:

While the merger is considered inevitable, the timing remains uncertain — likely 2027 or 2028 rather than 2025. From a valuation perspective, Tesla offers better math right now due to:

• Lower entry price relative to SpaceX
• Multiple near-term catalysts hitting simultaneously (CyberCab, FSD unsupervised, Optimus production ramp, Megapack demand)
• Merger premium baked into future valuation

That said, SpaceX remains a fascinating pure-play with zero competition in launch services and rapidly expanding into AI compute with data centers in space. Recent announcements include:

• V3 Starlink satellites (~60m long)
• IIA1 data center in space (~70m long)
• Elon Web Services (EWS) infrastructure buildout

A balanced approach involves building a 1/21 allocation to SpaceX while stacking Tesla as the primary position.

💡 The Art of Trimming Without Killing Winners

The psychological battle of selling winners often revolves around regret — either regretting not taking profits if a name crashes, or regretting capping upside if it continues to run.

Case Study: Marvell (MRVL)

An investor entered Marvell at $80 in January 2025 and sold 50% at roughly double the entry price. Currently, Marvell trades around $236, representing a 184% gain from the original entry.

The Pain: Selling half at 2x feels prudent, but watching the remaining half compound significantly creates FOMO about the trimmed position.

Framework for Managing Winners:

1. Rotate to Faster Horses

A detailed example involves rotating from Nvidia to Tesla in 2024:

• Nvidia equity sold at $160
• Capital rotated into Tesla LEAPs at $280
• Tesla subsequently moved to $410, delivering a 46% return
• Had the Nvidia position been held, it would have delivered a 30% return to $210

The rotation worked because Tesla was identified as the faster horse, and leverage amplified the return.

2. Reallocate on Dips with Leverage

During the April 2025 tariff tantrum, Nvidia crashed to $88. This presented an opportunity to re-enter with LEAP exposure, buying twice in the same day. Nvidia subsequently more than doubled to $210, recreating a large position that rejoined the long-term hold bag.

"You could trim, but you've got to rotate into that faster horse, and then reallocate on dips with leverage. That's the key."

3. Use 2030 Price Targets

Having multi-year price targets provides a framework for assessing whether a name is overheated or still has room to run.

Example 2030 Price Targets:

• Nvidia: Bear $800, Bull $1,200
• Marvell: Bear $550, Bull $1,200
• Micron: Bear $180, Bull $300

If a stock approaches the bull target early, it may signal an overheated condition suitable for trimming or hedging.

4. Hedge Instead of Selling

When ALAB spiked from $88 to $550 in a matter of weeks, selling wasn't the only option. Expensive options made selling hedges an attractive strategy, generating cash to rotate elsewhere while maintaining upside exposure.

5. Build Muscle Memory for Selling

Even selling 1-2 shares of a 100-share position builds the psychological discipline required to execute larger trims without emotional paralysis.

When to Hold Indefinitely:

Compounding is destroyed by excessive trimming. If a name has a 5x to 7x ROI potential over five years, holding through volatility often outperforms tactical trading.

The I13 Portfolio Philosophy:

The I13 is structured as a hold-through portfolio from 2028 to 2030 unless the thesis fundamentally changes. Examples of thesis-breaking events:

• Tesla builds a Terafab and displaces Nvidia in AI chip dominance
• CyberCab fails to generate scaled, high-margin software revenue by late 2027
• Humanoid robots (Optimus) fail to generate scaled revenue by late 2028

"If CyberCab does not generate scaled, high-margin software revenue by late 2027, the AI thesis for autonomy is dead. We leave. If humanoids do not generate scaled revenue by late 2028, that thesis is also broken. Then hope is not a strategy."

📊 CAGR Analysis: Tesla's Long-Term Dominance

Concerns about Tesla's near-term CAGR performance miss the bigger picture. A comprehensive CAGR table comparing Tesla, S&P 500, Bitcoin, and Gold over rolling 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10-year periods reveals:

• Tesla's average CAGR across all periods: 30.88%
• Bitcoin: 16.88%
• S&P 500: 16.13%
• Gold: 31.67% (recent strength driven by Chinese central bank buying)

Key Insight: If an investor randomly deployed capital at any point over the last 10 years, Tesla delivered the highest average CAGR.

Tesla's performance pattern: Quiet accumulation for ~4 years, followed by explosive 1-year rallies, then another 4-year base. The market is currently in the quiet accumulation phase, with multiple catalysts set to converge:

• CyberCab launch at scale
• FSD unsupervised rollout
• Optimus production ramp (targeting 130,000 units by year-end)
• Megapack demand surge
• Tera chip production

TAM Breakdown Recap:

• Robotaxis: $10 trillion
• Humanoid robots: $25-30 trillion

"When you look at what SpaceX has done with Starship and Starlink — if they can solve Starship, they can solve humanoids and CyberCabs very easily. That's why I have confidence."

📈 Position Sizing: How Much Concentration Is Reckless?

For a 25-year-old investor with a long time horizon and conviction in a deeply oversold name like MicroStrategy, the question of maximum concentration is critical.

Recommendation: Do not exceed 15% in a single name.

At 25, time is a major advantage, but preservation of capital and diversification still matter. A 15% allocation allows meaningful exposure without catastrophic risk if the thesis breaks.

For aggressive investors, 20% may be acceptable, but anything beyond that introduces unnecessary tail risk.

MicroStrategy (MSTR) Outlook:

Wall Street analyst price targets for one year out:

• High target: $570
• Average target: $275 (implying a ~3x from current levels)
• Low target: $130 (still >30% upside)

If the average target is realized, a 15% allocation could become a significantly larger portfolio weight through appreciation alone.

"You play where the puck is going to be. When you think about your allocations, you have to understand future price targets and the impact on allocations."

🤖 The Robotics ETF Question: Pure Play vs. Diluted Exposure

A recent interview highlighted the Robbo Strategy ETF (BOT), which charges a 0.75% expense ratio and has delivered a 46% return over the past year.

Portfolio Breakdown:

• 35% Standard Bots
• 15% Figure AI
• 15% Dyna
• Remaining allocation spread across Appronic, Upronic, Path, and others
• Zero Tesla exposure

Critical Issues with Thematic ETFs:

1. Fee Drag: A 0.75% expense ratio compounds negatively over time, eating roughly 10% of returns over a decade.

2. Dilution: Baskets of 20-50 stocks include laggards that dilute performance. Many names will fail or underperform.

3. Pre-IPO Names: Standard Bots and other holdings are private, with no public financials. Valuation and due diligence are opaque.

4. No Apex Predator Exposure: Tesla, widely regarded by top VCs and allocators as the leading humanoid robotics play, is absent from the fund.

Conclusion: The ETF structure is a Wall Street vehicle for farming retail fees. Pure-play exposure to the apex predator (Tesla) is preferable for those seeking concentrated robotics upside.

For those interested in Standard Bots specifically, private investment channels may offer better terms than paying ongoing management fees for a diversified basket.

📉 Amazon vs. Marvell: Rotation Decision

An investor holding Amazon with a 2x gain is considering rotating into Marvell at current levels.

Ratio Analysis (Amazon/Marvell):

The ratio currently sits at 1.04, down from a high of 3.8, indicating Amazon has underperformed Marvell by roughly 76-78% over the past year.

2030 Price Targets:

• Marvell Bull Case: $1,200 (6.3x from ~$200 current price)
• Amazon Bull Case: Unlikely to 6x from current levels

Assessment:

Marvell is a pure-play AI semiconductor story, while Amazon is diversified across retail, cloud, and robotics. Marvell has already significantly outperformed, but the question is whether it will continue to do so.

Caution: Marvell recently corrected from $330 to $226, though it appears to be stabilizing. The ideal rotation point was 6-12 months ago.

Recommendation:

• Run a capital gains tax analysis to understand break-even requirements
• Consider holding Amazon if it's a long-term conviction play
• If rotating, monitor Marvell's chart closely for re-entry timing
• Alternatively, consider Google or SpaceX as hyperscaler plays with cleaner risk/reward profiles

🎯 Final Thoughts: Discipline, Process, and Patience

The difference between successful long-term compounding and constant churn lies in discipline and process:

• Wait for big opportunities rather than overtrading daily noise
• Rotate strategically into faster horses with clear conviction
• Use price targets to assess whether names are overheated or still have runway
• Hedge, don't always sell when volatility spikes and option premiums are elevated
• Trim systematically to build selling discipline without capping compounders
• Know when hope becomes toxic and be willing to exit when theses break

"It took me a decade to learn: less is more. Wait for the big opportunities. Don't be a crackhead trader thinking you have God's gift. The big opportunity on Nvidia was $160, then $88 six months later. These are the things you wait for."

For those holding Tesla, the patience required to endure multi-year consolidation phases is justified by the scale of TAMs being pursued. Humanoid robots, robotaxis, energy storage, and compute infrastructure represent the largest addressable markets on Earth.

As the factory buildout accelerates and production ramps, the market will eventually reprice Tesla for the compounding revenue streams ahead. Until then, the strategy remains: stack, hold, and wait for the rampage phase. 🚀