Stop Hedging Nvidia and Palantir With SPY Puts. Here's What the Data Says Works Instead.

Key Concepts

• Correlation (Rho): A statistical measure indicating the extent to which two assets move in relation to each other.
• Beta: A measure of a stock's volatility (sensitivity) in relation to the overall market.
• AI Narrative: The specific company-driven story or fundamental driver that influences a stock's price independently of broader market trends.
• Beta-Weighted Risk: A risk management framework where position sizing is determined by the asset's sensitivity to market movements rather than just the dollar amount invested.
• Diversification Decay: The phenomenon where assets that are typically uncorrelated begin to move in lockstep (correlation spikes) during market crashes or periods of high stress.

---

1. Analysis of AI Stock Dynamics

The presentation challenges the conventional wisdom that all AI stocks move in unison. Research covering the period from 2020 to 2026 reveals a significant shift in market behavior:

• Correlation Trends: The average correlation between AI stock pairs dropped by approximately 23% between the pre-AI era (2020–2022) and the current AI era (2024–2026).
• Market Sensitivity (Beta): While correlations between individual AI stocks have decreased, their sensitivity to the broader market (beta) has risen to 75%.
• Divergent Behavior: Stocks like Microsoft, Google, and Amazon maintain high correlations to the market (>70%), whereas stocks like Nvidia and Palantir have decoupled, moving more based on their own specific "AI narratives" rather than general market signals.

2. Hedging Strategies and Frameworks

The speakers emphasize that hedging strategies must be tailored to the specific characteristics of the stock:

• Market-Correlated Stocks: For stocks that track the market closely, S&P 500 puts are an effective hedging tool.
• Non-Correlated Stocks: For stocks like Nvidia or Palantir that move independently, standard market-wide hedges (like S&P puts) are ineffective. Traders must analyze these stocks individually rather than relying on market-wide signals.
• The "65% Threshold": The team suggests using a 65% correlation threshold. Stocks with a correlation above 65% to the market should be treated as market-sensitive, while those below should be treated as idiosyncratic (driven by internal factors).

3. Portfolio Management and Risk

• Volatility Impact: Even if an AI stock has a low correlation to the market, its high individual volatility can significantly increase the overall volatility of a portfolio.
• Position Sizing: Traders are advised to move away from dollar-based sizing and toward beta-weighted risk sizing. This ensures that the position size is proportional to the actual risk exposure rather than just the capital allocated.
• The "Crash" Exception: A critical warning is provided: during market sell-offs or periods of extreme stress, the diversification benefits of these stocks disappear. Correlations tend to spike toward 90% or higher, meaning all AI stocks will likely move down together regardless of their individual narratives.

4. Notable Statements

• "Not all AI stocks are created equal." — The central thesis of the presentation.
• "If you trade them, don't try to pretend Google, Meta, Amazon, Palantir will move in the same directions or have the same strategy."
• "Diversification can disappear when the market gets stressed or when it's in the sell-offs."

5. Synthesis and Conclusion

The primary takeaway for traders is that the "AI sector" is no longer a monolithic trade. Because AI stocks have become less correlated with each other but more sensitive to the market, a "one-size-fits-all" approach to trading or hedging is dangerous.

Actionable Insights:

1. Classify: Categorize AI holdings by their correlation to the market and their individual beta.
2. Analyze: Research the specific "AI narrative" for each company, as these now drive price action more than general market sentiment for certain stocks.
3. Size by Risk: Use beta-weighted risk to determine position sizes.
4. Prepare for Stress: Acknowledge that in a market crash, all correlations will converge, rendering standard diversification strategies temporarily ineffective.