martin shkreli

Mathematical proof shows optimal bet size = edge - (1/edge); at 55% edge optimal is 10% of capital. Most traders size as if they have 4-5x Kelly edge, guaranteeing eventual blowup. Simulator demonstrates even 60/40 edge goes bust with overbetting.

Shkreli's simulator shows even a 60/40 edge leads to certain ruin if position size exceeds Kelly optimum; virtually all hedge funds and retail traders run 2-10x Kelly leverage, making blowups mathematical inevitability rather than bad luck. The only sustainable approach is chronic underbetting (half/quarter Kelly) with large cash reserves.

OpenAI, Anthropic, Microsoft, Google, and Meta are posting record business results, yet AI infrastructure stocks (chips, memory, neoclouds) are crashing because the marginal 5% of levered holders are forced to sell; price is set by liquidity needs not fundamentals, and the correction compressed a 3-4 year dot-com cycle into one month.

AI infrastructure trade (memory, chips, neoclouds) followed classic bubble pattern: smart money entered early, less sophisticated buyers chased 400% gains at peak, 4x leverage amplified 25% drawdown into total wipeout; fundamentals irrelevant to marginal 5% of shares setting price.

When a fund's equity approaches zero, prime brokers (Goldman, BofA) contractually take control and liquidate at any price to crystallize a known loss rather than risk a larger one; they preferentially allocate the distressed book to multi-strat platforms (Citadel, Millennium) that have the balance sheet and market-making capacity to absorb it, creating a recurring 'rescue buyer' franchise.

Historical pattern over 50 years shows hedge funds that add private/VC sleeves (like Tiger, SoftBank in 2021, and now Situational Awareness) typically fail because the skill sets, time horizons, and liquidity profiles are incompatible; Leopold's large illiquid Anthropic position became an anchor preventing portfolio adjustment during the liquidity crisis.

Market rewarded Meta and Microsoft for measured capex while punishing Google for aggressive spending; fundamental question remains whether massive AI infrastructure investment will generate sufficient returns or follow dot-com bust pattern.

Prime brokers (Goldman, BofA) rehypothecate client assets to earn financing spread; when blowups occur, they offload books to multi-strat platforms like Citadel rather than warehouse risk, making Citadel a systemic shadow bank that profits from distress once per decade.