Capital
Misallocation
The
Multiplicative
Indictment
The Causal Barbell: Part 1
How much alpha does your hedge fund need to beat the index? You supply the assumptions. The arithmetic does not negotiate.
Market — feeds every tab
Defaults
These are starting values. Move any slider; the indictment is yours.
- 10-year hold
- 1.5-and-20, pod-level netting on, pass-through skipped
- ½σ² variance drag — conservative if the book is left-tailed
- Monte Carlo seed 20260910, 10,000 shared draws
- Cascade impact k = 0.09
- Shiller real total returns, 1871–2023
Tab 5 · Shiller rolling windows
In 100% of 10-year windows since 1871, the fee load exceeded the hedge benefit
Share of windows the hedged book won, by decade
0 of 15 decades. The cells aren't missing data — at 5% fees and 2% alpha the hedge never cleared the index.
Win rate across fee × alpha — 570 windows
| Fee \ Alpha | 0% | 2% | 4% | 6% | 8% | 10% |
|---|---|---|---|---|---|---|
| 2% | 2% | 13% | 30% | 63% | 89% | 99% |
| 3% | 0% | 8% | 20% | 49% | 78% | 96% |
| 4% | 0% | 2% | 13% | 30% | 63% | 89% |
| 5% | 0% | 0% | 8% | 20% | 49% | 78% |
| 6% | 0% | 0% | 2% | 13% | 30% | 63% |
| 7% | 0% | 0% | 0% | 8% | 20% | 49% |
| 8% | 0% | 0% | 0% | 2% | 13% | 30% |
Navy ring is your current fee and alpha. Green: hedge wins ≥ half of history. Gold: some decades. Oxblood: none.
Terminal wealth ratio (hedged / unlevered) across windows
Where the long-short book won
None of the 570 10-year windows since 1871. Median terminal wealth is 0.54× the index. Losing windows ran 6.6% geometric at 13% vol. Raise alpha or cut fees in the grid — the arithmetic does not negotiate.