5D Library › Regime-Switching Fractional Kelly, Maximum-Drawdown Constraints and Lifecycle Leverage Allocation · 中文

Preface: Bet Sizing When the Weather Changes

Note

Please read this first. This book studies three things: why a full Kelly bet computed in good times so often ends in a deep drawdown when the market turns; what fraction of Kelly keeps a drawdown within a stated limit; and how leverage should change over a working life as human capital is turned into financial wealth. Its model parameters are illustrative or synthetic, meant to show structure; the only historical figure is the one-day fall of 19 October 1987; all calculations are generated by the programs printed in the book. This book is not investment, legal or tax advice.

Several claims about leverage are popular: that full Kelly on an optimistic estimate of the drift almost surely loses half or more when the cycle turns; that a fraction of Kelly between 0.15 and 0.35 keeps the maximum drawdown under 20% with high probability; that the young should borrow to hold equity and glide down to less as they age. This book writes them as arithmetic that can be run, then asks when they hold.

What the book covers

Chapter 1 derives the Kelly growth curve and shows how poorly the drift can be estimated, especially during a long bull run. Chapter 2 computes the chance that a Kelly bettor’s wealth halves, in continuous time and in a market with hidden bull and bear regimes. Chapter 3 asks how fast a bear regime can be recognized and what a regime-aware Kelly buys. Chapter 4 turns a drawdown limit into a fraction of Kelly and compares it with a floor on the running peak. Chapter 5 checks Ralph Vince’s optimal f. Chapter 6 builds the lifecycle balance sheet with human capital and tests early leverage when layoffs come with bear markets. Chapter 7 reads the Shiji’s “when things flourish they decline” and Feng Xuan’s three burrows. Chapter 8 gives a ten-year leverage table, gap risk and a drawdown warning rule.

Claims corrected after checking

  1. “Almost surely” needs twice Kelly. At the true Kelly, wealth ever halves with chance 0.500 and falls to a fifth with 0.200; only from twice Kelly is every low certain (Chapter 2, p03). But a full Kelly computed from bull months alone is 7.10 times wealth, and in the regime market it had a drawdown of half or more on every path and was wiped out on 0.640 of them within ten years (p04).
  2. Overbetting comes from the data. With ten years of data the standard error of the drift is 0.0506, as large as the drift; estimated after a three-year bull run, the median Kelly is 5.69 against a whole-history 1.46 (Chapter 1, p02).
  3. 0.15 to 0.35 of Kelly is too generous when regimes persist. For a 20% limit with chance 0.05 over ten years, the largest fraction is 0.20 in an i.i.d. market but 0.10 in the regime market (Chapter 4, p07); measured in leverage, a 20% limit allows 0.34 to 0.39 of wealth at volatility 0.16 whatever the Sharpe ratio (Chapter 8, p14).
  4. Regime filters are slow and fragile. Even with the true parameters, the median delay before a bear is flagged is two months, and 0.193 of bear spells are missed (Chapter 3, p05); with an optimistic model the protection of a regime-aware Kelly mostly disappears (p06).
  5. Optimal f is Kelly, and is hard to estimate. Growth turns negative above f = 0.594, twice the optimum of 0.300 (Chapter 5, p09); estimated from 30 trades, half of the estimate beat the full estimate: median TWR 2.31 against 1.75 (p10).
  6. Early leverage raises the median and the tail risk together. With layoffs in bear months, the two-to-one lifecycle rule had median wealth 1,725 at 65 against 1,477 for all equity, but a drawdown of half or more before 40 on 0.783 of lives (Chapter 6, p12).
  7. Two readings of the classics. “When things flourish they decline” describes a rising hazard, which simple regime models lack; Feng Xuan’s three burrows cut risk a hundredfold only if they fail independently, and only by half under a common shock (Chapter 7, p13).

How to read the boxes

The boxes separate published evidence, runnable Python computations (16 programs, standard library only, fixed random seeds, each actually run at build time with its real output printed beneath it), classical sources, notes and the author’s hypotheses; every hypothesis states what result would show it wrong. Programs and their output use ASCII characters only.

Disclosure

The author uses the works of 5D Library (including this book) as the methodological basis of community courses, the 5D Ascension Cards (Dimensional Shift Card Decks) and the Yafu AI knowledge base, and runs several websites, paid online calculators, subscriptions, reports, courses, business licences and advisory services, including a drawdown warning and leverage table calculator, which is exactly what this book studies; the author therefore has a direct commercial interest in its subject. This book promotes no software, course, card deck, financial product or service, uses no client’s financial data, and accepts no sponsorship from any financial institution. All parameters in the book are synthetic or illustrative. The hypotheses of Chapters 1 to 8 must be tested under pre-registered protocols, and the results published whether positive or negative.

Fan, Y. (2026). Regime-Switching Fractional Kelly, Maximum-Drawdown Constraints and Lifecycle Leverage Allocation: Cyclical Drift Estimation Errors, Ruin-Bounded Bet Sizing and Human-Capital Amortization [体制转换分数凯利、最大回撤约束与生命周期杠杆配置] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090256 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).