Algorithmic Short Selling With Python Strategies, signals, and risk management techniques for profitable short trades (Laurent Bernut) (z-library.sk, 1lib.sk, z-lib.sk)

【One-Line Pitch】 A practitioner's guide to treating short selling as a quantitative, algorithmic discipline—covering market philosophy, long/short portfolio construction, regime detection, and the Python tooling to test it all. Best for traders and quant-minded developers who already know Python and want a structured path into the short side of the book. 【Book Arc】 - **Opening (~0%–10%)**: Frames the market as an infinite, complex, random game and argues that short selling is best played algorithmically rather than by gut feel. - **Early (~10%–32%)**: Dismantles ten common myths about short sellers (pensions, volatility, "unlimited loss"), then introduces the two foundational long/short methodologies—the absolute method and the relative weakness method—and shows why the absolute approach underdelivers. - **Middle (~32%–55%)**: Builds the practical toolkit: downloading prices with yfinance, computing absolute vs. relative series, currency-adjusting global sector baskets, and using heatmaps to triage winners from losers. - **Late (~55%–85%)**: Extends into regime definition methodologies (breakouts, moving averages, shadows) applied to both absolute and relative series, plus sector-level aggregation and bullish/bearish counting. (Excerpts do not cover the exact chapter boundaries here.) - **Ending (~85%–100%)**: Turns to operations and self-review—TradeLog analytics, maximum adverse excursion, drawdown and consecutive-loss analysis, a psychology journal with AI-assisted weekly/monthly synthesis, and access to the code bundle. 【Key Takeaways】 - **Short selling is an algorithmic sport** (Early): Because most participants only play the long side, short sellers must do their own research; algos plow through data and strip out execution emotion. - **The "absolute method" is a lesser vehicle** (Early): It lags mutual and index funds in bull markets, only loses less in bear markets, and practitioners tend to freeze on the short trigger—net beta stayed positive even through the GFC. - **Relative weakness is the more coherent framework** (Early): A long/short book is the net sum of two relative books—outperformers on the long side, underperformers benchmarked to the inverse index on the short side. - **Relative series beat absolute series for cross-market work** (Middle): Converting local prices into a benchmark currency and comparing relative returns makes global sector comparisons (e.g., autos) tractable. - **Heatmaps are a triage tool, not a conclusion** (Middle): They let you sort winners from losers in seconds, but the real work—fundamental research to explain *why*—comes after. - **Regime definition is a design choice** (Late): The book illustrates multiple methodologies (breakouts, moving averages, shadows) for both absolute and relative series, implying no single regime filter is canonical. - **Risk is permanent capital loss, not volatility** (Early): The book channels Munger's framing and uses drawdown, MAE, and consecutive-loss analysis as the operative risk lens. - **Journaling and review close the loop** (Ending): TradeLog analytics plus a psychology journal with AI-assisted synthesis turn execution into a feedback system. 【Reading Tips】 - Skim the myth-busting chapter if you're already convinced short selling isn't villainous; deep-read the relative weakness method and the currency-adjustment workflow—those are the load-bearing ideas. - Treat the code blocks as templates, not gospel: the excerpts show heavy reliance on yfinance, pandas, numpy, and matplotlib, so have that stack ready before starting. - The regime-definition chapter repeats the same code block with different methodologies—read one carefully, then skim the rest for the differences. - Don't skip the ending chapters on TradeLog and the psychology journal; they're where the book connects signals to actual trading behavior. 【Coverage Limits】 This guide is based on stratified excerpts covering roughly the first half of the book plus the table of contents and ending chapters; specific signal formulas, backtest results, and mid-book chapter details are not fully represented.