The February 2026 Selloff › Why a 33% Loss Is Not the Opposite of a 50% Gain

The most basic reason selloffs destroy wealth faster than rallies create it is arithmetic. A stock that falls 50% must rise 100% to get back to where it started. A 33% decline requires a 50% recovery. A 75% loss needs a 300% gain. This is just how multiplicative returns work. | Loss | Gain Required to Recover | |------|--------------------------| | 10% | 11.1% | | 20% | 25.0% | | 33% | 50.0% | | 50% | 100.0% | | 75% | 300.0% | | 90% | 900.0% | The compounding trap makes this worse over time. A \$100 stock that suffers a negative 50% compound annual growth rate for five years falls to \$3.13. If it then achieves a positive 50% CAGR for the next five years, it only recovers to \$23.73. Not \$100. Not even close. The loss and the gain are nominally identical, but the loss came first and hollowed out the base from which gains compound. The historical record backs this up. The average bear market since 1928 has lasted 289 days and declined 35%. The average bull market has lasted 988 days and gained 112%. Bears move roughly three times faster per unit of decline than bulls move per unit of gain. The COVID crash took 33 calendar days to erase 34% of the S&P 500's value. The Nasdaq's 83% drawdown after the dot-com bubble took 15 years and 5,442 trading days to fully recover. Even among the 20 best-performing individual stocks over the past four decades, the median maximum drawdown was 72%, with a median peak-to-trough duration of nearly three years and a median recovery time of over four years. Winning stocks suffer catastrophic drawdowns too. They come back. Most stocks don't. The math has a psychological counterpart. Prospect theory established that humans feel losses roughly twice as intensely as equivalent gains. When a portfolio drops 30% in a week, nobody calmly calculates that they need a 42.9% gain to recover. The response is panic and the urge to sell. That behavioral response turns the arithmetic asymmetry into a self-reinforcing cycle: losses trigger selling, selling creates more losses, and the cycle runs until either prices fall far enough for new buyers to step in or a central bank intervenes.