Standard Deviation
标准差
Context: Term: Standard Deviation Standard deviation is a statistical metric that measures the fluctuation range between a player's actual profit results and expected profit, reflecting the influence of short-term luck components. In Texas Hold'em, standard deviation is used to assess the risk level of bankroll management: the larger the standard deviation, the more volatile your profit or loss swings, potentially indicating longer downswings or upswings. This often means your playstyle relies more on luck than skill. For example, a tight-aggressive player might have an hourly profit standard deviation of 100 BB, while a loose-aggressive player could reach 200 BB. The latter may experience greater bankroll fluctuations in the short term, requiring a more substantial bankroll to weather the variance.
Context: Term article: 标准差（Standard Deviation）
Overview
Standard Deviation (SD) is a key statistical measure in Texas Hold'em that quantifies the volatility of a player's winnings. It represents the degree to which a player's actual profit over a given number of hands deviates from their expected profit (based on skill advantage). The larger the standard deviation, the more short-term results are influenced by luck, and the greater the swings in bankroll a player may experience.
Calculation and Unit
In poker, standard deviation is typically expressed in units of "big blinds per 100 hands" (bb/100). For example, if a player has a standard deviation of 100 bb/100, then over every 100 hands, there is approximately a 68% probability that their actual profit falls within ±100 bb of their expected profit, and about a 95% probability that it falls within ±200 bb. Standard deviation is calculated based on historical profit data using the formula:
SD = sqrt( Σ (x_i - μ)² / N )
where x_i is the profit per 100 hands, μ is the mean profit, and N is the number of samples.
Relationship with Win Rate
Standard deviation and win rate together determine a player's long-term performance. Win rate measures skill advantage, while standard deviation measures variance. For example, a player with a win rate of 5 bb/100 and a standard deviation of 100 bb/100 needs a large number of hands to ensure profitability. According to the central limit theorem, the required number of hands is approximately (SD / winrate)² × 100. In this example, roughly (100/5)² × 100 = 40,000 hands are needed for about a 95% confidence of making a profit.
Applications
- Bankroll Management: A higher standard deviation requires a larger bankroll reserve to avoid the risk of ruin. It is generally recommended that the bankroll be at least 20 standard deviations (in units of per 100 hands) multiplied by the buy-in amount.
- Performance Evaluation: Short-term profits can be dominated by luck; standard deviation helps distinguish between luck and skill. For instance, a player who profits 20 bb/100 over 1,000 hands, with a standard deviation of 100 bb/100, has a 95% confidence interval ranging from -80 to 120 bb/100, so it cannot be concluded that their skill is superior.
- Game Selection: Different game types (e.g., tournaments vs. cash games) have different standard deviations. Tournaments typically have higher standard deviations due to steep payout structures.
Limitations
Standard deviation assumes that the profit distribution is approximately normal, but poker profit distributions often exhibit fat tails (higher probability of extreme outcomes), causing actual variance to be larger than theoretical values. Additionally, when the sample size is insufficient, the calculated standard deviation is unstable.