Variance in poker is the spread of actual results around the mathematically expected outcome. It is the main reason why even a strong player can run below expected value over the short term, and why weak players sometimes show good results over a month or two. Understanding variance is not an optional skill — it is a prerequisite for psychological stability and correct bankroll management.
What Is Variance
Technically, variance is a mathematical measure of how values spread around the mean. In poker it is measured through standard deviation, expressed in big blinds per 100 hands. The higher the variance, the wider the corridor of possible outcomes for any given number of hands. For NL Hold’em cash games, typical standard deviation values range from 80 to 120 bb/100. This means that on any given 100-hand sample, results can vary by ±80–120 bb from the expected.
Standard Deviation
Standard deviation is the primary practical tool for working with variance. A player with a winrate of 5 bb/100 and an SD of 100 can expect their actual result in 1,000 hands to fall within a very wide range. The more hands played, the narrower the corridor of expected results — but even 100,000 hands doesn’t fully eliminate variance. A professional who knows their SD doesn’t panic and doesn’t change strategy mid-downswing.
Expected Downswings (for winrate 5bb/100, SD 90)
- 10 buy-ins — several times per month
- 20 buy-ins — once every 1–2 months
- 30 buy-ins — once a year
- 50 buy-ins — once every several years
- 100 buy-ins — extremely rare
The Psychological Side of Variance
The main goal is to shift focus from results to process. A regular who evaluates every session by profit tilts constantly during downswings. A regular who evaluates themselves by the quality of decisions made remains stable regardless of the current result.
This is similar to how a professional athlete operates. A tennis player doesn’t assess every match purely by the score — they analyze technique, tactics, and physical condition. A poker player must also learn to separate the quality of play from the quality of results.
Tools for Analyzing Variance
- All-In EV in trackers — theoretical result based on actual equities
- Variance Calculator — calculating expected downswings
- Winrate graphs by session and week
- Standard deviation — key parameter for downswing modeling
All-In EV
All-In EV (also known as “EV line” in trackers) shows the theoretical result assuming equities run out. If a player’s actual result line lies significantly below the All-In EV line, they are running below expectation — this is a normal statistical phenomenon, not a sign of strategic mistakes. Conversely, if actual results are well above All-In EV, the player should be prepared for a likely reversion toward expectation.
Variance and Decision-Making
The primary danger of variance is its influence on decision quality. During a downswing, a player may start deviating from correct strategy: playing more cautiously, bluffing less, folding more thin value bets. All of this is a reaction to results rather than to game theory — a classic psychological leak.
The antidote: maintain a log of decisions. Analyze not whether you won or lost, but whether the decisions themselves were correct. If the decisions were right and the results are bad — it is variance. If the decisions were wrong regardless of results — it is a strategy problem.
Variance and Bankroll
Correct bankroll management is the main instrument for surviving variance. Standard recommendations: 20–25 buy-ins for NL cash, 50–100 for tournaments. These figures are not arbitrary — they are derived from standard deviation calculations for typical winrates and game formats. A player with proper bankroll feels even large downswings as a temporary fluctuation rather than a catastrophic event.
Common Mistakes in Working with Variance
- Evaluating decisions based on results (“I was right to call — I won”)
- Changing strategy during a downswing without a game theory basis
- Dropping down in stakes prematurely out of fear
- Ignoring All-In EV and other variance tools
- Assuming a downswing means “something is wrong with your game”
Variance in poker cannot be avoided. It can only be softened psychologically through understanding its nature, proper bankroll management, and shifting focus to process over results. Professional regulars work with this same reality year after year — large swings, prolonged downswings, uneven income. This is part of the profession, not a system failure.
One effective way to reduce the impact of variance is to increase your winrate, because a higher winrate narrows the confidence interval of actual results and shortens expected downswings. A large hand database from mining is one of the main tools for winrate growth at any stake: accurate stats on field regulars provide direct exploitative solutions and consistently improve results. The higher your winrate, the less you feel variance psychologically.