Poker Bankroll Requirements Calculator | Reverse RoR

This poker bankroll requirements calculator estimates the bankroll needed for a chosen risk of ruin in cash games. Enter your winrate, standard deviation and risk target to see the requirement in big blinds, whole 100 bb buy-ins and dollars.

For example, a winrate of 5 bb/100 and SD of 90 bb/100 gives requirements of 25, 32 or 38 whole buy-ins for 5%, 2% or 1% eventual risk. These are estimates under a fixed-stakes model with a constant positive edge, not a guarantee that a bankroll will last.

How to use the poker bankroll requirements calculator

1. Enter an estimated long-term winrate in bb/100, after rake. A short winning sample can overstate your edge.
2. Enter standard deviation in bb/100 from the same game and sample. Game buttons supply illustrative starting values.
3. Choose the eventual risk target. Enter 1 for 1%, not 0.01.
4. Enter the big blind in dollars, such as 0.50 for NL50. Leave it blank if you only need bb and buy-ins.
5. Compare the result with the modeled risk for 20 buy-ins. Recalculate after changing any input.

One buy-in means 100 big blinds throughout this page. A minimum of 2,427 bb is different from 25 complete buy-ins, which contain 2,500 bb. The tool shows both dollar amounts with their corresponding bankrolls. Numbers use a dot for decimals; correctly grouped thousands such as 1,000 are accepted.

What is Risk of Ruin?

Risk of ruin is the probability that a starting bankroll reaches zero at some point during indefinite play. Here the game stakes, expected winrate and standard deviation remain constant, and money is neither added nor withdrawn. A 1% target means a one-in-100 modeled probability of eventually reaching zero under those assumptions.

A 5% target corresponds to one in 20; 10% corresponds to one in 10. These are scenario choices. They do not establish a universal standard for professional or recreational players, and a lower target does not eliminate risk.

This calculation has no hand-count input because it has no time limit. It does not measure the chance of a losing session, a losing year, or a particular peak-to-trough downswing. Those questions require a time horizon and a different calculation.

How the Math Works

The calculator uses the Brownian approximation for a bankroll with positive drift:

r = exp(−2 × w × B / s²)

Solving for the required bankroll gives the reverse risk of ruin formula:

B = −s² × ln(r) / (2 × w)

Understanding the Variables

• B: starting bankroll in big blinds.
• w: expected profit per 100 hands, expressed in big blinds.
• s: standard deviation of a 100-hand result, expressed in big blinds; conventionally labeled SD bb/100.
• r: target probability as a fraction, so 1% becomes 0.01.
• ln: the natural logarithm.

Time is measured in blocks of 100 hands. You do not divide the final answer by another 100; divide by 100 only when converting big blinds into 100 bb buy-ins. Variance is already s²: the formula squares standard deviation once.

The unrounded result is the model threshold. The calculator separately rounds up to the next whole big blind and the next whole buy-in. Reference tables use the same upward rounding, so a displayed bankroll does not fall below the calculated threshold.

The underlying barrier-hitting result is given in Karl Sigman’s Notes on Brownian Motion, section 1.8. It is exact for that continuous stochastic model. Applying it to poker is an approximation: real results occur hand by hand, opponents change, and estimated winrates are uncertain.

Typical Standard Deviation by Game Type

The existing game presets are retained below as illustrative scenarios. They are not universal averages or promises about a particular format. Use your measured SD when available; playing style, stack depth and game conditions can change it substantially.

The multiplier is (s / 100)², holding winrate and risk target fixed. SD 140 therefore gives 1.96 times the unrounded requirement at SD 100. Comparing games with different winrates requires recalculating both inputs.

Other calculators may start with SD 90 for an NLHE example. That is a different scenario, not a different formula: changing 90 to 100 increases the bankroll threshold by about 23.46% at the same winrate and target.

Complete Bankroll Requirements Table

These cash-game scenarios use SD 100 bb/100. All bankrolls are rounded up to whole big blinds; the last column independently rounds the 1% threshold up to whole 100 bb buy-ins.

Bankroll at 5 bb/100 and SD 90

This shared example also appears in our cash-game variance tool. The requirement is the same when the inputs and model match.

The last three columns price the whole-buy-in bankroll. At NL100 and a 5% target, the minimum rounded to 1 bb costs $2,427, while 25 complete buy-ins cost $2,500. Both are displayed by the calculator; they refer to different rounding units.

Example: Planning for Professional Play

Suppose you estimate a net winrate of 4 bb/100 and SD of 100 bb/100, and select a 1% target:

B = −100² × ln(0.01) / (2 × 4) = 5,756.4627… bb

• Minimum rounded up to 1 bb: 5,757 bb.
• Whole-buy-in bankroll: 58 BI = 5,800 bb.
• At NL100: $5,757 for the rounded-bb minimum, or $5,800 for 58 buy-ins.
• At NL200: $11,514 for the rounded-bb minimum, or $11,600 for 58 buy-ins.

Selecting 0.1% instead gives 8,635 bb, or 87 whole buy-ins. A lower target increases the required bankroll; it does not account for rent, taxes, withdrawals or an uncertain edge. Money reserved for living expenses is outside the playing bankroll modeled here.

Why “20 Buy-ins” Is Often Not Enough

Whether 20 buy-ins meet a target depends on both winrate and SD. At 4 bb/100 and SD 100, a 2,000 bb bankroll gives 20.19% eventual risk of ruin. That exceeds each of the 1%, 2%, 5% and 10% preset targets.

The SD is 100 for every row. At 8 bb/100, 20 BI gives 4.08% risk: it meets a 5% target but does not meet a 1% target. At 5 bb/100 and SD 90, 20 BI gives 8.47%, which exceeds a 5% target even though the rounded requirement is only 25 BI.

The correct comparison uses the unrounded probability against the selected target. It cannot be inferred from a broad label such as “under 25 buy-ins” or from a universal claim that 20 buy-ins are safe.

NLHE vs PLO: The Variance Multiplier

For a winrate of 4 bb/100 and a 1% target, SD 100 gives 5,757 bb or 58 whole buy-ins. With the same winrate and target, SD 140 gives 11,283 bb or 113 whole buy-ins.

The ratio of unrounded thresholds is exactly 1.96. SD 160 gives a 2.56 multiplier relative to SD 100; doubling SD gives four times the bankroll. These are mathematical comparisons between specified scenarios, not evidence that every PLO player needs a fixed multiple of every NLHE player’s bankroll.

If your edge falls as variance rises, the effects compound. Halving winrate doubles the threshold; combining half the winrate with SD 140 instead of 100 gives a 3.92 multiplier.

Preparing for Downswings

A downswing measures a drop from an earlier bankroll peak. Ruin measures reaching zero from the starting bankroll. After winning 30 BI, losing 20 BI from that peak leaves you ahead of your starting point, so a 20 BI downswing does not necessarily mean ruin.

The following table converts downswing sizes into dollars. It states amounts, not probabilities or predictions that a downswing must occur:

To estimate the chance of those swings, specify a hand-count horizon in the variance calculator and downswing guide. Simulated maximum drawdowns depend on the horizon, sample size and observation grid. The reverse RoR formula alone cannot justify a claim that you should “expect” a 20, 40 or 50 BI downswing over a particular number of hands.

Uncertain winrate, rake and withdrawals

The poker bankroll requirements calculator treats your entered winrate as known and constant. If the true edge is smaller, the displayed risk target will be understated. Test several plausible winrates rather than treating one short-term estimate as established skill.

For example, reducing winrate from 4 to 2 bb/100 doubles the unrounded threshold at the same SD and risk target. At SD 100 and 1%, the rounded requirement rises from 58 to 116 whole buy-ins. A result based on a positive assumption cannot establish that you are a winning player.

Use winrate after rake. If your tracker already reports net winnings, subtracting rake again would count it twice. Include rakeback only when it is reliable, not already counted and available to the bankroll; a delayed promotion or uncertain leaderboard prize is not equivalent to a constant cash flow.

Moving down in stakes, making deposits, taking regular withdrawals, or changing games alters the process. This calculator does not model those policies. With zero or negative winrate and positive SD, no finite bankroll achieves an eventual risk below 100% under its fixed-stakes assumptions.

Related Poker Calculators

• Bankroll and risk calculators hub — choose a tool for bankroll, variance or bet sizing.
• Poker risk of ruin tool — start with an existing bankroll and estimate its risk.
• Cash-game variance and downswing simulator — explore results over a stated number of hands.
• Rake and rakeback calculator — examine costs and net winrate.
• Poker rake analysis — compare rake pressure across games.
• MTT bankroll calculator — examine tournament scenarios with their own inputs.
• Poker equity calculator — estimate winning chances in a hand.
• Pot odds calculator — compare a call’s cost with the pot.
• ICM calculator — estimate tournament prize equity.

Frequently Asked Questions (FAQ)