Poker Bankroll Management Calculator
Uses the standard risk-of-ruin formula to find the largest cash-game buy-in your bankroll supports, given your win rate, standard deviation and the risk of going broke you will accept. Essential for cash-game grinders sizing stakes responsibly.
Last updated: September 2026
Formula below · 3 sources (omnicalculator.com, pokerstrategy.com, Wikipedia) · Updated Sep 2026
Compare with similar
About this calculator
Bankroll management balances your win rate against the natural variance of poker so that your probability of going broke (risk of ruin) stays at an acceptable level. Treating results as a random walk with drift (the diffusion model in Chen and Ankenman's The Mathematics of Poker), the risk of ruin with a bankroll of B big blinds is RoR = e^(−2 × winRate × B / SD²), where winRate and SD are both in big blinds per 100 hands. Solving for B gives the bankroll you need: B = SD² × ln(1 / RoR) / (2 × winRate) big blinds. This calculator divides your current bankroll by that requirement to get the largest big blind you can afford and reports it as a 100-big-blind buy-in: maxBuyIn = 200 × winRate × currentBankroll / (SD² × ln(1 / RoR)). Variables: currentBankroll (dollars dedicated to poker), winRate (BB/100), SD (standard deviation in BB/100, typically 80–120 for no-limit hold'em cash games), RoR (the chance of going broke you accept). A higher standard deviation or a lower risk target raises the number of buy-ins you need; a higher win rate lowers it. Because ln(1/RoR) is only 2.3 at 10% and 4.6 at 1%, cutting risk from 10% to 1% doubles the bankroll you need. Edge cases: with a zero or negative win rate there is no safe stake (ruin is certain in the long run), so the calculator returns a message instead of a number. The model assumes a stable win rate and normally distributed results, which fits cash games; tournament results are far more skewed, so use tournament buy-in guidelines instead. Win rates are estimated with large error, so treat the result as a ceiling, not a target.
How to use
Example 1: $2,000 bankroll, 5 BB/100 win rate, 80 BB/100 standard deviation, 5% risk of ruin. Step 1: ln(1/0.05) = 2.996. Step 2: bankroll needed = 80² × 2.996 / (2 × 5) = 6,400 × 2.996 / 10 ≈ 1,917 big blinds, or about 19.2 buy-ins of 100 BB. Step 3: largest big blind = $2,000 / 1,917 ≈ $1.04, so the largest buy-in is about $104.32. Play $0.50/$1 no-limit with a $100 buy-in. Example 2: $10,000 bankroll, 3 BB/100, SD 100 BB/100, 1% risk of ruin. Step 1: ln(1/0.01) = 4.605. Step 2: bankroll needed = 100² × 4.605 / (2 × 3) ≈ 7,675 big blinds (76.8 buy-ins). Step 3: largest buy-in = 100 × $10,000 / 7,675 ≈ $130.29, so $0.50/$1 ($100 buy-in) is the highest standard stake that fits.
Frequently asked questions
How many buy-ins do I need for cash games versus tournaments?
Cash players typically need 20–30 buy-ins for recreational play and 50+ buy-ins for full-time grinders, because cash game variance is moderate and outcomes converge faster. Tournaments have much higher variance due to winner-takes-most payout structures, so professional MTT players often carry 100–300 buy-ins. The calculator weights your standard deviation heavily — tournament players typically report 150 BB/100 or more, which automatically suggests larger bankrolls. Sit & Gos sit between these two extremes (50–100 BI). Always err on the side of more buy-ins when moving up stakes, since the player pool gets tougher at higher levels and your effective win rate usually drops.
What win rate and standard deviation should I enter if I am a new player?
With fewer than 50,000 hands your observed win rate is not yet statistically reliable, so use a conservative estimate like 2–3 BB/100 rather than your tracked number. Standard deviation for a typical NLHE player is 80–120 BB/100 for cash games and 150–200 BB/100 for MTTs. Tracking software like PokerTracker 4 or Hold'em Manager 3 reports both metrics automatically once you import a meaningful hand history. Starting with conservative inputs protects your bankroll while your sample size grows toward statistical significance. Update the inputs as your sample grows past 100,000 hands.
What is an acceptable risk of ruin percentage?
Most professionals target 1–5% risk of ruin — a 1–5% chance of busting their entire bankroll before moving down stakes. Recreational players who can reload from a day job may tolerate 10–20%, since the financial consequences of going broke are smaller. A 5% target is a common baseline: strong protection without requiring an absurdly large bankroll. Lower targets (1% or below) dramatically increase the required buy-ins because the logarithmic relationship makes bankroll demand steepen quickly. The right answer also depends on your psychology — many players cannot stomach the swings of 5%-risk staking and prefer 1% even at the cost of slower growth.
What are common mistakes when applying bankroll management math?
Using an inflated win rate from a hot streak (e.g., +10 BB/100 over 5,000 hands) instead of a regressed long-term estimate makes recommended stakes look much higher than is safe. Underestimating standard deviation — entering 60 when your true value is 100 — produces dangerous under-rolled recommendations. Ignoring rake at micro-stakes overstates your effective win rate by 4–8 BB/100. Treating tournament and cash bankrolls as fungible without re-running the calculation for each game type leads to over-confidence; the formulas weight variance differently. Forgetting to subtract life expenses (rent, food, taxes) from a dedicated bankroll leaves you bankroll-poor when bills arrive. Finally, the BB/100 metric assumes 100 hands per session unit — at full-ring live poker with 30 hands/hour, results take many more hours to converge.
When should I NOT use this bankroll management formula?
Recreational players whose poker stake is a small fraction of total wealth (under 5% of net worth) can ignore bankroll math entirely — they should focus on enjoyment and table selection. The risk-of-ruin formula used here assumes your win rate is known exactly; because real win rates are uncertain, many professionals add a safety margin on top of its answer. Players running a backed or stake arrangement use entirely different math — their effective bankroll is the backer's, not their own. Live tournament players with a high proportion of $10k+ buy-ins should use tournament buy-in guidelines rather than this formula, because tournament results are too skewed for the normal-distribution model it assumes. Negative-win-rate players cannot solve their bankroll problem with sizing — work on the game first. Finally, in jurisdictions where you cannot legally withdraw from your account quickly, treat the 'bankroll' as illiquid and demand a larger buffer.