
Poker Bankroll Management for Live Players
The 20-40 buy-in convention and the three inputs that produce it: your edge, your format's swings, and what losing the roll would cost you.
Aug 23, 2026 · 6 min read · PokerInk Team
A bankroll is money set aside for poker only, sized so that a normal losing stretch cannot end your ability to play. For live cash games, the convention most players use is 20 to 40 buy-ins for your regular stake. For tournaments, the convention is far larger, 50 to 100 buy-ins or more, because tournament results swing harder.
Those ranges are not laws. They are outputs of three inputs, and once you see the inputs you know which end of the range is yours, and when the range itself is wrong for your situation.
One definition before anything else, because live games make it ambiguous: a buy-in is what you actually sit down with in your regular game. If your room's $2/$5 plays $1,000 deep and that is what you buy for, your buy-in is $1,000, and 30 of them is $30,000. Counting "buy-ins" against a number you never actually play is the quiet way to be badly underrolled.
Why a losing month proves nothing
Start with the uncomfortable fact the whole subject rests on. A winning player loses often. A good player is a favorite to book a win in any given session and still loses a large minority of them, and stacking sessions into a reliable profit takes longer than intuition expects; the road there includes losing weeks and losing months that mean nothing about your skill. Players call this variance. A bankroll is the tool that lets you keep playing through it.
The alternative is the pattern every card room regular has watched: a competent player runs normally bad for three weeks, is now playing with money that was supposed to be rent, tightens up into unprofitability, and disappears.
The three inputs that size a bankroll
Your edge. The bigger your win rate relative to the stakes, the faster you climb out of holes, and the fewer buy-ins you need behind you. A strong player in a soft game can live near the bottom of the conventional range. A thin winner needs the top of it. A losing player has no bankroll requirement, because no bankroll survives a negative win rate; the honest move at that point is moving down or studying, and the bankroll question waits.
The swinginess of your format. A passive full-ring game swings less than a lineup that raises and 3-bets everything. Big-bet games like PLO swing harder than no-limit hold'em and need a bigger multiple; applying hold'em buy-in counts to live PLO is a common way competent players go broke. Tournaments swing most of all, because the money concentrates in top finishes that any individual player rarely hits; this is why the tournament convention is 50 to 100+ buy-ins while cash sits at 20 to 40. Game selection moves this input and the last one at once: the same player has a different edge and different swings in the best game in the room than in the worst.
What losing it would cost you. A player with an income who can rebuild a bankroll in a few paychecks can run leaner. A professional whose bankroll is also the grocery money needs multiples of every number above. This input is about your life, and it outranks the other two.
The part most players skip
Every rule above consumes a number you probably do not have: your actual results. Not the sessions you remember, which skew toward the dramatic ones, but all of them. Memory keeps the $800 Friday and quietly drops the four $150 Tuesdays that paid for it, in either direction. We wrote a full article on why remembered win rates are wrong; the summary is that any stat built from hand-picked sessions flatters or terrifies, and both are expensive.
The fix is recording every session, buy-in and cash-out, before leaving the parking lot. A notebook does the job; we do it in PokerInk, which turns the complete record into dollars per hour. The tool matters less than the completeness. Be patient with what the numbers can tell you: 50 sessions gives a first honest read on your swings and the direction of your results, while an hourly you can actually lean on takes hundreds of hours to settle. The honest answer in the meantime is to size your bankroll as if your edge is smaller than you hope it is.
Working rules that follow
- Count buy-ins, not dollars. "I have 30 buy-ins at $300" transfers between stakes; "$9,000" does not.
- Move down before you have to. Pick the trigger in advance, at the bottom of the range you chose for yourself, and honor it. If your number was 30 buy-ins, you move down at 30, not after you are already playing scared at 12. Moving down feels like a demotion and works like a parachute.
- Move up on a schedule, not a heater. A shot at the next stake after your bankroll clears the top of your range at the new stake, with a pre-agreed retreat point, is a plan. Moving up because tonight went well is a tilt purchase.
- Keep poker money physically separate. A different account or a literal envelope. The moment poker money mixes with life money, every downswing becomes a decision about rent.
- Recount honestly after cashouts. Taking money out of the bankroll is fine, it is the point of winning, but the buy-in count you play by has to reflect what is actually there.
Honest bankroll rules need honest session records
PokerInk is free to start on iPhone, Android, and the web. See pricing.
Frequently asked questions
How many buy-ins do I need for live cash games?
The common convention is 20 to 40 buy-ins for your regular stake. Strong winners in soft games sit near 20; thin winners and anyone who cannot easily rebuild should sit near 40 or above. The range exists because the right number depends on your edge, your game's swinginess, and what losing the roll would cost you.
Do recreational players need a bankroll?
They need the separation more than the count: a fixed amount of money that is allowed to be lost at poker, kept apart from life money. The buy-in math matters less when a paycheck can refill the roll, but the wall between poker money and rent money matters for everyone.
What is variance in poker?
The natural swing between your results and your true win rate. Thin edges plus random cards mean even winning players have losing weeks and losing months. Variance is why bankrolls exist and why short-term results, good or bad, say almost nothing about skill.
When should I move up in stakes?
On a plan, when your bankroll clears your required buy-in count at the new stake, with a pre-set point at which you move back down. Moving up to chase losses or because one night went well is how bankrolls end.
How do I know my real win rate?
Record every session, buy-in and cash-out, with no exceptions. Compute dollars per hour and sessions won from the complete record. Memory-based estimates are biased toward memorable sessions, and any decision built on them inherits the bias. Even a complete record needs hundreds of hours before the hourly settles into something to lean on.



