Poker Bankroll for Live Cash Games
Live cash bankrolls work differently than online: softer games, higher rake, slower recovery. Here's how many buy-ins to keep, plus a per-stake table.
On this page · 7 sections
For live cash games, keep roughly 20–40 buy-ins for your stake. Live tables are softer than online, which supports the lower end of the range — but you play far fewer hands per hour, recovery from a downswing is slow, and the effective rake in small live games is brutal. Those forces pull the sensible number back up. Here’s how to size it and why live is its own animal.
Why live bankrolls follow different rules
The buy-in counts you see for online cash don’t transfer cleanly. Three things make live distinct:
- Fewer hands per hour. Live full-ring runs ~25–30 hands/hour; online multi-tabling can be 300+. Your win rate is measured per hour, and it takes far longer to realize your edge.
- Softer games. Casual live players call too much and bluff too little, so a solid regular’s edge is often larger — which supports a slightly smaller buy-in cushion.
- Higher effective rake. Small live games ($1/$2, $1/$3) often take a big fixed rake plus a jackpot drop and dealer tips out of small pots. That eats into win rate and, on the margin, argues for a bigger cushion.
The net effect: live players can run leaner in buy-ins than online grinders, but should never run thin, because a cold stretch takes real-world months to climb out of.
Buy-ins by live stake
Live buy-ins are quoted as a range (a min and a max), and most winning regulars buy in for the max. Use the max to size your roll:
| Live stake | Typical max buy-in | 25 buy-ins | Conservative (35 BI) |
|---|---|---|---|
| $1/$2 | $300 | $7,500 | $10,500 |
| $1/$3 | $500 | $12,500 | $17,500 |
| $2/$5 | $1,000 | $25,000 | $35,000 |
| $5/$10 | $2,000 | $50,000 | $70,000 |
If those totals look large, that’s the point — live buy-ins are big in dollar terms even when the buy-in count is modest. Money at these stakes must be truly separate from your living expenses.
Worked example: a live $2/$5 grinder
You want to play $2/$5 with a $1,000 max buy-in and you win at about $25/hour over a real sample.
- Bankroll target: 25 × $1,000 = $25,000 dedicated.
- Realistic downswing: even a strong live pro can run 15+ buy-ins below expectation — that’s −$15,000, or roughly 600 losing hours at your win rate. A 25-buy-in roll absorbs it with $10,000 to spare; a 12-buy-in roll would have you dropping down mid-swing.
- Move-down trigger: if the roll falls under 20 buy-ins ($20,000), drop to $1/$3 and rebuild — no ego.
The slow pace is the hidden tax here: 600 losing hours is a long time to sit stuck. The bankroll’s job is to keep that stretch from also being an existential one.
Live-only bankroll factors
A few things affect live rolls that never come up online:
- Tips and expenses. Dealer tips, food, parking, and travel are real costs that quietly lower your effective win rate. Track them.
- Table and game selection. Live edges come disproportionately from picking soft tables. Being willing to table-change is a bankroll tool. More on that in the cash game strategy hub.
- Time-based games. Some higher live games charge time (a set fee per half-hour) instead of raking pots, which changes the math for short sessions.
- Cash on hand. You need buy-ins physically available, not just a number in an app. Plan how you carry and store it safely.
Live vs. online in one glance
| Live cash | Online cash | |
|---|---|---|
| Hands per hour | ~25–30 | 60–300+ |
| Game softness | Softer on average | Tougher, especially higher |
| Buy-ins to keep | 20–40 | 20–30 |
| Recovery speed | Slow (months) | Faster (volume) |
| Hidden costs | Tips, travel, food | Site rake, few extras |
The buy-in counts are similar, but the experience of variance differs completely — online you grind through it, live you wait through it. That waiting is the part new live players underestimate: a cold month can feel like proof you’ve stopped winning when it’s really just a low-volume version of a routine downswing.
Common live bankroll mistakes
- Playing your whole roll on the table. Bringing every dollar you have to the casino turns one bad session into a bust. Bring a session bankroll — a few buy-ins — and leave the rest safely elsewhere.
- Treating a hot night as your win rate. One $2,000 upswing at $1/$2 does not mean you’ve mastered the stake. Live samples are tiny; judge yourself over months, not sessions.
- Ignoring the rake drag. In a $1/$2 game where $6 plus a jackpot drop leaves every small pot, your real edge is thinner than it feels. Factor it into both your win rate and your cushion.
- Refusing to drop down. Live egos run hot in a room full of familiar faces. Dropping from $2/$5 to $1/$3 during a downswing is correct math, not a public demotion.
Put it together
A live cash bankroll is 20–40 buy-ins of a stake you can genuinely beat, sized in real dollars because live buy-ins are large and recovery is slow. Lean on soft game selection to grow it, drop down the moment you dip below your threshold, and follow the same move-up discipline as any format — laid out in when to move up in stakes. For the format-agnostic buy-in math, start with how much bankroll you need, and see the full picture in the bankroll management hub.
Frequently asked
How much bankroll do I need for live poker?
Around 20–40 buy-ins for the stake you play. Live games are softer than online, which supports the lower end, but slower recovery and higher effective rake argue for keeping a solid cushion.
How much bankroll for live 1/2?
A live $1/$2 buy-in is typically $200–$300. At 25 buy-ins of a $300 max, that's roughly $7,500 dedicated to poker — money you can afford to have swing.
Is live poker lower variance than online?
Yes, usually. You play far fewer hands per hour and tables tend to be softer, so downswings measured in hands are smaller — but they take longer in real time because volume is low.
Do I need a smaller bankroll for live than online?
In buy-in terms, often slightly smaller because live games are softer. But live buy-ins are larger in dollars and recovery is slow, so the real-money figure can still be significant.