Table Stakes, Buy-Ins, and Rebuying

Assumptions: Unless a hand states otherwise, examples assume a 6-max online cash game at $0.50/$1 with 100 big blind stacks and no rake.

Before you play a single hand of cash poker, you make a decision that shapes every hand after it: how much money to put on the table. Most beginners click a default and never think about it again. That's a mistake — buy-in size determines what you can win, what you can lose, and even which starting hands are profitable. The rules around it are simple but strict, and one of them (no going south) carries real penalties if you break it live.

Table stakes: the money on the table is the whole game

Cash poker runs on the table stakes rule: only the chips in front of you when the hand is dealt can be bet, won, or lost in that hand. You can never reach into your pocket mid-hand, no matter what you're holding. The dramatic movie moment where someone slaps a watch on the table to call — illegal everywhere.

Table stakes cuts both ways, and both directions protect you:

• You can't lose more than what's on the table. If you sit with $100, your worst-case hand costs $100, period.
• You can't be bet off a hand for more than you have. If an opponent bets $500 and you only have $80, you can call all-in for $80 and contest a pot built from $80 of their money plus yours. (The side-pot mechanics were covered in the rules module.)

The flip side: you also can't win more from any opponent than you have in front of you. This is the concept of effective stacks — when a $100 stack plays against a $40 stack, only $40 per player is genuinely in play. Your extra $60 is just decoration for that hand. Remember this; it's the engine behind everything else in this lesson.

Minimum and maximum buy-ins

Cash tables post a buy-in window. Online at $0.50/$1, the typical window is 40bb to 100bb — $40 to $100. Some sites run "deep" tables (up to 200bb or 250bb) and some run "short" cap tables (max 40bb or 50bb), but 40bb–100bb is the standard you'll see most.

Always do this arithmetic in big blinds, not dollars, because big blinds are the unit every piece of strategy is written in: divide by the big blind. A $100 stack at $0.50/$1 is 100bb. The same $100 at a $1/$2 game is only 50bb — a medium stack, not a full one. Two players can have identical dollar amounts and completely different strategic situations because the blinds differ. When you move up stakes, your dollars shrink in bb terms even though the number on the screen looks the same.

Buy in for the full 100bb

Standard advice while you're learning, and this site's default assumption everywhere: buy in for the maximum, normally 100bb, and keep your stack near it. Three reasons:

1. All training material assumes it. Every range chart on this site is built for 100bb play. Sit with 40bb and the charts are quietly wrong for you — hands that rely on winning big pots later in the hand lose value, and all-in math changes.
2. Big hands need big stacks to get paid. When you flop a set against an overpair, your profit is capped by the effective stack. Deep, that's a full double-up; short, it's a nibble.
3. Your edge needs room to work. If you're learning to play well, you want the version of the game where skill — postflop play across three streets — matters most. Short stacks compress the game toward preflop all-ins, where there's less room to outplay anyone.

Short-stacking — deliberately buying in for the minimum, often 40bb or less — is a legitimate niche strategy: shorties shove early, deny opponents implied odds, and avoid hard postflop decisions. Some grinders did it professionally for years. But it's a different game with different charts, it wins less per hand even played perfectly, and it teaches you nothing about the full game. Learn 100bb poker first.

Worked example: $40 or $100 with 9♥9♦ waiting

You sit down at a $0.50/$1 6-max table, the client asks for $40–$100, and your first hand is going to be 9♥9♦ — a premium-ish pair you'd love to play a big pot with.

BTN

9♥♥9♥

9♦♦9♦

UTGunknown

BBunknown

1. 1.UTG raises to $2.50
2. 2.HJ folds, CO folds
3. 3.BTN (hero) calls $2.50
4. 4.

How strong is 9♥9♦? Against one random hand it wins about 72%. Against a classic raising hand like A♠K♣ it's a small favorite at about 56%. But against the hand you actually fear — pocket aces — it's crushed at about 19%. Those three numbers tell you how medium pairs make their money: not by racing big hands preflop, but by flopping a set and winning a huge pot from exactly the overpairs that dominate them before the flop. The 19%-vs-AA hand becomes a monster the times a nine hits the flop.

Now connect that to the buy-in. The set-over-overpair payday is worth whatever the effective stack is. Buy in for $100 against $100 stacks and the payday is up to 100bb. Buy in for $40 and the same magic flop pays at most 40bb, while the hand's costs — calling preflop raises, occasionally losing small pots — stay the same. You'd be keeping all of the price and amputating the prize. With a hand like 99 dealt first hand, the $100 buy-in is simply buying more upside on the same investment. Take the $100.

No going south: chips stay on the table

Once chips are on the table, they're committed to the game. The no going south rule (also called ratholing when someone does it) says: you may not remove chips or money from your stack while you remain in the game. Win a $300 pot at a live $1/$2 table and you cannot slip $200 into your pocket and keep playing with $100. Online, you can't reduce your stack at the table — the cashier won't let you.

The only way to take money off the table is to leave the game — stand up (or sit out and leave, online), cash out your whole stack, done. Most rooms also enforce a cooling-off period: leave a game with a big stack and you can't rejoin the same stakes minutes later with a short one. Live, ratholing draws floor intervention and, repeated, gets you barred; online, the software simply makes it impossible at the table.

Why does the rule exist? Fairness. Players who lost chips to you deserve a chance to win them back on equal terms. If winners could pocket profits mid-session while losers must keep full stacks at risk, the game becomes a one-way valve — everyone would lock up wins instantly, deep-stack poker would die, and the losing players funding the game would rightly feel hustled. Table stakes giveth (you can't lose your wallet) and table stakes taketh (your winnings stay in play).

Rebuying and topping up

The traffic flows freely in the other direction. Between hands, you may always add chips, subject to the table maximum:

• Rebuying usually means reloading after going broke or near-broke — busting your $100 and buying another $100.
• Topping up means adding partial amounts to return to full — turning $62 back into $100.

Both happen between hands only, never during one (table stakes again). Online clients have an "add chips" button and most offer an auto-top-up setting that refills you to your chosen amount every time you dip below it. Turn it on. Live, you slide cash to the dealer or buy chips from a chip runner between hands.

Worked example: sitting on 62bb after a lost pot

You bought in for $100, lost a chunky pot, and now have $62 — 62bb. The next hand is about to be dealt. Top up or not?

CO

A♠♠A♠

Q♠♠Q♠

BTNunknown

1. 1.Hero clicks Add Chips for $38 before the deal, restoring $100
2. 2.CO (hero) raises to $2.50
3. 3.BTN calls, blinds fold

Analysis

At 62bb, every all-in confrontation against the table's full stacks is capped at $62 — hero can never win more than 62bb in a hand no matter how perfect the setup. Adding $38 between hands is always legal and takes one click. If hero believes he has an edge in this game, playing it 38bb lighter than allowed just caps his best outcomes while leaving his blind costs identical.

The case for topping up is one sentence long: if the game is worth playing, it's worth playing with a full stack. At 62bb your downside per hand barely changes — disasters costing more than 62bb are rare — but your ceiling is chopped in every big-pot scenario, because effective stacks cap what full-stacked opponents can pay you. The moments that decide a session are precisely the ones where stacks go in, and you'd be showing up to them 38bb short.

There's a real exception worth naming honestly: if you've just lost a painful pot and you're rattled — tilted — the better move may be to sit out or quit, not to reload. Topping up is correct when your edge is intact. It's a leak when it's just feeding a bad session with fresh money. The rule of thumb: top up automatically when you're playing well; take a break when you're not. The software's auto-top-up handles the first case so you never play a hand short by accident.

Buy-in is not bankroll

A distinction that saves beginners real pain: the buy-in is what you put on this table tonight; your bankroll is the total fund you've set aside for poker, and the two should be wildly different sizes. Sitting at a $0.50/$1 game with your entire $100 poker fund as the buy-in means one normal cooler — aces into a flopped set, nothing you did wrong — ends your poker career until payday. Cash-game variance guarantees that even winning players regularly lose multiple buy-ins in a session, so the working rule of thumb is to keep a bankroll of at least 20–30 buy-ins for your stake and treat any single table's $100 as expendable ammunition, not the war chest. Bankroll management gets its own treatment later in this track; for now, the takeaway is simply that "buy in for 100bb" assumes the 100bb is money you can rebuy behind. If a max buy-in feels too big to lose twice in a night, the fix isn't a short stack — it's a smaller stake.

Live games run the same logic with different numbers: a typical $1/$2 room posts a $100–$300 buy-in window, so the "full" buy-in is $300 (150bb) and many regulars buy exactly that. The principle transfers unchanged — buy the maximum the game allows if your bankroll supports the stake, think in big blinds, and top up between hands just as you would online (slide cash or chips forward between deals; the dealer handles it).

Stack size changes which hands are profitable

A preview of an idea that grows throughout this curriculum: hands don't have fixed values — they have values at a stack depth.

• Deep stacks (100bb+) reward hands that can win huge pots when they hit hard: pocket pairs hunting sets, suited connectors, suited aces. These hands lose small pots often and win giant ones rarely; they need a giant available to win. The jargon is implied odds — money you expect to win on later streets.
• Short stacks (under ~40bb) kill implied odds and reward hands that are simply strong right now: big pairs and big-card hands like AK, AQ, KQ that win unimproved or make top pair. There's no "later" to get paid in.

That's the deepest reason the buy-in decision matters: it's not just bankroll logistics, it silently re-prices all 169 starting hands. The site's charts assume 100bb. Buy in for 100bb, keep yourself topped up, and the rest of your training will describe the game you're actually sitting in.