Poker is the only game in the world where the cards are almost secondary. Chess has the board. Blackjack has the chart. But poker — particularly Texas Hold'em — is a game of incomplete information played against incomplete human beings, and the human element is where the game lives. When you add a live camera to that equation, you get something genuinely rare: a competition where your face, your timing, your micro-expressions, and your composure are all live data points that your opponent can read in real time.
That's a very different context from an online poker room where you're a username behind a digital chip stack. On Shitbox Shuffle, when you raise preflop, the person who has to decide whether to call or fold can see your face while they think. They can watch you watching them. The psychological depth that makes poker fascinating in person translates to the video format more faithfully than any other card game, because poker's core challenge has never been about the cards — it's always been about the people.
This guide covers everything you need to engage competently with Texas Hold'em on video: the complete hand rankings from high card to royal flush, the full rule structure from preflop through the river, positional theory, starting hand selection, pot odds, bluffing in a live-camera context, bet sizing, and bankroll management for token play. Whether you're new to the game or returning after years away, the goal is to give you a foundation solid enough to play with genuine intent — and to understand what you're up against when you're matched with someone who has done the same work.
Poker rewards study faster than almost any other skill game. The foundational concepts are graspable in hours. The application of those concepts under live pressure — the actual playing — takes longer. But the gap between a player who has read this material and one who hasn't is enormous and shows up immediately in every single hand.
Before anything else about strategy matters, you need to know what beats what. Hand rankings in poker are fixed and universal — they don't change by game type, position, or table. A flush always beats a straight. A full house always beats a flush. These relationships are the grammar of every hand you'll ever play.
Several ranking relationships trip up new players. Flushes beat straights — five of the same suit is rarer than five in sequence. Full houses beat flushes — having three of a kind plus a pair is a stronger combination than all five cards sharing a suit. And within ranks, ties are broken hierarchically: two players with two pair compare their top pairs first, then bottom pairs, then the fifth card (kicker) if needed.
The ace plays a unique role in straights: it can be the highest card (10-J-Q-K-A) or the lowest (A-2-3-4-5, called a "wheel" or "bicycle"). It cannot wrap around — Q-K-A-2-3 is not a valid straight. In Hold'em, the kicker is often decisive — a player with A-K and a player with A-Q on an A-J-7 board both have top pair top kicker versus top pair second kicker, and the K beats the Q in a showdown.
In Texas Hold'em, you always use the best possible five-card hand from the seven available (two hole cards plus five community cards). You are never required to use both hole cards — you can use both, one, or technically neither (though playing the board is rare and usually not a winning proposition).
Texas Hold'em is structurally simple. The complexity emerges from the decisions, not the mechanics. A complete hand moves through four distinct stages, each with a betting round, and each stage revealing more information that changes what any given hand is worth.
Before any cards are dealt, two players post forced bets called blinds. The player immediately to the left of the dealer button posts the small blind (typically half the minimum bet), and the player two to the left posts the big blind (the full minimum bet). These forced bets create a starting pot that incentivizes play — without them, every player could simply fold until they receive a premium hand.
The dealer button rotates clockwise after every hand, ensuring that all players post blinds over time and that positional advantages cycle through the table. In a heads-up (two-player) game, the button is also the small blind and acts first preflop but last on all subsequent streets.
Each player receives two private cards (hole cards) dealt face-down. The player to the left of the big blind acts first — they can call the big blind amount, raise to a higher amount, or fold. Action moves clockwise. The small blind and big blind act last, with the big blind having the option to raise even if no one else has raised (this is called the "big blind option").
Preflop decisions are based entirely on your two hole cards and your position relative to the button. This is the moment to apply starting hand selection — which hands are worth playing, from which positions, and at what bet sizing. We'll cover that in detail in the starting hands section.
After preflop betting, three community cards are dealt face-up in the center of the table — this is the flop. These three cards are shared by all remaining players and are used in combination with hole cards to form hands. A second betting round occurs. The player closest to the left of the button who is still in the hand acts first — this order (first to act is earliest remaining position) holds for all post-flop streets.
The flop dramatically narrows the range of possible winning hands. After the flop, you have seen five of the seven cards that will determine your best hand (two hole cards plus three community cards). The fundamental questions at the flop are: do I have a made hand, a drawing hand, or air (nothing)? And how does my hand range interact with this board?
A fourth community card is dealt face-up — the turn. A third betting round follows. The turn often increases the size of bets, especially in limit Hold'em formats where bet sizes double on the turn. In no-limit and pot-limit formats, the practical effect is psychological — the pot is larger, so the stakes of each decision are higher. Drawing hands get closer to completing or failing: with one card to come, you can calculate your exact probability of hitting your draw.
The fifth and final community card — the river — is dealt face-up. The final betting round occurs. There are no more cards to come, no more draws to consider. The only questions remaining are: what is my best five-card hand, what does my opponent likely hold, and is this situation one to bet, call, raise, or fold?
If multiple players remain after river betting, a showdown occurs — all remaining players reveal their hands and the best five-card combination wins the pot. In many hands, showdowns never happen because one player's bet causes all opponents to fold. Winning without showdown — by making your opponents fold — is a core element of poker strategy and where bluffing lives.
At each betting round, the available actions are: Fold (discard your hand, lose any chips invested), Check (pass the action to the next player, available only when no bet has been made yet in the round), Call (match the current bet), Bet (place the first bet of the round), and Raise (increase the current bet). In no-limit Hold'em, bets and raises can be any amount up to your entire stack — this is the feature that makes the game so psychologically complex.
Position is the most underappreciated concept in Texas Hold'em for new players, and the most consistently important concept for experienced ones. Position refers to where you sit relative to the dealer button, which determines when you act in each betting round. Acting last — having already seen what every other player does before you decide — is a fundamental structural advantage that never goes away.
When you act last on the flop, turn, and river, you possess a stream of information your opponents don't have: you already know what every player before you chose to do. A check from the player to your right tells you something about their hand strength — it eliminates many strong made hands from their range. A bet tells you something else. You process all of this before you decide, giving you a structural advantage on every single street regardless of your cards.
This information advantage translates directly to money over time. The dealer button is the most profitable seat at any poker table, measured in big blinds won per hundred hands. Early position (UTG and UTG+1) is consistently the least profitable because you act first with the least information. This is why starting hand requirements are tighter from early position — you're committing chips with no information about how anyone else will play the hand.
In late position (button and cutoff), you can profitably play a wider range of hands because your positional advantage compensates for moderate hand strength. A hand like 7-8 suited is borderline from UTG but genuinely playable from the button, where its implied odds (the potential to extract large bets when you hit your draw) are much higher because you control the action on every street.
Position also makes bluffing dramatically more feasible. A bet from the button after all other players check is a credible story — you've seen everyone pass on aggression, suggesting weak holdings across the board. A bet from UTG into multiple unchecked players tells a much weaker story because you don't know yet whether anyone behind you is trapping with a strong hand.
One of the most consequential decisions in Texas Hold'em happens before you see a single community card: choosing which starting hands to play. Not every two-card combination is worth putting chips into the pot. Understanding hand tiers — which hands are premium, which are playable in certain conditions, and which are trash regardless of circumstances — is the foundation of preflop strategy.
These hands are strong enough to raise from any position at the table and to continue aggressively through most board textures.
These hands are profitable to play in the right conditions — primarily with position advantage, or against opponents who play too loose and will pay off strong hands.
Most starting hands in Hold'em are losing hands on average. Hands like 7-2 offsuit, 8-3 offsuit, Q-4 offsuit — combinations with no connected value, no flush potential, and no high-card strength — should be folded preflop in virtually all situations. The discipline to fold unremarkable hands is what separates recreational players from competent ones. Tight preflop play is not boring — it's the foundation that makes every subsequent decision cleaner.
Pot odds are the ratio between the size of the pot and the cost of a call. They're the tool that tells you whether calling a bet is mathematically profitable based on your estimated probability of winning the hand. Understanding pot odds doesn't require complex math — it requires a simple comparison of two numbers.
Here's the calculation: if the pot contains 100 tokens and your opponent bets 50, the total pot you'd be winning is 150 tokens, and your call costs 50. Your pot odds are 150:50, or 3:1. To break even on this call, you need to win at least 1 in every 4 times you're in this situation — 25% of the time. If your hand or draw wins more than 25% of the time, calling is profitable. If it wins less, folding is correct.
Your "equity" — the probability of your hand winning — comes from estimating your outs. An out is any card that completes your drawing hand. If you have four cards to a flush, you have nine outs (13 cards of that suit minus four you've seen). A quick estimation rule: multiply your outs by 4 on the flop (for two cards remaining) or by 2 on the turn (for one card remaining) to get your approximate equity percentage. Nine flush outs on the flop gives you roughly 36% equity — a profitable call in many pot odds situations.
Applied examples: facing a 50% pot bet when you have a flush draw gives you about 2:1 pot odds (need to win 33%) against roughly 36% equity on the flop — a profitable call. Facing a pot-sized bet with the same flush draw gives you 2:1 odds (need 33%) against 36% equity — still marginally profitable, but only barely. Facing a two-times-pot overbet against a flush draw gives you 1.5:1 odds (need 40%) against 36% equity — now it's a fold.
Pot odds thinking becomes more sophisticated when you incorporate implied odds — the additional chips you can expect to win on later streets if you hit your draw. A flush draw against a deep-stacked opponent who has shown strength is worth calling even with slightly unfavorable direct pot odds, because hitting the flush will likely produce a large additional bet or call on the next street. Implied odds justify looser calls in position; they justify tighter calls out of position where you're less likely to extract value when you hit.
Bluffing in poker is not deception in the ordinary sense — it is a rational strategy in a game defined by incomplete information. When the math says your hand is beaten, the only way to win is to make your opponent believe otherwise. The credibility of your bluff depends on two things: the story your betting tells about the hand you're representing, and the consistency of your behavior across bluffing and value-betting situations.
Video chat poker is a uniquely interesting bluffing environment because both players are visible. Every micro-expression, posture shift, and timing pattern is potentially observable — and potentially observable to you in your opponent as well. This creates a game of mutual observation that doesn't exist in online poker rooms.
The most important part of bluffing on video is managing your own observable behavior. Common tells in live poker — averting eye contact, swallowing, touching the face, changing breathing rate, speeding up or slowing down decision timing — translate directly to a video camera. If you bet quickly when bluffing and slowly when value-betting (or vice versa), a perceptive opponent will eventually notice the pattern.
The practical prescription is behavioral consistency: make all your decisions in roughly the same timeframe regardless of hand strength. Before betting or raising, take a consistent deliberation pause — two to three seconds of apparent thought — whether you have the nuts or nothing. This flattens the timing tells that observant opponents exploit. Your facial expression should aim for neutral engagement — not blank and robotic, which reads as studied performance, but the ordinary focused expression of someone thinking about a math problem.
What you want to avoid is the two extremes: performing confidence (the false bravado of a weak bluffer) and performing anxiety (the visible discomfort of someone doing something they're unsure of). Genuine confidence comes from knowing why you're making the play — bluffing because the board texture supports the story you're telling, your betting has been consistent, and you've identified a situation where your opponent is likely to fold the range of hands they're holding.
In the other direction, you have a continuous stream of information from your opponent's camera. The key is not to focus on individual physical tells — which can be misleading and is a skill that takes considerable live play to develop — but on patterns of behavior over multiple hands.
Timing tells are the most reliable because they're hard to suppress consistently. A player who takes significantly longer to call than to fold, or who acts very quickly when they're strong and deliberates when they're weak, is broadcasting information in both directions. If you notice an opponent consistently delays before big river bets, the delay itself may be meaningful. Equally, a player who acts at roughly the same speed regardless of hand strength is harder to read and likely more experienced.
Verbal behavior is also available in video chat and genuinely informative. Players who talk more when they're bluffing and go quiet when they're strong (or the reverse) are patterning in ways you can track. The specific content of what people say is less important than the correlation between verbosity and hand strength — once you notice a pattern, it has predictive value.
"The best bluff doesn't require reading anyone's face. It tells a story so coherent, so consistent with everything that came before, that folding is the only logical response."
Effective bluffing is not a willingness to bet with nothing — it's a disciplined selection of situations where betting with nothing has a higher expected value than checking and folding. Those situations share common characteristics: your opponent has a range full of hands that can't call a bet (missed draws, weak pairs, overcards with no pair), the board texture favors your perceived range (e.g., you raised preflop and the board came K-K-2, which connects with hands like A-K that you'd raise with), and the pot size makes the bluff economically sensible relative to your stack and theirs.
Semi-bluffs — bets made with drawing hands that aren't currently the best but could improve — are generally preferable to pure bluffs because you have equity as a backup. A flush draw bluff that gets called still wins 36% of the time. A pure air bluff that gets called has zero equity. Semi-bluffing correctly is one of the highest-frequency profitable plays available in Texas Hold'em.
Bet sizing in no-limit Hold'em is a form of communication. Every bet size tells your opponent something about your range of hands — or, if you're skilled, it tells them something deliberately misleading. Getting bet sizing right serves two goals: extracting maximum value from strong hands and minimizing losses with weak or drawing hands, while maintaining enough ambiguity that your bet size doesn't immediately reveal your hand category.
A standard preflop open raise is 2.5 to 3 times the big blind from late position, increasing slightly to 3–4x from early position to compensate for the positional disadvantage. Re-raises (three-bets) are typically 3 times the initial raiser's bet, or 2.5 times if you're in position. These sizings are common enough that they don't give away hand information while still building a pot proportionate to hand strength.
One of the biggest preflop mistakes is raising larger with strong hands (trying to "protect" them) and smaller with weak or bluffing hands. This makes your sizing a direct tell. Your open-raise with pocket aces should look identical to your open-raise with 7-8 suited from the button. Balanced sizing across your range is more important than any individual hand-optimal sizing.
Three sizing zones serve most post-flop situations well. A small bet of 25–40% of the pot works on dry boards (boards with few draws, like K-7-2 rainbow) where you want cheap information or want to keep in hands that have little equity against you. A medium bet of 50–70% of the pot is the default continuation bet — large enough to put pressure on drawing hands and hands with no pair, small enough to not overcommit on boards that might have hit your opponent. A large bet of 80–100%+ is appropriate on wet boards with many draws, where you want to price out flush and straight draws and signal genuine strength.
On the river, where no more cards are coming and there are no draws remaining, sizing shifts. Larger river bets (75–125% of pot) are appropriate both for maximum value with strong hands and for bluffs on boards where draws missed — in both cases you want the bet to be large enough to produce a decision with real consequences. Small river bets often look like blocking bets from players with medium-strength hands who don't want to face a raise.
The principle underlying all sizing decisions: your bet should accomplish a goal. Value bets want calls from worse hands. Bluffs want folds. Protection bets want to charge draws. Know your goal before you size, and let the goal determine the size — not the size of the pot or the magnitude of your hand in isolation.
Poker variance is substantial — more so than blackjack — because the skill advantage needs many hands to express itself, and in any individual session, even a significant skill edge can be erased by bad luck in a few key hands. Managing your token bankroll intelligently is not optional — it's part of competent play.
The standard bankroll guidance for cash game poker is 20 to 30 buy-ins for your chosen stake level. A buy-in is typically 100 big blinds. If you're playing in a 10-token big blind game, a full buy-in is 1,000 tokens, and a responsible session bankroll to sustain that stake level comfortably would be 20,000 to 30,000 tokens. This cushion allows you to absorb losing sessions — which are inevitable — without being forced to move down in stakes or quit the game entirely.
For casual token wagering on Shitbox Shuffle, where the social and entertainment context differs from a grinding poker session, a more modest framing is appropriate: bring 5 to 10 buy-ins to any stake level you choose, and set a session loss limit of 2–3 buy-ins. If you lose that in a session, the cards aren't running in your favor today. Stopping prevents the emotional escalation that often accompanies continued play after a bad run.
The most bankroll-destructive single action in poker is the emotional all-in — going all-in not because the math supports it but because you're frustrated, because you want to recover losses quickly, or because the session has been running bad and you want to "let it ride." This decision-making pattern turns manageable losses into catastrophic ones. No individual hand is worth your entire stack unless the math — stack depths, pot odds, equity — genuinely supports it.
Recognizing tilt is easier said than done in a live video context, because the social pressure of being on camera creates additional emotion. If you catch yourself making decisions for emotional rather than strategic reasons — calling a big bet because folding "looks weak," bluffing off chips because you've lost three hands in a row and want to assert control — stop the session. The tokens will be there tomorrow.
Most beginner mistakes share a root cause: playing too many hands, in too many positions, with too little regard for relative hand strength. The corrections below address the most costly specific patterns.
There is a reason poker has been played for money between strangers for over two hundred years, and it has very little to do with the money. Poker is an unusually accurate character test. The pressures of the game — uncertainty, variance, loss, the temptation to deceive, the discipline required to fold — expose patterns of behavior that normal social interaction carefully conceals.
In a one-hour poker session with a stranger, you will learn more about how they handle adversity than you would learn in a week of casual conversation. You'll see whether they maintain composure under sustained pressure or whether they chase and escalate. You'll see whether they respect their own analysis or whether they abandon it the moment it becomes emotionally inconvenient. You'll see whether they're gracious or resentful when the cards go against them.
This is what makes video chat poker uniquely compelling. The camera captures the person, not just the player. When your opponent takes a bad beat — the kind of loss where the math was in their favor and luck intervened — their reaction in the seconds afterward tells you something true. Not performed, not curated, just real. And you're watching it on screen, in real time, knowing that they're watching you just as closely.
Playing well in this context means playing the person as much as the cards. Reading their behavioral patterns across hands, calibrating the credibility of their bets against how they've played before, and understanding that their decision-making process is as much an emotional journey as a strategic one. The best video chat poker players are good at poker, yes — but they're also genuinely curious about the person on the other side of the screen. That curiosity gives them information. Information gives them edge.
Pot odds are the single most important mathematical concept in poker — not because they're complex, but because they convert every "should I call this?" situation from a gut feeling into an arithmetic problem. Once you can calculate pot odds automatically, a significant category of mistakes disappears from your game entirely, because the math either justifies the call or it doesn't.
The calculation is simple: pot odds are the ratio of the pot size to the cost of the call. If the pot contains 100 tokens and your opponent bets 50, the total pot you'd be calling into is 150 tokens, and it costs 50 to call — so your pot odds are 3:1, or 25%. That means you need to win this hand at least 25% of the time to make calling profitable in the long run.
Outs are the cards remaining in the deck that will complete your drawing hand. If you have four cards to a flush after the flop, there are nine flush cards remaining in the deck — so you have nine outs. On the flop (with two cards still to come), multiply your outs by four to get a rough equity percentage: nine outs × 4 = 36% chance of completing the flush by the river. On the turn (with one card to come), multiply by two: nine outs × 2 = 18%.
Compare that equity to your pot odds requirement: if you need 25% equity to call and you have 36%, the call is profitable. If you need 33% equity and you only have 18%, you fold — the math doesn't support continuing regardless of how much you want to hit the flush.
| Drawing Hand | Outs | Flop Equity (%) | Turn Equity (%) |
|---|---|---|---|
| Open-ended straight flush draw | 15 | 54% | 30% |
| Flush draw | 9 | 36% | 18% |
| Open-ended straight draw | 8 | 32% | 16% |
| Two overcards | 6 | 24% | 12% |
| Gutshot straight draw | 4 | 16% | 8% |
| One overcard | 3 | 12% | 6% |
Implied odds extend pot odds by accounting for the tokens you expect to win on future betting streets if you complete your draw. If a flush draw is slightly unprofitable based on current pot odds alone — say you need 25% equity and you have 18% on the turn — but hitting the flush will almost certainly win you a large river bet from your opponent, the actual expected value of calling is higher than the current pot odds suggest. Implied odds are highest against opponents who tend to stack off with strong second-best hands (like sets or top two pair) when flush and straight cards complete on the river.
This is one of the most counterintuitive truths in poker for beginners: position — where you sit relative to the dealer button — has more impact on your long-run profitability than most starting hand decisions. Professional poker players exploit positional advantage so systematically that they can profitably play far weaker hands from the button than they can from early position. Understanding why explains the entire architecture of poker strategy.
Position matters because poker is an information game. Acting last in a betting round means you have seen every other player's decision before making your own. You know whether the flop was checked around (weakness) or bet into (strength). You know whether your continuation bet was called or raised. This informational advantage compounds across every street — preflop, flop, turn, and river — meaning a late-position player makes better-informed decisions on four separate occasions per hand compared to an early-position player who is betting blind into the unknown.
In late position (button or cutoff), you can profitably open a wider range of starting hands, because you'll have the informational advantage on every postflop street. Hands like suited connectors, small pocket pairs, and suited aces that are losing propositions from early position become playable from the button. You can also make more profitable bluffs from late position — you've seen the weakness signals before betting, rather than stabbing blind.
In early position, tighten your opening range to primarily premium hands and strong broadway combinations. You'll be acting first on every postflop street against the remaining players, which means you'll frequently be betting without knowing whether you're ahead. The compensation for this disadvantage is playing only hands strong enough to withstand that uncertainty.
One of the most important skills in poker — arguably more important than memorizing hand rankings — is learning to read what betting patterns tell you about your opponent's hand strength. This isn't mystical. It's pattern recognition built on a simple foundation: most players, especially beginners, bet in ways that correlate with their hand strength. They bet large when they have something and bet small or check when they don't. Learning to read these signals doesn't require years of experience — it requires attentiveness to the action from hand one.
In video chat poker specifically, you have an additional layer that's unavailable in online text-only poker: the camera. You can watch your opponent's physical responses to community cards, their hesitation before betting, their body language when they receive their hole cards. These are additional data points — not definitive, but informative — on top of the betting pattern information the game itself provides.
| Betting Pattern | Likely Interpretation | How to Respond |
|---|---|---|
| Bet → Check → Bet (flop bet, turn check, river bet) | Often a polarized range: strong hand or missed draw that turned into a river bluff | Consider pot odds carefully; if the river card completed obvious draws, the bet is more likely a value hand |
| Bet large on every street without hesitation | Usually strong hand value-betting for maximum extraction | Give significant credit; need strong hand or strong pot odds to continue |
| Preflop limp, then call raises passively | Weak holdings or speculative hands (suited connectors, small pairs) hoping to hit cheap | Continuation bet frequently on dry boards; they often miss and fold |
| Check-raise after initially checking | Strong hand (set, two pair, top pair with strong kicker) or aggressive bluff | Respect this line heavily early in a session; beginners rarely bluff-check-raise |
| Minimum bet (betting the smallest allowable amount) | Testing the waters, often with a medium-strength hand; sometimes a value hand that doesn't know how to size | Raise to get information or take down the pot; min-bets rarely represent the nuts |
| Instant call without any hesitation | Drawing hand with adequate pot odds, or medium pair that doesn't want to raise but can't fold | Be more cautious on completing draw cards; continue to value-bet if draw misses |
Watch every hand you're not involved in with the same attention you give hands you're playing. Most players check out when they fold. This is one of the most expensive mistakes in poker: the hands you're not playing are your primary source of information about how your opponents think and bet. When you see someone bet big on the flop and get called, then check the turn — note it. When you see someone limp in four consecutive hands from early position — note it. You're building a behavioral model that pays dividends in every hand you play against them.
Poker is played by human beings using a human brain — which means all of the cognitive vulnerabilities that affect every other high-stakes decision-making context affect poker as well. The difference between a technically skilled player and a winning player is often not the quality of their strategy but the quality of their mental state while executing it.
Tilt is the state of playing emotionally rather than strategically, typically triggered by a bad beat (losing a hand you had every right to win), a series of coolers, or a perceived slight from an opponent. The physical sensation of tilt — the tightness in your chest, the urgency to recover losses immediately, the frustration at someone who played poorly and got lucky — is real, but the decisions it produces are reliably worse than your baseline. Recognizing tilt as a state rather than a justified reaction is the beginning of managing it. When you notice yourself making a bet motivated by frustration rather than calculation, that's the moment to stop and reset.
Decision fatigue is less discussed but equally destructive. After playing for two or three hours, the quality of decisions measurably declines — not because you know less, but because the brain's capacity for effortful deliberation has been taxed. The solution isn't willpower; it's time limits. Set a session length in advance and stick to it. The last hour of a four-hour session is almost always the worst poker you play all day.
Patience is not passive — it's active. Patient poker means folding 70–80% of starting hands without frustration, because you understand that selective engagement is the structural edge. Every hand you fold correctly is a small win: you preserved your stack for the spots where you have genuine advantage. Players who experience folding as losing ground rather than choosing good spots are fighting against the basic math of the game.
Poker bankroll management is more demanding than bankroll management for most other casino games because poker is a game of skill played against other players — meaning your results include both variance (luck) and edge (skill). A player with a genuine edge can still lose over significant sample sizes due to variance alone. Bankroll management is the discipline that keeps you in action long enough for your edge to express itself.
The standard guidance for cash game poker: maintain a bankroll of at least 20–30 buy-ins for the stakes you're playing. This means if you're playing at a level where the standard buy-in is 100 tokens, your total playing bankroll should be 2,000–3,000 tokens before you sit at that table regularly. This cushion exists specifically to absorb the variance that even skilled players experience over hundreds of hands.
In blackjack with basic strategy, the variance per session is relatively predictable because you're playing against a fixed mathematical house edge. In poker, you're playing against humans whose skill levels vary wildly, and the single-hand variance can be enormous — even with 80% equity going all-in preflop, you lose 20% of the time. Over a session, that variance stacks. A 20-buy-in downswing is statistically possible for a genuinely winning player. A player with insufficient bankroll will go broke before that variance resolves itself, even if their edge over the field is real.