How to win at poker: what winning means and how to get there

You win at poker over thousands of hands, not one session, and almost all of that profit comes from other players’ mistakes. So the recipe is short: play in games where opponents make more mistakes than you, play fewer hands but play them aggressively and in position, bet for value far more than you bluff, keep the rake you pay low, and never play with money you can’t afford to lose.

Each of those points hides real detail, and each has its own guide on this site. This page is the map: what “winning” means in numbers, which habits matter most, and how to tell skill from a lucky month. We’ll link the deeper guides as we go rather than repeat them.

What winning at poker actually means

Cash game players measure results in big blinds won per 100 hands, written bb/100. A player who wins 5 bb/100 at a $0.25/$0.50 game makes $2.50 for every 100 hands on average.

Tournament players use ROI: profit as a share of buy-ins, so a 20% ROI means $20 profit per $100 in buy-ins. Our guide to poker winrate shows how to calculate both. Measure both after rake and before rakeback: rake is a cost you can’t avoid, while rakeback is a bonus that may change when the site changes its program.

The catch is how noisy poker results are. Over 100 hands, a typical six-max No-Limit Hold’em player’s result lands around 90 big blinds above or below their true win rate.

That spread is called the standard deviation; 90 bb/100 is the assumption we use below, and your own figure depends on your style and game. At $0.25/$0.50 it means swings of about $45 either way around an expected $2.50. Even a solid winner can be behind for a long time, and the chart below shows how long.

So a 5 bb/100 winner still has more than a one-in-four chance of being down after 10,000 hands. A few thousand hands tells you almost nothing about your skill, good or bad. Tournaments are noisier still: a player with a 20% ROI can go dozens of events without a single cash. That’s why winning players judge decisions, not results; our guide to variance shows how big the swings get.

Ten thousand hands sounds like a lot. Online at a single table it’s well over 100 hours of play; live, at about 25–30 hands an hour, it’s over 300 hours.

Choose games you can beat

Your profit is the gap between your skill and your opponents’ skill, minus the rake. Moving to a game with weaker opponents widens that gap faster than any new move you could learn. Strong players who sit only with stronger players lose.

You can spot a soft table in a few orbits:

• Many players see the flop, often by calling instead of raising (limping).
• The average pot is large for the stakes, a figure many online lobbies show.
• Players call big bets with weak pairs and rarely fold once they’ve put money in.
• Showdowns reveal hands that should have been folded before the flop.

Online, the lobby helps: sort tables by average pot or players per flop, join waiting lists for the tables that look softest, and leave when the weak players leave. Live, ask the floor for a table change if your table is full of regulars.

The site matters too. Sites differ in rake, rakeback and who plays there: some cater to recreational players with fast-fold games and no tracking software allowed, others to regulars. Our reviews of GGPoker, PokerStars, ACR Poker and CoinPoker compare those details.

Leave tables where everyone plays tight and aggressive. Winning players keep a list of opponents who give money away and look for their tables first; an hour against them is worth more than a week of study at a tough table.

Play fewer hands, and play them aggressively

Most losing players play too many hands and play them passively: they call before the flop with weak offsuit hands, then call again hoping to hit. Winning players fold most hands, and when they enter a pot they usually raise.

Raising gives you two ways to win: the best hand at showdown, or everyone folding. Calling gives you only one. Opening with a raise instead of limping also builds a pot when you have the stronger range and takes the initiative into the flop. Our starting hands chart shows which hands to open from each seat at six-max and full ring.

There are exceptions. Completing from the small blind can be fine in some live games, and over-limping behind several limpers with small pairs and suited connectors makes sense when the pot is multiway and opponents pay off big hands. But “raise or fold” is the right default.

Sizing matters as well. Online, opening raises of 2–2.5 big blinds are standard; live, where players call more, 3–4 big blinds plus one more for each limper is common. When someone has already raised, re-raising (a 3-bet) with your strongest hands and a few good bluffs usually beats calling: it takes the pot down often and gives you the lead when called. Calling a raise out of position with a hand like K♣ J♦ is one of the most common small leaks, because you’ll often flop a hand that’s hard to continue with.

Use position

The player who acts last after the flop sees everyone else’s decision first. That makes every decision easier: you get free cards when opponents check, you bet when they show weakness and you fold cheaply when they show strength.

That’s why the button is the most profitable seat at the table and the blinds lose money for almost every player. Play more hands from the cutoff and button, fewer from early seats. When you can choose a seat, sit to the left of the loosest players, so they act before you and you raise or isolate them with position. The detail by seat is in our guide to poker positions.

Bet for value more than you bluff

At low and mid stakes, value bets win far more than bluffs. A value bet is a bet made because worse hands will call. Many players at small stakes call too often, which means bluffing them fails and betting good hands pays.

Bluffs still have a place, mainly against opponents who can fold and on boards where your story is believable. Our guides to value betting and bluffing cover sizing and frequency. Against players who never fold, the rule is simple: bet your good hands bigger and stop bluffing. The full method is in how to play against calling stations.

Fold to big raises from passive players

The other side of value betting is knowing when your good hand is no longer good. At low stakes, a player who has only called all hand and suddenly raises the river almost always has a strong hand. Many players can’t fold top pair there and give back what they won elsewhere.

The price looks tempting. Say the pot is $20, you bet $10 on the river and a passive opponent raises to $35. You call $25 to win a final pot of $90, so you need to be ahead about 28% of the time.

Against a player who raises the river only with two pair or better, one pair is ahead far less often than that. Fold, note the read, and move on. Against aggressive players who raise as bluffs, it’s a different decision.

A winning hand, street by street

To see these ideas together, here is a full hand against a typical small-stakes caller. It’s a $0.10/$0.25 six-max game. The player in the big blind has called down with second pair twice in the last hour; he calls a lot and rarely raises.

You open to $0.60 on the button with K♦ J♦, the small blind folds and the big blind calls. With the small blind’s $0.10 in the middle, the pot is $1.30.

Top pair, good kicker, on a dry board. A small bet of about a third of the pot keeps all his weaker jacks, sevens and pocket pairs in, and he calls. The pot is $2.20.

The 2♦ changes nothing. Against someone who calls with second pair, a bigger bet now earns more from the same hands, so you bet $1.40, about two thirds of the pot. He calls. The pot is $5.00.

Many players check the river here, afraid of the eight. That’s a mistake against this opponent. Suppose he calls three streets with every jack, seven and pocket pair he’d have played this way, plus a few suited hands that paired. Against that range, K♦ J♦ is ahead of about two thirds of his calling hands: QJ, JT, J9, pocket pairs such as TT, 99 and 66, and hands like 9♠ 8♠ or A♠ 7♠.

Only AJ, two pair, sets and T9 beat you. A bet that’s called by worse hands more often than by better ones makes money. He could raise, but a player who almost never raises gives you an easy fold if he does. He calls with J♥ 9♥ and you win an $11.50 pot.

Now change one detail: you hold Q♠ T♠ instead and miss everything. A $3.25 bluff into $5.00 needs him to fold 39% of the time ($3.25 ÷ $8.25).

He just called two bets with any pair; he won’t fold that often. The winning play is to check and give up. Same board, same opponent, opposite decisions, both based on the same read.

Plan the hand before you act

Losing players decide one street at a time and often find themselves on the river with a hand they can’t fold and can’t bet. Winning players think a street ahead. Before betting the flop, ask what you’ll do on the turn if your opponent calls: which cards you’ll bet again, which ones you’ll check, and what you’ll do if they raise.

A quick plan prevents the most expensive situations: building a big pot with a medium hand, or bluffing once and giving up on a card that was perfect to keep betting. In the hand above, the plan on the flop was simple: bet small now, bet bigger on any blank turn, value bet the river unless an ace, king or straight card changed things.

Know the basic math

You don’t need advanced maths to win, but three numbers come up in every hand:

• Pot odds: the share of the final pot you have to put in. Facing a half-pot bet, you need to win 25% of the time to call profitably. Our pot odds guide has the full table.
• Outs: the cards that improve you to the best hand. Multiply by 4 on the flop and 2 on the turn for a rough percentage; the exact figures are in our outs table.
• Equity: your share of the pot against your opponent’s likely hands. See equity in poker.

Compare your equity with the price you’re offered and many calls decide themselves. A flush draw on the turn has about 20% equity; facing a half-pot bet that needs 25%, it’s a fold unless you’ll win a lot more when it hits.

That “unless” is called implied odds: the extra money you expect to win on later streets when your draw comes in. It’s large against opponents who pay off and much smaller against tight players who fold the moment the flush card lands, or when the stacks behind are short. Count on implied odds only when you can name the hands your opponent will pay you with.

Read your opponents

Most players fall into a few rough types, and each one needs a different approach:

• Calling stations (loose, passive): value bet more and bigger, don’t bluff.
• Tight-passive players: steal their blinds, believe their raises and fold your marginal hands to them.
• Loose-aggressive players: let them bet, call down lighter with good bluff-catchers, trap with strong hands.
• Tight-aggressive regulars: avoid them when you can; when you can’t, play solid, balanced poker.

Online, use the site’s notes and colour tags: one line per player (“calls down with second pair”, “never bluffs the river”) is enough to change a decision weeks later. Watch what players do, not what they say: how often they enter pots, raise, fold to bets and show weak hands at showdown. Our guides to reading opponents and beating tight players go further.

Rake: the opponent you always play

Every pot you win pays a fee to the site or casino. Online sites usually take about 5% of the pot up to a cap, and at micro stakes the cap is rarely reached: a 100 big blind pot at PokerStars’ NL2 to NL10 tables pays the full 5 big blinds. Live, 10% up to a cap plus a jackpot drop is common, which on small pots is an even larger share.

Over many hands that adds up to more than many players’ win rate. Two players with the same skill can finish one in profit and one at a loss purely because of where and at what stakes they play. Ways to keep more of it:

• Play sites and stakes where the rake is lower relative to the blinds.
• Use rakeback and rewards programs; they return part of the rake you pay. During our checks in November 2026, PokerStars Rewards returned 15–40% depending on level and GGPoker’s Ocean Rewards roughly 16–80%, so the same play can cost very different amounts.
• Avoid calling into small pots with weak hands; those are the pots where the rake takes the biggest share.

Our guide to rake in poker compares the caps at the major sites.

Bankroll rules by format

Bankroll management keeps a normal downswing from ending your poker. Below are the guidelines most winning players use as a starting point. A buy-in means one full stack (100 big blinds) for cash games and one entry for tournaments.

The tournament figure is a floor, not a target. In our simulation of a strong MTT player, the median worst downswing over 1,000 tournaments was 91 buy-ins; our MTT guide shows the full results. Move down a level when you fall below the minimum for your stakes, and only move up when the bankroll allows it with room to spare.

Some players take a shot at a higher level with a fixed slice of the bankroll, for example five buy-ins, and move straight back down if they lose it. That keeps a bad run at the higher stakes from turning into a big hole.

Your bankroll is money set aside for poker, separate from rent, bills and savings. Never top it up with money you need for anything else, and treat a session spent chasing losses as a reason to stop. If poker stops being fun or costs more than you planned, our responsible gambling page lists where to get help.

Keep tilt out of your decisions

Tilt is any emotional state that makes you play worse: anger after a bad beat, frustration after a long losing run, or overconfidence after a big win. Every player tilts; winners notice it early and stop.

Practical habits help more than willpower. Set a stop-loss, such as three buy-ins in a session, and quit when you hit it. Take a short break after any big pot, won or lost.

Know your warning signs: calling bets you’d normally fold, playing hands you’d never open, or wanting to “get it back” from one player. Session length matters too: most players make worse decisions after a few hours, so set a stopping time as well as a stop-loss, and keep it on winning nights. Our guides to tilt and the poker mental game go into detail.

Live, online and home games

The fundamentals are the same everywhere, but each setting rewards different adjustments.

• Live casino games deal about 25–30 hands an hour, so results take a long time to mean anything. Opponents are often looser and more passive than online, and they call more. Value bet more, bluff less, and make bigger opening raises when several players tend to call.
• Online games deal two to three times as many hands per table, and you can play more than one table. Players are usually tighter and more aggressive, especially at higher stakes, so sound preflop ranges and position matter more.
• Six-max and full ring differ too. At six-max you’re in the blinds more often and play more hands from late position, so the game is looser and more aggressive; full ring (nine players) rewards patience and tighter early-position ranges.
• Fast-fold games, such as Zoom or Rush & Cash, move you to a new table after every fold. You see even more hands, but you rarely face the same opponent twice, so reads matter less and solid default ranges matter more.
• Home games rarely have rake, which helps everyone. They’re often loose and friendly; bluffing a table of curious callers rarely works, while patient value betting does. Keep stakes at a level where nobody is hurt by losing.

Track results and study

You can’t fix leaks you don’t know about. Record every session: date, stakes, hands or hours, and result. Over time, look at bb/100 and, if your software shows it, all-in EV (what your all-ins should have won given your equity) to separate bad luck from bad decisions.

Be careful with tools. GGPoker, CoinPoker and WPT Global ban third-party trackers and HUDs, while some other sites allow them under rules. Where trackers are banned, use the site’s own statistics, such as PokerCraft on GGPoker.

A short record after each session is enough:

• date, site, game and stakes;
• hands played or hours at the table;
• result in big blinds and in money;
• two or three hands you weren’t sure about;
• how you felt: tired, tilted, focused.

Once a month, read the notes back. Patterns show up quickly: losing sessions that start late at night, results that drop when you play too many tables, or the same type of hand costing money again and again.

For study, review the hands you weren’t sure about after each session rather than only the big losses. Learn the theory behind sound defaults with our guides to GTO strategy and studying with solvers, and work on post-flop play with postflop strategy. An hour of focused review a week beats ten hours of autopilot play.

How to win at poker FAQ