The rules of Texas Hold’em take about three minutes: two hole cards, five community cards, four betting rounds. What Texas Hold’em is already covers all of it.
The strange part is this: once you know every rule, you still don’t know what to do.
You know the rules and you know which hands beat which, but nothing tells you what to do with them — so the most intuitive thing a beginner does is play by the size of the two cards in front of them.
“I have aces, so I am never folding” is the most expensive habit that comes out of it.
That is not a gap in your studying. In most games, learning the rules is learning the game — the rest is practice.
In Hold’em, learning the rules is where the actual problem starts: the player across from you just bet 500. What do they have? Is my hand worth 500 against that?
That gap is where the fun lives. You are always missing a piece of information, so bad cards can still win and good cards can still lose; luck takes a large share of any single night, and yet skill is what pulls the results apart over time. This article takes those apart, and answers a question that comes up a lot along the way: why are so many people in Taiwan playing this now?
Six things this covers: Never enough information｜Bad cards can win｜Every hand resets｜Luck gets a share｜Why now, in Taiwan｜Skill is trainable
1. You always decide on incomplete information
Chess and Go are perfect information games. Every piece is on the board, you and your opponent are looking at the same data, and the only difference is who calculates further.
Hold’em is not. You never see your opponent’s two cards until showdown — and most hands never reach showdown. All you get is this: the community cards, how much they put in, where they are sitting, and how they played the last few hands.
So the question changes at the root. It is not “is my hand strong,” it is “given the actions they took, which hands could they be holding?” The answer is never one hand; it is a set of them. That set is called a range, and it is the shared vocabulary of every serious poker discussion. Our preflop guide is a catalogue of exactly that: which set of hands a good player has, by position and by the action in front of them.
Incomplete information has a consequence that gets overlooked: the game cannot be memorised.
Perfect-information games can be. Checkers was solved in 2007 (Schaeffer et al., Science): with no mistakes on either side, the result is always a draw.
It is not that poker has no optimal strategy either. The simplest variant — heads-up, limit betting — was worked out in 2015 (Bowling et al., Science, titled Heads-up limit hold’em poker is solved). The six-handed no-limit game people actually play has nothing of the kind, and yet AI has already beaten the humans: Pluribus took on 15 professionals across two formats in 2019 and won both, one of them over a full 10,000 hands (Brown and Sandholm, Science).
None of that turns into an answer you can copy at the table, for two reasons:
- Nobody plays the optimal strategy perfectly. GTO is the unbeatable line if your opponent is also perfect, and real tables are full of people who lean in different directions: one calls too much, one never bets, one only raises with a monster. Against an opponent who leans, playing the equilibrium is not the most profitable line — you are supposed to lean back the other way. Multiway it is starker still: with three or more players, playing a Nash equilibrium can be a losing strategy because the others need not be sitting in the same equilibrium, which is exactly why Pluribus’s authors gave up theoretical guarantees on purpose
- The cards come out at random. Same seat, same AK: this time the flop is A72 rainbow, next time it is KQJ with two of a suit and you and your opponent have swapped situations entirely. However complete your strategy is, the spots it meets almost never repeat — what you can prepare is a set of principles, not a table you can look everything up in
Which means: what you face at the table is not a puzzle with a known answer. It is an opponent you have to re-estimate every time you sit down.
2. Bad cards can win, and that is the game's strangest route
There are only two ways to win a hand: your cards are better at showdown, or everybody folds before you get there.
The second one does not require good cards.
That route exists because of the first section: they cannot see your hole cards, so all they can do is work backwards from your actions. What you have and what you look like you have become two separate things — and betting is the only way you get to speak. Most card games have no such layer: bad cards are simply bad, and all you can do is play them out.
Bluffing gets described as psychology, acting, reading souls. In practice it is closer to arithmetic. Bet 100 into a pot of 100: when they fold you win that 100, when they don’t you lose the 100 you put in. So the bet needs a success rate of
100 ÷ (100 + 100) = 50%
Which is to say: whether to make that move is something you can calculate, and the smaller the pot and the bet, the lower the success rate you need. It is a technical action with a price on it, not a talent.
The other side of it matters just as much: because people do bet without the best hand, you cannot fold every time you don’t have one either. Every decision therefore carries two layers — they might have it, they might be representing it — and you have to divide your responses between those two worlds. That is the part most board games have no equivalent of, and the reason this one survives being played over and over.
One misunderstanding worth killing early: a bluff widens your options; it is not a free pass. It has a price, and the more players you have to convince, the more the success rates multiply against you. When to use it and when to put it away is something later articles will get to.
3. Every hand resets, so mistakes are cheap
A hand is usually over inside five to ten minutes. Then the chips are redistributed, your position moves one seat, and you start again.
That is unusual. A game of Go runs for hours, and a bad opening is something you carry to the end of it. A bad start in Hold’em ends the moment you fold: it costs you between nothing and one big blind, and two minutes later you have two new cards and better position.
The trade is decision density. Four betting rounds per hand, and on each one your options are the same six actions — check, bet, call, raise, fold, all-in. Play two hundred hands in an evening and you have made several hundred decisions.
| Length of one game | Cost of one mistake | Decisions in an evening | |
|---|---|---|---|
| Go | One to several hours | Can decide the whole game | Hundreds, all inside one game |
| Texas Hold’em | Five to ten minutes | Usually one hand | Hundreds, spread over 200 games |
And because position rotates every hand, the same two cards are worth different amounts depending on where you are sitting — the most expensive thing beginners overlook. Our position guide sets out how a hand relates to the seat you are in, and why the same two cards are not worth playing once the seat changes.
“A minute to learn, a lifetime to master” is a description of this structure. You are not waiting for one big hand; you are making a long series of small decisions, every one of which somebody has already studied.
4. Luck gets a share, so good decisions lose
The cards come out at random, and that has a very practical consequence: anybody who sits down has a chance. Someone who learned the rules this afternoon can beat a ten-year player tonight, which is close to impossible in chess or snooker. That share of luck is why the barrier to entry is low and why any single hand is worth watching.
The same fact from the other side: you get the money in as an 80% favourite, the remaining 20% still arrives sometimes, and you lose the hand.
The decision was not wrong. The result was. Most games have no equivalent of this. Lose at chess and you played worse; lose a race and you were slower. Hold’em draws a sharp line between decision quality and today’s result — and one evening of winning or losing says close to nothing, because one evening is nowhere near enough hands.
So the game forces a different standard on you. Not “did I win tonight,” but “given what I could see at the time, was that decision profitable?” That question has a mathematical answer, and an easier one than people expect: pot odds covers the basic form, where you never need an exact equity number, only the answer to “how often do I need to win for this to be worth it?”
It is also part of why people keep studying it. If the result cannot serve as feedback, all you have left to examine is the process.
(How much of the short run is luck, and how long it takes for skill to show through, is something later articles will get to. The one thing worth taking from here: judging your play by one night’s results is a broken method, regardless of the answer.)
5. How many more people in Taiwan are playing now
None of the four things above has changed since the first World Series of Poker in 1970. What changed is how many people can reach the game.
Taiwan’s recent growth has public numbers behind it. From The Reporter’s January 2025 feature on university poker clubs:
- 20 universities and colleges started poker clubs within six years, with four more in preparation as of January 2025
- Those clubs draw 20 to 30 people to a typical session
- A joint tournament between seven northern universities in 2024 took over 200 student entries
- The number of civil poker associations went from around five in 2017 to more than 40
Tournament scale is the other measure. The Asian Poker Tour’s 2025 Taipei stop drew 2,547 entries to its Main Event, the largest field in that tour’s history since it began in 2007 (poker.org, May 2025).
Those figures all describe the same shift: the game moved out of private games and into clubs, associations and published tournament structures.
There is a quieter change that matters more: the cost of information came right down. Ten years ago, understanding ranges, odds and position meant knowing somebody who already played. Now the videos, the range charts and the discussion are all public, and a student can cover the fundamentals from a dorm room. The path described in that report runs the same way — watch first, find a game second.
What this activity is under the law, and where the line between it and gambling sits, are two entirely different questions, and this article leaves them alone. This section answers one thing only: why you suddenly know several people who play.
6. The one that matters most: skill can be trained and improved
The sections above are about why the game is hard and why it stays interesting. Plenty of games are both. What separates Hold’em is that the hard part can be broken into rules, written down, and then practised.
The skill gap is measurable in the data. Levitt and Miles, in a 2014 Journal of Sports Economics paper, tested it against results from the 2010 WSOP: players identified as highly skilled before the events began averaged a return on investment above +30%, against −15% for everyone else.
That number is not an income claim, and it should not be read as “play well and you’ll make money.” What it demonstrates is narrower and more interesting: the same people do better than everyone else, repeatedly, and by a margin large enough to show up in results. In an activity where luck takes a real share, that is not a given.
So what does practice actually consist of? Take the first layer: which hands to play, and how wide you can open from each position. That has been studied to the point where it can be written down as rules — the 15 preflop range charts on this site set the detail out grid by grid, one chart per position, open or fold. Those charts are not drawn from feel; they are expanded from the default preflop ranges in RushCraft’s own source code.
Then there is the step after practice: checking whether you actually did it. That is what stats are for. VPIP tells you the share of hands you really played, as opposed to the share you think you played. A guide tells you when an action is correct; your own numbers tell you when you actually took it. The distance between the two is where your leaks are.
Back to the question this started with: what makes Texas Hold’em fun?
- Information is always incomplete: you cannot see their cards, so the question is always what they might have, never what you have
- Bad cards can win: betting is the only way you get to speak, and whether it pays is something you can calculate
- Every hand resets: mistakes are cheap, decision density is high, and you make hundreds of them a night
- Luck gets a share: anybody who sits down has a chance, at the cost of results being useless as feedback
- The scene has grown: clubs, associations and public tournament structures, and more study material every year, most of it out in the open
- Skill is trainable: the hard parts can be written down, practice moves them, and the movement is measurable
If the rules are still shaky, start from what Texas Hold’em is — it ends with a map of every guide on this site in reading order.
When you want to check how you are actually playing, RushCraft turns your hand history into the numbers above.
What holds people’s attention here is not the hand you won. It is looking back the next day and finding the decision you could have made better.