Poker Tells That Actually Hold Up Live

Tells are real, and they are the most overrated part of live poker at the same time. A tell is a change in behavior that correlates with the strength of a player's hand: how fast they act, where their eyes go, what their hands do with chips. Live players give off genuine information this way. But bet sizing and the story of the hand tell you far more, far more reliably, and a player who hunts for eye twitches while ignoring a third barrel on a four-flush board has the priorities backwards. This article covers tells in that order of honesty: what to trust, what to watch, and how to stop giving off your own.

Baseline first, or you are reading noise

A behavior means nothing until you know what the player does normally. Some people's hands shake with any two cards. Some stare at everyone. The tell is never the behavior; it is the change from that player's baseline. So spend the first orbit collecting baselines while you play your normal game; you still get dealt in. Note how each player handles chips, how long they take on easy decisions, whether they talk. Reads collected before you have a baseline are guesses.

The tells worth watching

The literature on this is two books deep. Mike Caro's Book of Poker Tells, from 1984, organized the classics around one principle: weak means strong, strong means weak. Actors overact. The player who sighs, shrugs, and says "I guess I call" while sliding chips in is rarely weak, and the one who slams chips down with a stare is rarely strong. Zachary Elwood's Reading Poker Tells is the modern treatment, built around attention and timing rather than theatrics, and his follow-up on verbal tells covers the speech plays live poker is full of: unprompted chatter in a big pot leans strong, and "will you show if I fold" is a chapter of its own. The patterns most experienced live players agree carry signal:

• Instant calls. A snap call means the decision was already made; the fold was never live. As a strength read it is weaker than it looks, because live players snap-call sets and two pair all the time, especially multiway. Treat it as a lean toward draws and medium hands only once you know the player.
• The flop glance at chips. A player who looks at their own chips the moment the flop lands, before looking at anyone else, often connected with it. The glance is involuntary planning. On someone who stares at chips every street, it is nothing.
• Freezing. Players with big hands often go still and quiet while the decision is live, managing themselves. After the chips are in, the same player may loosen up and chat. Read stillness against the moment it happens and against baseline, or not at all; terrified bluffers freeze too.
• Rechecking hole cards on a flush flop. A lean toward one card of the suit, not two: players holding a made flush know, players holding one heart check which one. Worthless on someone who rechecks every street anyway.
• Decision speed changes. The player who has taken five seconds on every turn decision and suddenly takes forty is having a genuinely new kind of problem. What kind depends on the player, which is the baseline's job.

Every one of these is a lean, never a verdict. They move a close decision; they do not overturn a clear one.

Population reads beat individual reads at most stakes

At low and mid stakes live, the most profitable "tells" are about the player pool rather than any individual. Most players in most rooms do not bluff enough on rivers. Most size their bets by hand strength, small with medium hands, big with strong ones, and most are not thinking about what their sizing reveals. Playing correctly against those pool tendencies is worth more than every eye-movement read combined, and it asks far less of your observation: mostly the discipline to notice that this passive player is the one who finally raised, and to fold.

Your own tells

The defense is routine, and acting is what Caro built a book on catching. Same posture, same handling of cards and chips, same amount of time on decisions whether the choice is trivial or agonizing, same amount of talk. A deliberate few-second pause before every action, easy hands included, flattens your timing tell for almost nothing; the rest of the routine has to cover what your hands and posture still leak. Where you look is the other half: pick a spot, the pot or the board, and make it your default during big moments.

Reads are data, and data wants a record

A read is a hypothesis about one specific player. The only way to find out if it was right is to see the showdown, and the only way to keep the answer is to write it down, between hands, not mid-hand. A notebook does the job; our own habit is logging the hand in PokerInk with the read in the notes, and the showdown either confirms it or kills it. Sessions later, the notes on a regular say something no single session could: this player's forty-second tank was a flush both times, the snap-caller showed exactly the medium hands the pattern predicts. Reads that never get checked against showdowns stay guesses, and confident guesses are the expensive kind.

Log the read with the hand, then check it at showdown

PokerInk is free to start on iPhone, Android, and the web. See pricing.