Hand Reading Poker: Narrow Ranges by the River

Hand reading is bookkeeping, not clairvoyance. The mistake most players make isn’t that they guess badly; it’s that they start guessing at the wrong moment. They watch a river bet land, feel a jolt of suspicion, and try to produce a single hand out of thin air. By then the useful information has already gone by, unrecorded.

The alternative is boring and it works. You start with a set of hands your opponent could hold before the flop, and you subtract from it on every street based on what he actually did. No street produces an answer. Each one produces a smaller list. By the river you’re not asking “what does he have,” you’re asking a much easier question: of the hands still on the list, how many are betting and how many are folding?

That question has a number attached to it, and the number is usually the thing that decides the hand. Most bad river calls aren’t the result of a bad read. They’re the result of never having counted.

Your opponent’s opening range is a starting number, not a read

Begin with the loosest useful fact: how often this player raises from this seat. A typical winning regular opens something like a quarter of hands from the cutoff. That’s roughly 330 two-card combinations out of the 1,326 possible. Against a live $1/$3 player who opens anything with a face card, it’s more.

You don’t need that number to be right to three decimals. You need it to be roughly right, and you need it to be a range rather than a hand. The value of starting here is that it forces you to include the hands you’ll want to forget later. Every river where you’re sure he can’t have the nuts, you’re usually wrong because you quietly dropped a suited connector from his range on the flop for no reason.

Two adjustments matter more than precision. Position tightens people, and so does a cold call. A player who flats a raise from middle position has a very different list than one who raised himself: fewer big pairs, more mid pocket pairs and suited broadway cards, almost no trash. In a field where you’ll never see the same opponent twice, this is the bulk of what you get. The 2026 WSOP Main Event back in July drew 9,208 entries, and the nine who made that final table got there mostly by reading strangers from actions alone.

The flop tells you more about what he doesn’t have

The first useful cut is usually a negative one. When a player checks back the flop in position, he’s telling you something reliable: he almost never has a strong hand he wanted to build a pot with. Cap his range there and keep it capped. That cap is worth money on every later street, because it means you can bet big on turns and rivers without much fear.

Bet sizing does the same work in reverse. A small flop bet from the preflop raiser on a board like king-nine-four with no flush possible is close to automatic for his entire range. It narrows almost nothing. A large bet on that same board narrows a lot, because there’s no draw worth charging and no reason to build a big pot with middling hands. Big bets on dry boards mean fewer hands, not scarier ones by default. Sizing tells you the shape of a range more than its strength, which is why what small and large bets actually mean is worth working out for each opponent separately.

The other flop habit worth building: notice which draws the board contains before anyone acts. Write them down mentally. Two cards to a flush, two open-ended straight possibilities, whatever they are. On the river you’ll need to know exactly which of those missed, because missed draws are the only hands most players are willing to bluff with.

Turn barrels are where the range actually gets cut

Flop bets are cheap and frequent. Turn bets are neither. When a player fires a second large bet, his range typically drops by half or more, and it drops asymmetrically. Weak pairs leave. Ace-high leaves. What stays is strong made hands plus whatever draws the turn card left alive.

This is why the turn card itself deserves a moment of thought before you react to the bet. A turn that completes nothing changes the range very little; the second barrel just removes the junk. A turn that completes a flush changes everything, because now half the aggression can come from hands that were calling on the flop rather than betting. Ranges don’t just shrink on the turn. They get re-sorted by who was drawing.

Against most live opponents, the turn barrel is also the single most honest action in poker. Recreational players will fire a flop bet with anything. Very few of them fire again on the turn without a real hand or a real draw. If you learn nothing else about your table, learn who barrels turns and who doesn’t.

Counting combinations on the river before you decide

Here’s the worked version. You’re on the button with king-queen. The cutoff, a competent regular, raises. You call. Flop comes king, nine, four, no flush possible. He bets small, you call. Turn is the deuce of hearts. He bets three-quarters pot. You call again. River is the jack of spades. He shoves for about the size of the pot.

Now count. His turn barrel on that board leaves roughly: two combinations of aces, three of kings, twelve of ace-king, three of pocket nines, three of pocket fours. Add some king-jack, which just made two pair. That’s north of twenty-five value hands, and almost every one of them beats you. You hold one king and one queen, which removes a few combinations from his set of kings and his ace-king. Not many.

Against a pot-sized bet you need to be right about one time in three. So you need him showing up with roughly thirteen bluffs to make calling break even. Now ask the only question that matters: which thirteen? No flush missed. The only straight that got there is queen-ten, and that’s a value hand. Ace-queen and ace-ten might barrel the turn, but they usually check the river instead of jamming. Realistically he has three or four bluffs, and that’s if he’s aggressive.

Fold. Not because the bet felt strong, but because the arithmetic wasn’t close. Run that same count on a board where the flush draw bricked and the answer flips completely, because suddenly there are fifteen missed draws in his range and only twenty value hands. The board, not the bet, decides it. Which cards you hold matters too, since holding a blocker to his most likely value hand shifts the count in your favor.

Where the process quietly stops working

Two situations break it. The first is the player who doesn’t have a range in any meaningful sense. He plays his own two cards and nothing else, calls the flop because he has a pair, bets the river because he still has it. Combination counting against him is precise nonsense. What you want instead is a simple behavioral read: does he ever bluff a river? If the answer is no, and against a large share of low-stakes players it genuinely is, you can fold everything short of a strong hand and be right nearly every time.

The second is multiway pots. Ranges interact. When three players see a flop, the second caller’s range is shaped by the first caller’s, and the guy in the big blind behind them is choking on all of it. Counting combinations against two opponents at once is more work than it’s worth at the table. Simplify: assume someone has it. Multiway pots reward hands that don’t need a read, and hand reading is a heads-up tool wearing a general-purpose label.

The last failure is subtler. People update their read on the river and forget what they concluded on the flop. A player who checked back the flop did not suddenly acquire a set. If your river story requires ignoring an earlier action, it’s not a read. It’s a feeling with a hand attached to it. Peter Clarke’s The Grinder’s Manual is worth reading for the street-by-street discipline alone.