What is a set in poker? Odds, set mining and how to play one

A set in poker is three of a kind made with a pocket pair and one matching card on the board. Hold 8♠ 8♥, see a flop of 8♦ K♣ 3♠, and you have a set of eights. You’ll flop one about once every 8.5 times you see a flop with a pocket pair, and because two of your three cards are hidden, a set often wins the biggest pots of the session.

The hand’s strength comes from disguise as much as rank. Below are the exact odds, the maths behind set mining, how to play a set on different boards, and the spots where even a set should slow down.

Set vs trips

Both hands are three of a kind, and they rank the same in the poker hand rankings: above two pair, below a straight. The difference is how you make them.

• Set: a pocket pair plus one matching board card. 8♠ 8♥ on 8♦ K♣ 3♠.
• Trips: one hole card plus a pair on the board. A♠ Q♠ on Q♥ Q♦ 5♣.

The board tells the story. With trips, everyone can see the pair of queens, so opponents holding a queen are wary and players without one slow down. With a set, the board shows one eight; nothing warns your opponent that you hold the other two.

Trips also bring a kicker problem. Your A♠ Q♠ beats Q♣ J♣ only because of the ace; holding Q♠ J♠ instead you’d split the pot, and with Q♠ T♠ you’d lose. Two players can’t hold the same set, because there are only four cards of each rank.

A set on a paired board is already a full house. Holding 6♠ 6♥ on K♠ K♦ 6♣ gives you sixes full of kings, which loses to any opponent holding a king (kings full of sixes). If the board pairs your set rank instead, such as 6♠ 6♥ on K♦ 6♣ 6♦, you have four of a kind. Our guides to the full house and four of a kind cover those hands.

Odds of flopping a set

With a pocket pair, two cards of your rank are left in the 50 you haven’t seen. The chance that at least one of them lands on the flop is 11.8%. Put the other way, you miss 88.2% of the time.

The table shows what happens if you see more cards. These are exact figures for a pocket pair against a random board, counted by computer:

“Set or better” includes the rare flops that give you more than a set. Of the 11.8%, a plain set is 10.8%, a full house (a set on a paired flop) 0.7%, and four of a kind 0.2%.

People quote this as “1 in 8” and “8 to 1”, and the two aren’t the same. One in 8.5 means you hit once every 8.5 tries. Odds of 7.5 to 1 against means 7.5 misses for every hit. Both describe 11.8%; mixing them up leads to wrong pot-odds sums.

For comparison, with an unpaired hand such as A♠ Q♦ the flop gives you trips (two of your rank on the board) only 1.35% of the time. That’s another reason sets come up far more often than trips.

Set mining: when the call pays

Set mining means calling a raise with a small or medium pocket pair mainly to hit a set, and folding the flop when you miss. Small pairs rarely win unimproved against a raiser, so the call is a bet on that 11.8%.

It only works when you can win enough on the hands where you hit. That extra money is called implied odds, and it depends on the stacks behind and on your opponent paying you off.

The math behind 10x, 15x and 20x

Many players use a rule of thumb: call to set mine when you and your opponent both have at least 10 to 20 times the call left in your stacks. Here’s where those numbers come from.

Take a $1/$2 game. A player opens to $6 from the cutoff, you hold 5♣ 5♦ on the button, and the blinds will fold. Before your call there’s $9 in the pot ($6 plus both blinds). You call $6.

• You miss 88.2% of the time and lose the $6 (you fold the flop).
• You hit 11.8% of the time and win the $9 already in the pot plus whatever more you can win.

To break even, the hits have to cover the misses: 0.882 × $6 = 0.118 × ($9 + extra). The extra comes to about $36, which is 6 times your call. If your opponent paid you every single time you flopped a set, 6x would be enough.

He won’t. Sometimes he misses the flop and folds to your first bet; sometimes he has a weak hand and pays one small bet. If he pays the full amount on only part of your sets, the required stack grows:

That’s the 10x to 20x range in plain numbers. The 10x end fits loose opponents who stack off with top pair; the 20x end fits tight players who fold unless they hit something big. The “5 and 10” rule says the same thing in percentages: call when the raise is 5% of the smaller stack or less, think twice between 5% and 10%, and fold above 10%.

What moves the number

The break-even maths leaves out several things that cost you money, which is why most players use a buffer:

• Position. Out of position you’ll get less value from your sets and lose more when you miss; ask for more behind.
• Size of the raise. Facing a 3-bet, the call is bigger but the stacks aren’t, so the multiple drops fast. A $20 call with $200 stacks is only 10x.
• Players still to act. A squeeze behind you forces a fold before you see the flop.
• Multiway pots. More callers mean more dead money and more players who might pay you, which helps set mining.
• Set over set and draws. Some of your sets lose, and the rule doesn’t count those losses.
• Rake. The site takes a cut of every pot you win. Our guide to rake shows how much.

Pocket pairs aren’t only set-mining hands, either. Pairs like 8♠ 8♣ and above win plenty of pots unimproved; our starting hands chart shows when to raise them instead of calling.

Set mining in tournaments

Tournament stacks are measured in big blinds, and they shrink as the blinds rise. With 25 big blinds and a raise to 2.2 big blinds, your call is about 1/11 of your stack, at the edge of the useful range even before you allow for missed payoffs.

Your chips also aren’t worth their face value near the money, so stacking off with a set is right far more often than calling off with the hands that pay you. With short stacks, small pairs usually play better as raises or all-ins than as set-mining calls.

How to play a flopped set

A set beats most hands on most flops, so the question is how to get paid and how to avoid giving free cards to the hands that can outdraw you. The board texture decides most of it; our guide to reading the board explains texture in detail.

• Dry board, such as K♣ 7♦ 2♠ with a set of sevens: few draws exist, so you lose little by letting your opponent catch up. Small bets that keep his weaker kings in, or a check if he’s the aggressive one, both work. A slow play gives a free card, but here the free card rarely beats you.
• Wet board, such as J♥ T♥ 4♣: flush and straight draws are everywhere, and each free card gives them a chance. Bet or raise, and use a bigger size. Draws will call, and they pay you when they miss.
• Paired board, such as 9♠ 9♦ 5♣ with 5♥ 5♠: you have a full house, and the main danger is a bigger full house. Opponents with a nine have trips and will put in money; there’s no hurry.
• Monotone board, such as Q♠ 8♠ 3♠: a single spade in your opponent’s hand already makes a flush. A set has about one in three chances to fill up by the river, so it’s still strong, but big pots against heavy action often mean you’re behind.

The check-raise is a strong option out of position, especially on wet boards where the pre-flop raiser bets often. In position, a raise on the flop does a similar job.

Slow playing a set feels clever and costs money when it’s used on the wrong board. Slow play only when few cards can beat you and your opponent is likely to bet for you; on boards full of draws, bet.

Worked hand: a set on a wet board

To see these decisions in one hand, here’s a common spot from a $0.25/$0.50 six-max game, with $50 stacks (100 big blinds). The cutoff opens to $1.25 and you call on the button with 7♥ 7♦. The blinds fold, and the pot is $3.25.

You flopped middle set on a board with two spades. Only pocket kings beat you right now, but every spade draw has outs. Against A♠ Q♠, the nut flush draw with overcards, your set has 74% equity on the flop. That’s a big favourite, and also a hand that loses about one time in four if you let it see two cards cheaply.

Raising to $4 charges the draws and builds a pot with the hands that will pay: kings, which he’ll rarely fold, and the flush draws themselves. He calls. The pot is $11.25 and you each have $44.75 behind.

The 4♥ is close to a blank: no flush, and it helps only a few straight draws such as 6-5. Against A♠ Q♠ you’re now 84% to win, and the draws pay a lot to see one card. Bet about two thirds of the pot. He calls, and the pot is $26.25 with $37.25 behind.

Had a spade come instead, the plan changes. His flush draws would now be made hands, and a big bet would get called mainly by flushes. Betting smaller or checking and calling one bet keeps the pot from getting out of hand when you might be beaten.

The river misses every draw. He checks, and you bet $19.50, about three quarters of the pot, aiming at the kings that have called twice. A missed draw won’t call, so a size that a king can call matters more than a shove. He calls with A♠ K♦ and you win a $65.25 pot.

Look at what the hand needed: about $33 in profit from a $1.25 call. That’s the payoff set mining counts on, and it came from an opponent with a strong hand and stacks deep enough to pay.

Improving to a full house

Even when an opponent makes a straight or flush, a set isn’t drawing dead. If the board pairs, your set becomes a full house; if your fourth card comes, you have quads.

On the turn you have 7 outs: the last card of your rank and three cards of each of the other two flop ranks. If the turn misses, you get 10 outs on the river, because the turn card adds three more cards that pair the board.

You’ll often see 27.8% quoted for the flop-to-river figure. That number assumes 7 outs on both streets and forgets the extra three outs a blank turn gives you. The correct figure is 33.4%, almost exactly one in three; our outs table shows how to count outs like these.

That’s a real chance, and it changes some decisions. Facing an all-in on the turn from a player who’s just made a flush, you need a price of about 22% to call with your 10 outs. Against a pot-sized bet, which needs 33%, you’re folding; against a small one, calling can be right. Our guide to pot odds covers the sums.

Set over set and when to slow down

Set over set is the cooler players remember: two sets on the same flop, and the smaller one loses almost everything. When two players hold different pocket pairs, both flop a set in about 1.0% of hands.

The more useful number is conditional. Once you’ve flopped a set, a specific opponent who also holds a pocket pair has flopped a set of his own 8.5% of the time. Worse, if his pair is higher, you’re nearly dead. With 7♠ 7♥ against J♠ J♥ on J♦ 7♣ 2♠, your equity is 4.3%: you need the last seven.

That 8.5% is too small to fold a set on most dry boards; you’d give up far more against all the hands you beat. But some situations make a fold or a smaller pot reasonable:

• Four cards to a flush or straight on the board, and a tight opponent putting in big money.
• Heavy action on a board where your set is the bottom one, such as 2♦ 2♣ with the board showing K♠ Q♠ 2♥ J♠.
• A player who almost never raises suddenly raising the turn and river.

Any one of these alone isn’t enough. Together, against a player you know plays straightforwardly, they’re good reasons to call instead of raising, or to fold to a final shove. The nuts guide explains how to tell which hands beat yours on each board.

Losing players reach the river with a set and pay off every raise without looking at the board. A set on four spades is a bluff-catcher against many opponents; treat it as one.

Sets in Omaha

In Omaha you hold four cards and every player has many more ways to make straights, flushes and higher sets. A set is still a good hand, but the bottom set on a connected flop is often in trouble, and a set in Pot-Limit Omaha is frequently up against a big draw with 13 to 20 outs. Our guide to wraps in Omaha shows how strong those draws are.

Top set with a backup draw is the hand Omaha players want. Bottom set on a wet board, without a redraw, is a hand to play carefully. The rules for using your cards in Omaha are in our Omaha rules guide.

Set in poker FAQ