The moment a pair lands in front of you
You have placed a bet, the cards are out, and you are looking at two cards of the same rank. The dealer is showing a card, and the table is waiting. That moment is not a guess. It is a decision point with a published answer, and the answer changes depending on the rules printed on the felt or in the game information panel.
Splitting means turning one hand into two. You place a second wager equal to your original one, the two cards are separated, and each card becomes the first card of its own hand. You then play each hand independently: hit, stand, double, or in some rule sets split again. The reason the decision matters is that you are not just changing how you play; you are changing how much money is at risk on that round.
What makes splitting confusing for new players is that the correct choice is rarely about whether the pair itself is good. It is about what the dealer is showing and what the rules allow you to do after the split. The same pair can be a split in one situation and a stand or a hit in another.
What a split does to the hand
When you split, you are giving up a two-card total and replacing it with two starting hands. Sometimes that improves your position; sometimes it makes a bad situation more expensive. A pair of eights is the clearest example. As a single hand, 16 is a weak total against almost any dealer upcard. Split, those eights become two hands starting at eight, and an eight is a much better building block. A pair of tens is the opposite: 20 is a strong finished total, and breaking it into two hands starting at ten is usually a way to turn one good hand into two uncertain ones.
The second wager is the part people forget. Splitting doubles the amount of money you have on the table for that round, and if you are allowed to resplit, the amount can grow again. That does not make splitting wrong; it makes it a decision about expected loss per unit wagered, not just about the cards. A player who splits every pair because it feels active is multiplying their exposure on hands where the mathematics does not justify it.
After the split, the rules take over. Most tables allow you to draw as normal on each new hand. Some allow doubling after a split, which makes splitting more attractive because it gives each new hand the same options as a regular starting hand. A few rule sets restrict drawing on split aces, commonly to one card per ace, which changes the value of splitting aces in the first place.
The pairs worth separating
Basic strategy for splitting is usually presented as a chart: your pair on one axis, the dealer's upcard on the other. The chart is not a list of opinions. It is derived from the exact rule set in use, including the number of decks, whether the dealer stands or hits on soft 17, whether doubling after a split is allowed, and whether resplitting is permitted. A chart built for one rule set can be wrong for another.
That said, a few pairs are stable across most common rule sets. Aces and eights are the two pairs that almost every published chart tells you to split regardless of the dealer's upcard. Aces split into two hands that each start with an ace, which is the strongest starting card in the game, even when only one additional card is allowed. Eights split to escape a weak 16. At the other end, tens and fives are almost never split: tens because 20 is already strong, and fives because a starting total of 10 is a good doubling hand.
| Pair | Typical basic-strategy treatment | Reason in one line |
|---|---|---|
| Aces | Split against any dealer upcard | Two hands starting with the strongest card |
| Eights | Split against any dealer upcard | Escapes a weak 16 total |
| Tens | Never split | 20 is already a strong finished hand |
| Fives | Never split | A starting total of 10 is a strong doubling hand |
| Nines | Split against most upcards, stand against 7, 10 or ace | 18 is strong, but two nines can beat a weak dealer card |
| Sixes | Split against weak upcards, hit otherwise | 12 is weak; two hands starting at six can improve |
The middle pairs are where the dealer's upcard does the work. A pair of nines is worth 18, which is a good total against a dealer showing 7, 10 or ace, so the chart usually says stand. Against a weak upcard like a 5 or 6, two hands starting at nine are worth more than one hand stuck at 18, so the chart says split. Sixes, fours, twos and threes are split mainly against weak dealer upcards, because the alternative is a low total that needs work anyway.
Where the table rules change the chart
Every split decision is conditional. The same pair against the same dealer card can flip from split to hit or stand when the rule set changes. The two rules that move the chart most are doubling after a split and the dealer's soft-17 behaviour.
Doubling after a split matters because it gives each new hand the full set of options. If the table does not allow it, splitting becomes less valuable in the marginal cases, because a split hand that starts well cannot be pressed as hard. The dealer's soft-17 rule changes the dealer's final-hand distribution, which shifts the borderline decisions. A chart built for a dealer who stands on all 17s is not automatically correct at a table where the dealer hits soft 17.
The number of decks matters too. With fewer decks, the composition of the remaining cards changes, and some split decisions that are correct in a six-deck shoe are incorrect in a single-deck game. This is why a printed strategy card usually states its assumptions in small type. If the card does not state them, it is incomplete.
Before the first hand, read the table's posted rules or open the game's information panel. Look for four things: whether doubling after a split is allowed, whether resplitting is allowed and to how many hands, whether split aces receive one card or can be played normally, and what the dealer does on soft 17. Those four lines determine which chart applies.
How the split changes the house edge
Splitting does not change the house edge of the game. The house edge is a property of the whole game under a given rule set, and it applies to every wager you place. What splitting changes is how many wagers you place and at what expected cost per wager. A correct split on a pair of eights puts a second bet on the table, but it does so in a spot where the expected loss on that second bet is lower than the expected loss on playing the 16 as a single hand. An incorrect split on a pair of tens puts a second bet on the table in a spot where the expected loss is higher than simply standing on 20.
That is the whole arithmetic. A player who splits correctly is not winning more; they are losing less per unit wagered on the hands where the decision arises. The difference is small in any single round and material over many sessions. The same logic applies to every basic-strategy decision: the chart does not promise a win, it minimises the expected cost of the decisions you are offered.
For a player who wants the exact figures, published reference sources list the expected return for each hand and each possible action under a stated rule set. Those numbers are specific to the rule set, and a figure quoted without the rule set is not reliable. The game information panel and the table's posted rules are the reader's own reference for the game they are sitting at.
Mistakes that cost more than the cards
The first procedural mistake is splitting without checking the rules. A player who splits aces at a table where split aces receive one card, believing they can play them normally, has made a decision based on a different game. The second is splitting a pair because the pair feels unlucky. A pair of fours is not a bad hand because it is a pair; it is a total of eight, and the chart treats it as an eight. The third is forgetting that the second wager is real money. Splitting doubles the stake on the round, and a player who splits marginal pairs repeatedly is increasing their total wagered without a corresponding reason.
A smaller mistake is asking the dealer what to do and expecting a strategy answer. Dealers can explain the rules, but they are not strategy advisors, and their answer will not be tailored to the rule set in a way you can verify. The chart is the tool. The rules on the felt are the input. The decision is yours.
FAQ
Should I always split aces and eights?
Under the rule sets that most published basic-strategy charts cover, yes. Aces give you two hands starting with the strongest card, and eights let you escape a weak 16. The exception is a rule set so unusual that the chart changes, which is why the chart you use must match the posted rules.
Can I split a pair of tens if the dealer shows a six?
The chart almost always says no. A total of 20 is already a strong hand against any dealer upcard, including a six. Splitting turns one strong hand into two hands starting at ten, each of which still has to be played out, and it doubles your wager to do it.
Does splitting change my odds of winning the hand?
It changes the expected cost of the round, not the built-in house edge of the game. A correct split puts a second wager on the table in a spot where the expected loss is lower than the alternative; an incorrect split does the opposite. Over many sessions, correct split decisions reduce the total expected loss on the hands where pairs appear.
Play responsibly
Gambling should be treated as paid entertainment, never as a way to earn income or recover losses. 1-800-MY-RESET (1-800-697-3738)
18+ or 21+ depending on where you are; follow the minimum age that applies to you.
This article is general information about how these mechanics work. It is not legal advice and not a recommendation to gamble or to use any particular operator. Availability and legality differ by jurisdiction — check the rules that apply where you are.