You have two cards of the same rank, and the dealer is waiting
The moment you look down at a pair, you are holding one of the few blackjack decisions that changes the structure of your bet rather than just its size. You can play the hand as a single total, or you can split it into two separate hands, each starting with one of the matched cards and receiving its own additional card. The choice is not a matter of feel. It follows a published decision chart, and that chart is only as good as the exact rule set printed on the table where you are sitting.
What splitting does to your money
When you split, you place a second wager equal to your original bet. The dealer separates your two cards and deals one new card to each. From that point, you play two independent hands in sequence. Each hand can hit, stand, or in some rule sets double down, and each hand settles against the dealer on its own. You are not doubling your chance to win a single hand; you are buying a second hand at the same stake, with all the variance that implies.
The decision matters because some pairs are strong as a single starting total and weak when separated, while others are weak together and stronger apart. A pair of eights totals 16, one of the worst starting hands in the game. Split, and you begin two hands at 8, each with a realistic chance to build toward a competitive total. A pair of tens totals 20, one of the strongest hands you can hold. Split, and you have thrown away a near-certain winning position for two uncertain ones.
The sequence of a split from start to settlement
A split follows the same round structure as any blackjack hand, with one extra step at the decision point. The dealer deals your initial two cards. If they match, you announce the split and place the matching wager beside your original bet. The dealer separates the cards and deals one additional card to the first hand. You play that hand to completion: hit, stand, or double where the rules allow. Only then does the dealer deal the second card to your other hand, and you play it the same way. When both hands are finished, the dealer reveals the hole card and plays out their own hand, settling each of your hands separately.
One procedural point trips up new players more than any other: you cannot touch the cards in most shoe-dealt games. Announce your intention clearly, place the second bet, and let the dealer handle the cards. In hand-held games where you do touch the cards, use one hand and place the split bet before you move anything.
How the main pairs break down
Pair decisions fall into a small number of patterns. The table below shows the standard treatment for each pair under a common rule set, but treat it as a starting point, not a universal law. The exact chart changes with the number of decks, whether the dealer stands or hits on soft 17, and what you are allowed to do after a split.
| Pair | Usual action | Reason in plain terms |
|---|---|---|
| Aces | Always split | Two separate hands starting at 11 are far stronger than a soft 12. |
| Eights | Always split | 16 is a weak total; two hands starting at 8 give you a real chance. |
| Tens | Never split | 20 is a strong hand; splitting risks turning a likely win into two uncertain hands. |
| Fives | Never split | 10 is a good starting total for doubling, not for splitting into two weak 5s. |
| Fours | Split only against specific dealer upcards | 8 is weak, but two 4s are also weak starts; the dealer's card decides. |
| Nines | Split against most upcards, stand against a few | 18 is strong, but two 9s can build higher; the dealer's 7, 10 or ace changes the math. |
| Twos, threes, sixes, sevens | Split against weak dealer upcards, hit otherwise | These pairs are weak as totals and benefit from a second chance only when the dealer is vulnerable. |
The pattern is not symmetrical. Some pairs you always split, some you never split, and the rest depend on the dealer's upcard. That is why a chart is the practical tool at the table, not a memory of a few rules of thumb.
Where the chart comes from
A published basic-strategy chart is not a list of opinions. It is a calculation: for every possible player total and dealer upcard, the chart lists the action that minimises expected loss over the long run. The calculation depends on the precise rules in force. A chart built for a game where the dealer stands on soft 17 will differ from one built for a table where the dealer hits soft 17. The number of decks changes the composition of the remaining shoe. Whether you may double after a split, and whether you may re-split pairs, changes the value of splitting in the first place.
This is why no single chart is correct everywhere. The right move is to read the table rules before you sit down, find a chart that matches those rules, and use it. The table rules are printed on the layout or available from the dealer. The chart is a reference document, not a secret.
Rule variants that change the split decision
Three rule variants affect splitting more than any others. The first is whether the dealer hits or stands on soft 17. A dealer who hits soft 17 is more likely to improve a weak hand, which makes aggressive splitting slightly less attractive against certain upcards. The second is double after split. When the table allows you to double down on a split hand, splitting becomes more valuable because a strong first card on a split hand can be leveraged. The third is re-splitting. Some tables let you split a pair, receive another matching card, and split again; others allow only one split. Each of these changes the expected value of the initial split decision.
| Rule variant | What it changes | Effect on splitting |
|---|---|---|
| Dealer hits soft 17 | Dealer improves weak hands more often | Splitting against weak upcards becomes slightly less attractive. |
| Double after split allowed | You can double a split hand after its first card | Splitting becomes more valuable; more pairs become worth splitting. |
| Re-split allowed | You can split a second time if another match appears | Splitting becomes more valuable; the initial split has more upside. |
| No re-split | One split only | Splitting is less flexible; the chart is more conservative. |
None of these variants changes the house edge on the game as a whole by more than a small amount, but they change specific decisions. A player using a chart for the wrong rule set will make small, repeated errors that add up over a session.
Procedural mistakes new players make
The first mistake is splitting without placing the matching bet first. The dealer will not separate your cards until the second wager is on the table. If you announce a split and then fumble for chips, you slow the game and invite a miscommunication. Have the bet ready before you speak.
The second mistake is treating a split pair as one hand. After the split, you have two separate wagers and two separate decisions. Many new players play the first hand, then forget that the second hand still needs to be played, or they try to act on both at once. Play each hand in the order the dealer presents it, and wait for the dealer to indicate that the first hand is settled before moving to the second.
The third mistake is splitting tens because it feels like more action. A pair of tens is a 20, and 20 wins more often than it loses against every dealer upcard. Splitting tens is one of the most expensive departures from the chart a player can make. The only reason to do it is if you are deliberately choosing higher variance for its own sake, and you should know that is what you are doing.
What the chart cannot do
A split chart tells you which action minimises expected loss. It does not tell you that you will win the hand. Splitting eights against a dealer ten is still a losing situation in the long run; it just loses less than playing the 16 would. The chart is a method for reducing the rate at which the house edge grinds down your bankroll, not a way to remove the edge. Every hand you play, split or not, carries the same built-in disadvantage over a large number of rounds. The chart changes how fast you lose, not whether you lose.
This is the honest frame for any blackjack strategy discussion. You are choosing between a bad option and a worse one in some situations, and between a good option and a worse one in others. The chart exists because the difference between those options is real and measurable, even when both are negative.
FAQ
Should I ever split tens in blackjack?
The standard chart says no. A pair of tens is a 20, which is a strong hand against every dealer upcard. Splitting turns one strong hand into two uncertain ones. The only reason to split tens is a deliberate choice to increase variance, and you should understand that you are trading expected value for action.
Why do you always split aces and eights?
Aces split into two hands starting at 11, each with a strong chance to build a competitive total. Eights split into two hands starting at 8, which is far better than playing a 16. Both pairs are much weaker as a single total than as two separate starting hands.
Does the split rule change if the dealer hits soft 17?
Yes, for some pairs. A dealer who hits soft 17 improves weak hands more often, which makes splitting against certain weak upcards slightly less attractive. The exact changes appear on a basic-strategy chart built for that rule set. Always match your chart to the table rules printed where you are playing.
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.