Blackjack is the only game on a casino floor where your decisions change the arithmetic. Play it badly and the house takes something like two per cent of everything you stake. Play the chart correctly and that figure drops to roughly half a per cent. Nothing else about the game matters as much as that gap, and closing it takes an afternoon of memorisation.
The rules, briefly
You are dealt two cards. The dealer takes two, one of them face up. Number cards count as their face value, picture cards count as ten, and an ace counts as one or eleven, whichever helps you. The aim is to finish closer to 21 than the dealer without going over. Go over and you lose immediately, whatever the dealer does afterwards.
That last clause is the entire structure of the game. You act first, so when you bust, your stake is gone before the dealer plays a card. This is where the house edge comes from. It is not the dealer being luckier. It is you being exposed to the risk of going over first.
Once your hand is settled, the dealer plays to a fixed rule with no choices at all: draw until reaching 17 or more, then stop. A blackjack, meaning an ace with any ten-value card on the first two cards, pays more than an ordinary win and is settled immediately.
Your four decisions
Every hand comes down to picking one of four actions.
- Hit takes another card.
- Stand keeps what you have and passes to the dealer.
- Double doubles your stake in exchange for exactly one more card and no further choices.
- Split turns a pair into two separate hands, each getting a second card and a matching stake.
Some tables also offer surrender, which folds the hand and returns half the stake. It is rarer online than in a physical casino, and where it exists it is worth using in the narrow set of spots the chart specifies.
Basic strategy is a lookup table, not a philosophy
Basic strategy is the mathematically optimal action for every combination of your hand against the dealer's up card. It was worked out by simulation decades ago and has not changed since, because the maths has not changed. It is not a system for winning. It is the play that loses least.
The chart is usually printed as a grid: your total down the side, the dealer's up card across the top, and a letter in each cell. Learning it looks harder than it is, because most of the grid collapses into a handful of rules.
Hard totals
A hard total is a hand with no ace, or one where the ace has to count as one. Stand on 17 or higher, always. Hit anything 11 or below, always, because you cannot go over. Between 12 and 16 you are in the awkward zone, and the rule is to look at the dealer's card: stand if they show 2 through 6, hit if they show 7 or higher.
The reason is that a dealer showing a low card is more likely to end up drawing into a bust. You are not standing on 15 because 15 is good. You are standing because letting them draw is better than drawing yourself.
Soft totals
A soft total contains an ace counting as eleven, so the hand cannot bust on the next card. Soft 19 and above stands. Soft 18 stands against a weak dealer card and hits against a strong one. Soft 17 and below hits, and doubles in several spots where the dealer is showing 4, 5 or 6.
Soft hands are where most casual players lose value, usually by standing too early. A soft 17 cannot be damaged by another card, and treating it like a hard 17 gives away a real amount over time.
Pairs
Two rules are absolute. Always split aces, because two hands each starting with an ace is far stronger than one hand totalling 12. Never split tens, because 20 is already close to the best hand available and breaking it up to chase something better is a losing trade.
Eights are the one that catches people out. Sixteen is the worst total in the game, and splitting it into two hands each starting with eight is better than playing it, even against a dealer showing a nine or a ten. The split is not good. It is less bad.
The rule variation that costs you the most
Blackjack traditionally pays 3:2. A £10 blackjack returns £15 in winnings. A growing number of tables pay 6:5 instead, which returns £12 on the same hand. The rest of the game looks identical. The felt says blackjack, the chart still applies, and most players never notice.
That single change adds roughly 1.4% to the house edge. To put that in proportion: perfect basic strategy on a good table gets the edge down to about half a per cent, so a 6:5 payout roughly quadruples what the game costs you. No amount of correct play compensates for it.
Check the payout before you sit down, every time. If a table pays 6:5, the correct response is to find another table. This is the single highest-value habit in the game, and it costs nothing to apply.
Other rules that shift the number
Several smaller variations move the edge in one direction or the other. A dealer who stands on soft 17 is better for you than one who hits it. Being allowed to double after splitting helps you. Fewer decks in the shoe helps you. Being able to surrender helps you.
None of these individually approaches the significance of the 3:2 payout, but they compound. A table with good rules across the board and one with poor rules across the board can differ by several tenths of a per cent, which is the difference between blackjack being one of the cheapest games in the building and being unremarkable.
Insurance, and why we never take it
When the dealer shows an ace they will offer insurance, a side bet that they have blackjack. It pays 2:1 and it is presented as protection.
It is a separate bet on whether the dealer's hole card is a ten, and the odds offered are worse than the odds of that happening. It loses money over time regardless of what you are holding. The one exception is a card counter who knows the remaining deck is rich in tens, which is a different game from the one described here. For everyone else it is a straightforward decline.
What basic strategy will not do
Playing the chart perfectly does not make blackjack profitable. It brings the house edge down to somewhere near half a per cent under good rules, and half a per cent of everything you stake is still a loss over a long enough run. The game is cheaper than most, not free.
What the chart does buy you is time. A smaller edge means your bankroll lasts longer for the same amount of play, and the swings are driven more by variance than by systematic leakage. If you want the entertainment to cost less per hour, this is the lever. There is no second lever.
Live dealer blackjack against the software version
Online blackjack comes in two forms and they behave differently enough to change how you should play. The software version deals from a virtual shoe that is reshuffled after every hand. The live dealer version streams a physical table where a real shoe is dealt down before reshuffling.
Basic strategy applies identically to both, because it is derived from the composition of a full deck and that starting point does not change. What does change is pace. A software table will deal as fast as you can click, which means a bad session compounds quickly. A live table runs at the dealer's speed with a betting window between hands, and that enforced rhythm is worth more than it sounds if discipline is the thing you struggle with.
Live tables also carry higher minimum stakes, since a physical dealer serving a limited number of seats is expensive to run. Check the rules displayed at each table rather than assuming they match the software game in the same lobby. The payout and the dealer's soft 17 rule are set per table, and operators mix them within a single casino.
Card counting, and why this guide stops short of it
Counting works. It is not illegal, it does not require unusual memory, and the arithmetic behind it is simpler than most people assume. As high cards leave the shoe the remaining deck shifts in the house's favour, and as low cards leave it shifts toward the player. Tracking that drift and raising your stake when the count is favourable is how the edge flips.
Almost none of that survives contact with online play. Software blackjack reshuffles every hand, so no count can accumulate. Live dealer tables do deal from a real shoe, but the penetration is shallow, the cut card comes early, and operators reshuffle well before the point where a count would become reliable. Betting patterns are also logged in a way they are not at a physical table.
We cover counting properly in a separate guide because it is a legitimate topic. For online play specifically, treat it as background knowledge rather than a strategy. The chart is the whole of your edge here.
Bankroll and session structure
The house edge sets what the game costs per pound staked. Bankroll management sets how long you get to play for that cost, and it is the part most players skip entirely.
A workable rule is to bring an amount you would be content to lose, then size your base bet at a small fraction of it, somewhere in the range of one to two per cent. On a hundred-pound session that puts the base unit at one or two pounds. It looks unambitious. It is also the difference between a session that lasts two hours and one that ends in twenty minutes on a normal run of bad cards.
Doubling and splitting matter here too. Correct basic strategy will occasionally ask you to put a second stake on the table, and a bankroll sized so tightly that you cannot afford to double is a bankroll that forces you into incorrect play. Leave room for it.
Decide the exit points before the first hand. A loss limit that ends the session, and a win level at which you stop and bank it. Both are trivial to write down beforehand and unreasonably hard to invent halfway through a losing run.
The mistakes that cost the most
Four errors account for most of the money lost by players who otherwise know the rules.
- Taking insurance. Covered above. It is a losing side bet dressed as caution.
- Standing on soft totals. A hand that cannot bust should almost never be stood on below 18.
- Mimicking the dealer. Copying the stand-on-17 rule ignores that you act first and lose ties by busting. It is one of the most expensive habits in the game.
- Chasing with stake size. Doubling the bet after a loss does not change the odds of the next hand. It changes how fast the bankroll disappears.
Side bets belong in the same category. Perfect Pairs, 21+3 and their variants carry a house edge several times that of the main game. They are entertainment priced as such, and treating them as part of a strategy is a mistake the strategy cannot absorb.
A fifth mistake is subtler and harder to see in the moment: deviating from the chart because the last few hands went badly. The cards have no memory of what they did to you three hands ago, and the correct play against a dealer six is the correct play whether you are up or down. Every departure from the chart costs something small. Enough of them together undo the entire reason for learning it.
Set a session limit before you start, keep the chart open until you no longer need it, and leave any table paying 6:5.