Blackjack Basic Strategy for On-Chain Play
Most players hand the house free money by guessing. Basic strategy is the mathematically optimal play for every blackjack hand — solved by computer decades ago — and it drags the house edge down to roughly 0.5%, better odds than almost anything else in a casino. This guide covers the whole framework: hard totals, soft totals, the doubling windows, surrender spots, pairs, and the dealer-strength mental model that ties it all together.
Play blackjack →What basic strategy actually is
Basic strategy is a decision table that gives the single best action — hit, stand, double or surrender — for every combination of your hand and the dealer's up card. It is not opinion and it is not a betting system: it was solved with exhaustive computer simulation by comparing the expected value of every legal action in every situation, over millions of simulated hands. According to the Wizard of Odds, a player using perfect basic strategy faces a house edge of only about 0.5% under standard rules, versus 2–3% or worse for an average untrained player. That gap is the difference between grinding and getting bled. Crucially, basic strategy requires no memory of past hands and no card counting — it is a pure lookup from two pieces of information you can always see: your total and the dealer's up card. Learn it once and every decision at the table becomes automatic.
Hard totals: the backbone of the chart
A hard total is any hand without an ace counted as eleven. The rules compress nicely. Hard 8 or less: always hit — you cannot bust and your hand is too weak to stand. Hard 9: double against dealer 3 through 6, otherwise hit. Hard 10: double against 2 through 9, hit against a ten or ace. Hard 11: double against everything except an ace. Hard 12: stand against 4, 5 and 6, hit against everything else — including 2 and 3, the spot most players get wrong. Hard 13 through 16: stand when the dealer shows 2 through 6, hit when the dealer shows 7 through ace. Hard 17 and up: always stand. The logic behind the stiff hands (12–16) is simple: against a weak dealer card you refuse to bust first and let the dealer's forced draws do the damage; against a strong card the dealer probably has a made hand, so you must risk busting to compete.
Soft totals: the ace changes everything
A soft hand contains an ace counted as eleven, which means one card can never bust you — and that safety net makes soft hands far more aggressive than their number suggests. Never stand on soft 17 or less; a soft 17 that draws a ten simply becomes hard 17, so hitting is free upside. Soft 18 is the famous trap: it looks strong but is only a favourite against a weak dealer. Stand on soft 18 against 2, 7 and 8; hit it against 9, ten or ace; and double against 3 through 6 where the table allows soft doubling. Soft 19 and above: stand. The classic soft doubling windows are ace-2 and ace-3 against 5–6, ace-4 and ace-5 against 4–6, and ace-6 against 3–6. Where soft doubling is not offered — Llamabet currently doubles on hard 9, 10 and 11 — just hit those hands instead; the cost is small, and the hit-or-stand rules above never change.
The doubling windows: where the edge claws back
Doubling down is how you get extra money on the table precisely when the math favours you, and it is worth a large slice of basic strategy's value. The three hard doubling windows are 9, 10 and 11 — exactly the totals Llamabet lets you double. Eleven is the crown jewel: roughly one card in three makes it 21, and almost any ten-value card leaves you unbeatable, so double 11 against every up card except an ace. Ten is nearly as strong: double against 2 through 9, but keep it to a hit when the dealer shows a ten or ace and may already have you beat. Nine is the marginal window: double only into dealer weakness, 3 through 6. Skipping these doubles is one of the most expensive polite-looking mistakes in the game — you are declining a bet the simulation says you should be raising. A natural blackjack pays 3:2 on Llamabet, so the payout structure rewards playing these windows at full value.
Surrender and pairs, briefly
Late surrender returns half your stake and folds the hand, and it is correct in exactly the worst spots — the hands whose expected loss played out exceeds 50% of the bet. The list is short: surrender hard 16 against a dealer 9, ten or ace, and hard 15 against a ten. Surrendering feels like defeat, but keeping half your stake in a spot where you expect to lose more than half is pure profit over time. Do not surrender anything else. As for pairs: where splitting is offered, the timeless rules are always split aces and 8s, never split tens or 5s. Llamabet's current ruleset is hit, stand, double and surrender, so play pairs as their hard total — a pair of 8s is hard 16 (surrender against 9, ten and ace, stand against 2–6, hit against 7), and a pair of tens is a hard 20 you simply stand on.
The mental model: weak dealer, strong dealer
If the full chart feels heavy, one model organises most of it. The dealer has no choices — on Llamabet the dealer hits 16 and stands on 17, every time, by rule. That rigidity is why the game is solvable, and it splits the up cards into two regimes. Dealer 2 through 6 is weak: the dealer is likely sitting on a stiff total and forced to draw at it, with 5 and 6 the most bust-prone cards in the deck. Against weakness you stand on your stiffs, double your 9s, 10s and 11s liberally, and let the dealer's forced draws destroy the dealer. Dealer 7 through ace is strong: the dealer likely finishes with a made 17-plus, so you hit until at least 17, keep doubles disciplined, and surrender the worst 15s and 16s. Ask one question before every decision — is the dealer weak or strong? — and the correct play usually follows.
Costly deviations, practice, and why on-chain changes nothing
The common leaks are well documented: standing on 16 against a ten out of fear (hit or surrender it), refusing to hit 12 against a 2 or 3, skipping the double on 11, and mimicking the dealer by always hitting to 17 — a strategy that quietly plays at several times the optimal house edge. Playing hunches after a losing streak is the most expensive leak of all, because past hands carry zero information here. Practice is cheap: the minimum bet is 1 SUI, so you can drill the chart at low stakes until every decision is reflex. And the chart works identically online and on-chain: every Llamabet hand is dealt from a fresh four-deck shoe whose shuffle is committed to the Sui blockchain before you bet and revealed after, so there is no count to track and no pattern to exploit — but also no way for the deck to be rigged against your correct plays. Basic strategy is the ceiling, and provable fairness guarantees you are actually playing under it. It will not make you a guaranteed winner — the house keeps its ~0.5% — so set a bankroll, bet a small fraction of it, and never chase.
Frequently asked questions
Does basic strategy guarantee a win?
No. It minimises the house edge to about 0.5% but does not flip the game in your favour — the house keeps a small edge. Basic strategy makes you lose the least over time, maximises your chances in any single session, and is the strongest legal play available in online blackjack.
Can I count cards on Llamabet?
No. Each hand is dealt from a freshly shuffled, provably-fair four-deck shoe committed on Sui before the bet, so past cards carry no information about future ones. With counting off the table, perfect basic strategy is the optimal approach — and the mathematical ceiling.
What is the best first move to memorise?
Double down on 11 against everything but an ace, and stand on 13–16 only when the dealer shows 2 through 6. Those two rules cover the highest-value and most frequent decisions, and they are where untrained players leave the most money on the table.