Should I always double 11?
Against 2 through 10, yes, in every common ruleset. Against an ace the trainer doubles; some S17 charts hit. Either play is close, and both are far better than standing.
Guru
Updated
Doubling down means doubling your bet in exchange for exactly one more card. It is the only play in blackjack where you get to put more money on the table after seeing your hand and the dealer's upcard, so it is where almost all of a basic strategy player's positive expectation comes from. Missing correct doubles is the most expensive category of mistake after mishandling hard 16.
The rules below are for a 4-deck game with the dealer standing on soft 17, where you may double on any two cards and after a split.
Hard 11 is the best doubling hand in the game. Any 10-value card makes 21 and any 7, 8 or 9 makes a strong hand. The trainer's chart doubles 11 against every upcard including the ace; some S17 charts hit against the ace, and the difference is tiny.
Double hard 10 against 2 through 9. Against a 10 or ace the dealer's hand is too strong to justify the extra bet, so hit. Double hard 9 against 3 through 6 only. Against a 2 the dealer busts too rarely for a one-card 9 to be worth doubling, and against 7 or higher you are a clear underdog.
Soft doubles exploit a dealer who is likely to bust while you hold a hand that cannot. Against a 5 or 6 you double every soft total from 13 to 18. Against a 4 you double soft 15 to 18. Against a 3 you double soft 17 and 18. Against a 2 you never double a soft hand.
If the table does not allow doubling on your soft hand, or you cannot afford the second bet, fall back to the non-doubling play: hit soft 13 through 17, and stand on soft 18. The trainer applies the same fallback automatically when a double is not legal.
Never double hard 12 or above. One card from a stiff total is more likely to bust you than to make a hand, and you have given up the option to stop. Never double hard 8 or below; even against a 6 the expected gain from the second bet is negative.
Never double soft 19 or soft 20. These are strong hands that stand, and the one-card restriction removes the only reason to draw. Doubling a soft 19 against a 6 is a well-known expensive mistake by players who overgeneralise 'double soft hands against a 6'.
The cost of doubling is that you cannot hit again. On hard 11 that cost is tiny, because you would almost never hit twice from 11 anyway. On hard 9 against a 6 it is larger, which is why 9 doubles against fewer upcards than 10 or 11. Basic strategy has already weighed that cost for every cell, so you do not need to.
Over a long session, a player who takes every correct double and skips every incorrect one gains around 1.5 percent of their total action compared with a player who never doubles. That is roughly the entire house edge of a good game, which is why doubling discipline matters more than almost anything else you can learn.
| Hand | Double Against | Otherwise |
|---|---|---|
| Hard 11 | Every upcard | n/a |
| Hard 10 | 2-9 | Hit vs 10, A |
| Hard 9 | 3-6 | Hit |
| Soft 17-18 (A,6 / A,7) | 3-6 | Hit soft 17; soft 18 stands vs 2, 7, 8, A and hits vs 9, 10 |
| Soft 15-16 (A,4 / A,5) | 4-6 | Hit |
| Soft 13-14 (A,2 / A,3) | 5-6 | Hit |
| 5,5 | 2-9 | Hit (never split) |
Against 2 through 10, yes, in every common ruleset. Against an ace the trainer doubles; some S17 charts hit. Either play is close, and both are far better than standing.
Only where the table allows double after split (DAS). The trainer allows it. If you split 8s and draw a 3 for hard 11, double it.
Take the fallback play: hit hard 9, 10 and 11, hit soft 13 through 17, stand on soft 18. Then size your bets so this does not happen; a base bet you cannot double or split is too large for your bankroll.
Doubling for less than the original bet is allowed at most tables but is never optimal. If the double is correct, the full double is better; if it is not, do not double at all.
Reading strategy helps, but repetition is what locks it in. Use the trainer to practice hand-by-hand decisions with immediate EV-focused feedback.