2 Player Spades (Two-Hand Spades)

Standard Spades deals thirteen cards to each of four players and moves straight into bidding. That cannot work with two people. A straight deal would either give each player twenty-six cards or leave half the deck sitting unused with no way to decide which half.

The variant solves it with a draw phase. The full fifty-two card deck goes face down between the two players as a stock, and the players alternate turns pulling cards off it, choosing which ones to keep, until both have assembled a thirteen-card hand from scratch. You are not dealt a hand in this game. You build one, while your opponent builds theirs out of the same stock.

Once both hands are complete the draw phase ends and the rest of the variant is ordinary Spades. Spades are permanent trump. You must follow suit if you can. Nobody may lead a spade until spades have been broken. Bids score ten a trick with one point per overtrick, and a missed bid costs ten a trick.

Shuffle all fifty-two cards and set them face down as a stock. The non-dealer draws first. Turn the top card and look at it. You now have exactly one decision to make: keep it or refuse it.

If you keep it, the card goes into your hand and you must immediately take the next card off the stock and place it face down on a discard pile without looking at it. If you refuse it, that card goes face down on the discard pile and the next card off the stock goes into your hand sight unseen. You are stuck with whatever it turns out to be. Either way exactly two cards leave the stock and exactly one enters your hand.

Turns alternate. Each player takes thirteen turns, sees twenty-six cards pass through their hands, and finishes holding thirteen. The other twenty-six cards sit dead in the discard piles for the whole deal. The arithmetic is exact, which is why the procedure works: fifty-two cards, two players, thirteen turns each, two cards a turn.

Some tables play a looser version in which the second card is turned face up and must be taken, which strips out the blind element entirely. Both versions are played. Settle which one you are using before the first draw, because they reward completely different levels of risk.

When both hands are full, bid. The non-dealer bids first and the dealer bids second, and there are thirteen tricks to split between exactly two contracts. There is no partner number to add to yours, so whatever you say is the whole contract. Make it and you score ten a trick plus one for each overtrick. Miss it and you lose ten a trick bid.

Thirteen tricks split between two hands means the average contract sits at six or seven, roughly double what you would bid in the four-player game. Beginners coming from partnership Spades chronically underbid here because a bid of six still feels enormous. It is not. It is the par result.

Nil is normally still available at one hundred points, but it is a very different bid with nobody covering you. Your opponent holds half the live cards, wants you busted, and can duck any trick to force you in. Plenty of two-hand tables drop nil altogether or price it at fifty, and that is a reasonable house call rather than a mistake.

Two-hand games commonly run to two hundred or three hundred rather than five hundred, because contracts are twice the size and points arrive twice as fast. Bag penalties are often relaxed for the same reason: when two bids frequently total less than thirteen, somebody banks overtricks nearly every hand through no fault of their own.

Twenty-six cards never enter play. This is the single largest strategic difference and most players never adjust for it. In four-player Spades, if you have seen twelve spades you know exactly who holds the thirteenth. In two-hand Spades the missing ace of hearts is as likely to be face down in a discard pile as in your opponent's hand.

Elimination counting therefore does not work. What replaces it is better: you personally looked at a large slice of the deck during the draw. Every card you saw and refused is a card you know for certain is out of play. Track those, and by the middle of the hand you often know more about the live deck than you ever would at a four-player table.

The practical consequence is that kings and queens are worth more than your instincts say, because the card that would beat them may simply not exist in this deal. Second-round winners are routine. Leading a king early to see whether the ace shows up is a normal and cheap probe.

Working the other way, your aces are less safe than they look. With only twenty-six cards live across two hands, both players are void in something by the middle of the hand, and a bare ace of diamonds gets ruffed far earlier than it would in the four-player game.

The keep-or-refuse choice is the only genuinely new skill in the variant, and it is a straight comparison between a card you can see and a card you cannot. Refusing is not free. Refusing means a random card enters your hand instead. So the question is never whether the card is good, it is whether the card is better than an average unknown one.

Near-automatic keeps: every ace, and every spade above the nine. Spades are trump, and in a thirteen-trick head-to-head game trump length converts to tricks almost directly. Near-automatic refusals: middling cards in a suit where you already hold five or six, because a sixth heart adds almost nothing and takes a slot that a spade or an ace could have used.

The calculus flips near the end of the draw. With two turns left and only three spades, a spade you would have shrugged at on turn two becomes an obvious keep, because you can no longer count on another one coming. Shape matters more than rank once your remaining turns run low.

Watch your opponent's decisions. You cannot see the card, but you can see what they did with it, and that is information. A card kept on the very last turn usually filled a hole. A long pause followed by a refusal usually means a close call, which means their hand is already shaped and they were protecting it.

Every trick is decided by exactly two cards, so the second-best card in a suit wins far more often than it does with four players fighting over it. Queens win tricks. Jacks win tricks. Bid accordingly.

The lead is a liability in this variant. Your opponent always plays second, always knows exactly what they need to beat, and can duck for free whenever the trick is not worth taking. Winning a trick means leading the next one, which hands that advantage straight back. Strong two-hand players are perfectly happy to give away cheap early tricks and cash their winners late, once the opponent's remaining cards are largely known.

Because trumps are the only way to beat an ace, and because both players are usually void in something by trick seven, the player still holding spades at trick ten normally decides the contract. Do not burn your last two trumps to win tricks you were going to get anyway. Break spades on your own terms, and keep something back for the end.

Play it when there are two of you and you want actual Spades rather than a substitute. It keeps every rule that defines the game and adds a build phase that most players end up enjoying on its own terms. It is also the best card-counting practice available, because tracking the draw is directly rewarded.

What it loses is the partnership. No signalling, no covering a partner's nil, no negotiating a team contract, no shared bags. If those are the parts of Spades you actually like, two-hand Spades will feel thin no matter how well it is designed.

The alternative for two people is to fill the other two seats with computer players and run standard partnership rules. That is a legitimate way to play, but it is a different game from two-hand Spades, and none of the draw-phase strategy on this page transfers to it.