Nil is the bid that makes Spades interesting. You declare you will take zero tricks, and if you manage it, your side collects 100 points. Fail by a single trick and you lose the same 100.

It is a two-hundred-point swing decided by thirteen cards, and most players either never bid it or bid it on hands that were never going to work.

What a nil hand actually looks like

The instinct is "no high cards". That is necessary and nowhere near sufficient, because the card that breaks most nils is not an ace. It is a middling spade or a stranded low card in a suit that runs out.

The real requirements:

  • Few spades, and low ones. Three or fewer, nothing above a five or six. Spades are trumps - you cannot duck a trump trick by discarding, so a lone eight of spades is a loaded gun.
  • No aces, and no unsupported kings. A king with three small cards behind it is survivable; a bare king is a trick you will be forced to win.
  • Balanced side suits. This is the one people miss. A seven-card suit is fatal, because once everyone else is void you will be leading it, and leading means winning.
  • Low cards you can actually play. A hand of twos and threes only helps if the suits are distributed so you always have one to follow with.

The last two together explain most failed nils. The hand looked harmless, then a suit ran dry, and the player was left on lead with nothing safe to play.

The endgame is where nils die

Early tricks are easy - everyone has cards in every suit and you simply play low. The danger builds as suits empty out.

Count, from the start, how many cards you hold in your longest side suit and how many the other three players are likely to hold. If you have five diamonds and the suit is otherwise thin, you will eventually be the only person with diamonds, and being the only person with a suit means winning tricks in it.

The defensive move is to shed length early while others can still overtake you. Play your longest suit at every safe opportunity rather than hoarding low cards for a rainy day.

That is a planning problem: you are not optimising the current trick, you are avoiding a position you can see arriving six tricks away. It is the same shape of thinking that play Checkers online demands, where the losing move is usually made long before the position looks lost, and the discipline is reading a forced sequence before entering it.

Covering your partner

When your partner bids nil, your job changes completely and most partnerships handle it badly.

You are now playing two roles: making your own bid, and protecting theirs. Protection means winning tricks your partner might otherwise be forced to take - so you want high cards and you want to be on lead.

Concretely: bid your own hand slightly up, because you will be spending tricks on defence. Lead high, not low. If your partner shows out of a suit, stop leading it. And when in doubt, take the trick - a bag costs about nine points in the long run, while a broken nil costs a hundred.

That trade is the single most useful number in the game. It means that while your partner is nil, bags are essentially free. Take everything.

Defending against a nil

The opponents' job is to force a trick onto the nil bidder, and the technique is to lead suits where they are likely to hold a card they cannot duck.

Lead low into the nil bidder, not high. A low lead means the bidder must play something, and if the other cards played are lower still, they win. Leading an ace simply lets them discard a card they wanted rid of anyway.

Watch their discards. The first suit a nil bidder shows out of is the suit they were terrified of. The last suit they hold is usually where the vulnerable card lives.

This is reading an opponent from incomplete information, which is the defining skill of the draw-and-discard family. Anyone who has worked out what a rival is collecting when they play Gin Rummy online is doing exactly this, and in the fuller partnership version - play Canasta online - the discard pile becomes a shared public record that rewards the same attention.

Blind nil, and when a swing is correct

Blind nil is declared before you look at your cards, usually for double the stakes. It is not a stunt bid - it is a variance instrument, and the correct time to use it is narrow and identifiable: when you are far enough behind that playing a normal hand well still loses.

If you are 200 points down with the game nearing its end, a normal hand is worth perhaps 30 points of progress. A blind nil is worth 200. The expected value is worse and the win probability is higher, which is the entire point.

Conversely, bidding blind nil from in front is one of the clearest strategic errors in the game. When you are ahead, you want the game to be as boring as possible.

Choosing your variance by position rather than by mood is a principle that runs through every game with a scoreboard. It is why a trailing player should take the long-odds line and a leading one should not - the same logic that decides whether to chase a category over on Yahtzee, where a thin chance at a big score is only correct when the safe line is not enough.

Variants that are all nil

Suicide requires one partner to bid nil on every single hand, which turns the whole game into the covering problem described above - it is the best possible practice for it. Blind centres the blind variant. Whiz forces you to either bid your spade count or go nil, which removes the judgement and sharpens the execution.

Next time you play

Play a session of Suicide specifically to practise covering, then take nils in the standard game only on hands that meet all four preconditions above. Most players discover their success rate roughly doubles, mostly by bidding it less often.

The rules hub has the exact scoring, the nil explainer and the blind nil FAQ cover the basics, and the glossary names the terms. Put it into practice on the main game.

And for why making your bid by too much is quietly worse than missing it, see the companion piece on bags.