Review of Hanabi & Ikebana

Review by: Dale Yu

Hanabi & Ikebana
Designer: Antoine Bauza
Publisher: Les XII Singes
Ages: 8+
Time: 20-30 minutes

At this past Essen, most folks were aware of the Bauza game called 7 Wonders, but that fantastic game was not the only new title available at the fair.  Hanabi & Ikebana is a delightful collection of two card games in a small plastic container which use the same components.  The two games are quite different in gameplay. Hanabi, which is the Japanese term for fireworks, is a cooperative style game where the players work together as a team to produce a successful fireworks display.  Ikebana. a term that refers to the traditional Japanese art of flower arranging, is a competitive card drafting game where each player tries to create the most beautiful bouquet (composed of the cards in his hand). The terms go together quite well as they share an ideogram when written in Kanji – Ikebana, is written with the characters for living and flower, and Hanabi (firework), is written with the characters for flower and fire.

Both games use the same 55-card deck.  There are 10 cards in each of the five colored suits (red, black, yellow, green and blue) with the distribution being: 1, 1, 1, 2, 2, 3, 3, 4, 4, 5.  There is also a multicolored “wild” suit with 5 cards, one of each rank 1 thru 5.  A baggie full of colored plastic chips is also included in the box, and both games also use these for different purposes.


Initially, I was introduced to Hanabi, the cooperative game, by my brother.  He had heard me state on a number of occasions that I didn’t care for cooperative games, and he wanted me to try this one out.  One of Bauza’s other recent designs, Ghost Stories, is among the very few cooperative games that I can tolerate – so I at least thought that this one might be to my liking.

In Hanabi, the players try to work together to put on a beautiful fireworks show.  The ultimate goal in the game is to display five fireworks, one of each rank in ascending order in each of the five colors in one pass through a 50-card deck.  (The standard game of Hanabi does not use the 5 multicolored cards.)  The team must do this before they make three mistakes – if they do this, they automatically fail.  If the players make it through all the cards without finishing the firework display, they can judge their relative success by adding up the highest card played in each of the five colors.

In the four player game, each player is dealt 4 cards to start the game.  Importantly, the cards are held with the backs toward the player so that you can never see your own cards while you are able to see the cards of everyone else.  Additionally, 8 blue chits are placed in the middle – one of these is spent each time a clue is given – as well as 3 red chits which are used to track incorrectly performed fireworks.  Then, starting with the player who is wearing the most different colors of clothing, play goes clockwise around the table.

On your turn, you always have the following three options:

  • Give a clue based on another player’s cards
  • Play a card to the table
  • Discard a card

When you give a clue, you first take one of the blue tokens on the table and return it to the box.  You then look at any one opponent’s hand and choose between giving information about cards of a certain color or cards of a certain rank.  Once you have chosen a color or a rank, you then point out all cards that fit that criteria to the player.  An example would be “You have two blue cards, here and here” – using your finger to point ou the specific cards.  It is important to remember that you must always give complete information with the hint, all cards that match your chosen criteria must be identified.

You could also choose to discard a card.  If you choose to do this, you simply discard a card faceup onto the discard pile and that card is now out of the game forever.  The main benefit of doing this is that each time you discard a card, you add one blue token to the center of the table – so that your team can now give an additional clue.  After your turn, you also draw a card from the deck to replenish your hand to four.

The final option is that you can play a card to the table as part of the fireworks display.  To do this, you simply choose a card from your hand and play it faceup on the table.  If it is exactly one higher than the previous card on the table in that suit, it is a valid play and it stays on the board.  Thus, at the start of the game, when there are no cards on the table, the “1” of each suit is a valid play.   The decision to play a card should not be made lightly though… The team automatically loses the game after the third misplayed card, so hopefully you know the identity of the card you are choosing to play!  After playing a card (whether successful or not), you draw a card from the deck to replenish your hand to four.  One final bonus of playing cards is that if you manage to successfully complete a firework by playing the 5 card successfully in that color, the team gets a bonus blue chip added to the table.

The game ends automatically if the third red token is discarded (from invalid card plays) – and the players lose.  The players automatically win if they are able to finish all the fireworks up to a value of 5.  The last game end condition draws near when the draw deck is exhausted.  After the last card is drawn, each player gets one more turn and the game ends after that round.  If this happens, the players total up the highest card played in each suit to generate their score.

That’s all there is to Hanabi. It is different from many other cooperative games because it is almost impossible for any one player to “quarterback” the rest of the team because no one will ever know their own hand.  This is reinforced by the fact that the rules strictly prohibit giving hints or suggestions when it is not their turn.

The big challenge in Hanabi is making sure that each clue gives useful information.  If you plan to win the game (by finishing all 5 suits), you’ll need to play 25 cards to the table.  Best case scenario for discarding cards in a 4p game then is 13 times – this assumes that the final 4 turns of the game (once the deck is all drawn) are all successful plays to the table.  With the 8 blue chits that you begin the game with, this gives your team a max of 21 hints given over he course of the game while still allowing your team to achieve the full victory.

Anytime someone gives you a clue, you really need to work hard to remember all the information given to you.  I prefer to not move cards once they’re in my hand, though I’ve seen other players successfully remember information by moving cards around in the their hand.

Even though clues are given to a specific player, it’s still very important for all players to be paying attention.  Oftentimes, multiple clues can be given the the same person to help pin down a specific card…  so it behooves you to try to remember what each player is supposed to know about their hands.  It’s also a good idea to remember the clues given so that you don’t repeat a clue.  Again, given the limited number of chances you have to give out clues, you don’t want to “waste” one by giving away information that should already be known.

You can sometimes also use finesse clues to try to get your teammate to play a specific card.   For instance, if the player to play right after me has a valid play in his hand, and it’s the only card in his hand of a specific rank, I might point at it and say “This card is your only 3”.  Eventhough he may not know which color that card is, it would only make sense for me to give him that information if he could then immediately play it.  So, hopefully, that player will then choose that card to play on the next turn.

On the whole, I find this a very enjoyable cooperative game.  The rules of the game force the players to work together without allowing any one player to dictate play.  This removes the biggest obstacle for me with any cooperative type games.   The rules also limit the kibitzing between turns because players are not allowed to give suggestions when not their turn.  Thus, the prolonged strategy discussions that bog down many other cooperative games do not occur in Hanabi.  Like most other cooperative games, it also certainly helps to play with the same players.  There are certainly some useful hint conventions that only develop with repeated plays.

The game, though, can be disastrous if any of the players isn’t paying attention (or isn’t able to remember all the clues).  The margin for error in this game is slim – a trait also seen in Bauza’s Ghost Stories, and it’s easy to squander away a hint opportunity by repeating information or giving a sub-optimal clue if you’re not completely on the ball.

Though the game obviously takes a fair amount of concentration to play well, I’ve not heard too many complaints that it’s too cerebral.  Most games take between 20 to 30 minutes, and this is about the right amount of time investment for this game.  For me and the gamers I usually play with, it’s been a challenging game, the best score I’ve managed to accomplish in 5 four-player games is 21.  The rules do suggest adding two extra blue chips if the game is too hard.  And, for those super gamers, it suggests a harder game where you use only one red chip instead of three!


Ikebana, the other game included in the box, uses a completely different central mechanic than Hanabi.  In Ikebana, each player is a florist working to make the most beautiful bouquet of flowers.  There are four rounds in the game, and in each round, players construct a hand of five cards that represents their flower bouquet.

Ikebana uses the full 55 card deck (described above), and all of the plastic chits are used as currency in the game (each color is worth a different denomination).  At the start of the game, each player gets 6 points worth of chips.  The cards are shuffled and a draw pile is created.  Players then have two options to choose from at the start of their turn.  Once they choose an option, they continue down that road until they add a card to their hand.

The first option is to draw cards from the deck.  The active player draws the top card from the deck.  If he likes it, he simply places it in his bouquet (face up on the table), and his turn is over.  However, if he does not like it, he may discard it into his personal discard pile (placed in front of him), and he must pay one point to the next player in turn order.  Then, he draws a new card.  If he likes it, he keeps it, and his turn ends.  If he does not want this card, he discards it and then pays 2 points to the next player around the table (the second player to the left).  This pattern continues until the player chooses to keep a card.  The costs for discarding continue to increase by one with each card, and the recipient continues to rotate clockwise around the table (and can never be the active player himself).

The second option is to purchase a card from any other player’s discard pile.  Of course, this can only be an option if there are discarded cards available for purchase.  You may only buy the top-most card in an opponent’s discard pile.  The cost for the chosen card is a number of points equal to the face value of the card chosen.  It is paid directly to the play from whom the card is purchased from.

Again, remember that once you choose your method of acquiring a card, you are committed to it for that turn… Therefore, once you start drawing cards, you can’t change your mind and then try to buy from someone’s discard pile.

Once a player has chosen a card for the turn and added it to his face-up bouquet, the draw pile is passed to the next player who then gets to choose how he will acquire a card.  This continues until all players have finished their five-card bouquet.  At this point, the round is scored.  Each player evaluates his own hand, calculates the score and then takes an equivalent value in chips for the score.

There are two components to scoring a hand – a base score and a multiplier.  To calculate the base score, you look at the different combinations that you can make – limited by the fact that a card may only be a part of a single combination for the hand.

  • Straight of 1,2,3,4 gives 8 points
  • Straight of 2,3,4,5 gives 12 points
  • Straight of 1,2,3,4,5 gives 15 points
  • Cards in pairs, trips, quads or quints score their face values added together

So, a hand of 2,2,5,5,5 would score 19 (2+2 for the pair of 2s, and 5+5+5 for the trip 5s).  A hand of 1,1,2,3,4 would most score 8 (for the straight), you would not add points for the second 1 because the other 1 can not be part of a pair as it is already being used in the straight.

Now, you calculate the multiplier for the bouquet.  What is important here is the color or colors of the cards you have collected.  When evaluating colors, you can designate a wild card as any color you like.

  • 4X multiplier if all your cards are of a single color
  • 3X multiplier if you have one card in each of the five different suits
  • 2X multiplier if you have exactly two colors represented in your hand

To get your final score for the round, you take your base score and apply any multiplier that has been earned – and you take a corresponding amount of chips from the bank.  The cards are then all collected, shuffled, and the player with the highest score in the previous round starts the next round.  This continues until four rounds have been played.  Whichever player has the highest score at the end of the four rounds is the winner.

Ikebana is more complex than it looks at first glance.  At first blush it seems like it’s mostly luck – based on getting cards that you want early on in your draw.  But, advance planning can be highly rewarded.  The lowest score for a round I can see is 2 (1,1,2,3,5 with 3 colors) while the highest possible score is 88 (4,4,4*,5,5* in a single color for 4X) – so there is a lot of difference in scores if the cards work out right.  This potential difference makes it worthwhile sometimes to search for the card that fits your strategy best.  You must constantly be evaluating your possible scores because both components of the score are important.

Knowing what other people are collecting can help you decide on the path that you plan to take – either by taking cards defensively to keep them away from your opponents or by staying away from particular ranks or suits that other people are also vying for.  You might also pass on a certain high valued card hoping that someone will buy it from you (and give you those victory points in payment).

The game has the potential for slowing down near the end of a round as all the information is out on the table to be calculated (and recalculated), but I haven’t seen too much of this in my 5 games of Ikebana.

Overall, I like Ikebana a little bit better than Hanabi, and I think this is mostly because it is competitive rather than cooperative.  But I’d play either if asked!  While Hanabi & Ikebana may be hard to find as this point, it was recently announced that Cocktail Games will be releasing a new version of Hanabi.  While it doesn’t sound like the full set of chits will be included, I’m fairly sure that a resourceful gamer would be able to find surrogates to also be able to play Ikebana!

Comments from other Opinionated Gamers

Larry Levy (1 play of Hanabi):  Like Dale, I’m not a big fan of cooperative games.  But Hanabi is one of the very few examples of the genre that I like.  One of the reasons is the one Dale gives, that it’s impossible for one player to dominate the game.  Another is that the bizarre, but necessary restrictions on telling fellow players what you have in your hand don’t apply, since you don’t know what you have in your hand.  But I think the thing I like most about it is that it’s a group logic puzzle, and a tough one.  It’s kind of a reverse deduction game, where the hard part is figuring out what clues to give, not on deducing what the answers mean (although guessing what the answers imply is a big part of things as well).  It’s a different kind of thinking man’s game and a very welcome one.

My one concern is replayability, that the game might start to feel samey.  That’s just a hunch, as with only one play, I obviously don’t have enough data to draw this conclusion.  However, many of the game’s fans on the Geek are very enthusiastic, so that’s probably not a valid worry.  In any event, I can recommend this one for fans of coops and thinky games alike.

Patrick Brennan (2 plays of Hanabi):  Unlike these other jokers, I am a big fan of cooperative games, and this is a very good, tense co-op game using just a pack of cards. The tension comes because you can’t see your own cards before you play them. You’re reliant on other players spending turns informing you about details of your hand (ie these are 1’s, or this is green). But you only get so many shots at this before people either have to play correctly to a pile (make 3 mistakes and it’s game over) or discard a potentially needed card in order to replenish the supply of chips which allow the team to do a helper turn. The game elevates to you trying to provide / glean the maximum mount of information with the fewest possible hints. Which then builds the tension – when he said this was a 2, did he imply it was a playable 2 or a discardable 2? Given what else I see around the table, and what’s on the 5 score piles, I’m going to say it’s … playable. Arghhh!!! So you somehow need to build a series of conventions where the same hint means different things in different situations. The game is fun in a quiet, slow, stricken way and is aimed purely at fans of deduction games. I’m not sure how much replay there’ll be once the learning curve is ridden (it’s just cards and abstracted fun after all), which puts it in the “good, not great” area. The real downside is that there’s no room for the game chit-chat or shared emotion that are the usual drivers of co-operative games (this may be what makes it more appealing to non co-op lovers), as anything you do can be construed as providing clues above and beyond what you’re allowed to provide. It can be wrenching watching someone draw the wrong conclusion from hints given so far and reach for the wrong card; to keep a straight face and not groan out loud, or make a face, or cough or simply yell out “nooooooooooo”. Regardless, or maybe because of, I’m looking forward to the evolving discovery process and seeing how nailable this is with more play with the same group of players.

Patrick Korner (12+ plays of Hanabi): This game provides more bang for buck than any other cooperative game out there. Just a pack of cards and a few chips, and yet the gameplay experience is tense and enjoyable. Much of the fun comes after having played the game with the same group a few times, as you’ll start to recognize what certain players mean in certain situations when they tell you something. It’s also very cool to watch the group’s clue-giving start to expand beyond simple Level 1 information and turn into something much more interesting. When you get told that “these cards are blue” and can use that information to figure out two or three other things about your hand, then you’ve reached deductive nirvana. One of my favourite games of 2010.

Jonathan Franklin: Hanabi is a wonderful pair of games, and no, I don’t mean Hanabi and Ikebana..  Many won’t like Hanabi because it feels as if it was developed on the planet Vulcan. It is about as far away as you can get from Cat and Chocolate as possible.  There cannot be any tabletalk because anything anyone said causes ripples of inference around the table.  At the same time, there are gloriously excruciating moments where you are torn between giving one of two clues or discarding.  If you are playing with serious players, there can be quite a bit of discussion about whether you did or did not make a mistake in doing X.  The game is well suited for post-mortems.  Whether that is a positive or a negative should help decide if Hanabi is a game for you.

You might ask why I called Hanabi a pair of games.  There is the game itself, but there is also a metagame, creating conventions and practices that optimize your team’s play without breaking the spirit of the game.  What is the spirit of the game?  Groups really vary.  Some approach Hanabi like bridge and come up with highly artificial conventions.  Other groups try to stick to natural conventions which definitely makes the game more intuitive to new players.  If you are a programmer, love negative inferences, and/or creating codes, definitely grab some friends and a Hanabi deck.  I have not played Ikebana. I love Hanabi when I am in the mood for it, but it is not a game I am always in the mood to play.

Ratings summary from the Opinionated Gamers (Hanabi)
I love it! (1) Brian Yu, Patrick Korner
I like it. (2) Dale Yu, Larry Levy, Patrick Brennan, Jonathan Franklin
Not for me…

Ratings from the Opinionated Gamers (Ikebana)
I love it! (1) Dale Yu
I like it. (1) Brian Yu
Not for me…

About Dale Yu

Dale Yu is the Editor of the Opinionated Gamers. He can occasionally be found working as a volunteer administrator for BoardGameGeek, and he previously wrote for BoardGame News.
This entry was posted in Reviews. Bookmark the permalink.

5 Responses to Review of Hanabi & Ikebana

  1. Jeff says:

    Can anyone speak to how these games work with different numbers of players? The box says 2-5, but it sounds like they might be quite different with just two.

    • patrickkorner says:

      I can only speak for Hanabi. It is much harder with more people as you get fewer turns to do things in. Yes, you see more cards, but it feels harder to set things up. I am pretty sure that my best scores have come from two player games.


  2. Dale Yu says:


    I don’t know if I agree that it is less difficult with only 2 players…

    With more people in Hanabi — you do get a bit more opportunity for “positional play” where you can give a directed clue to the next player in turn order to get a card to the table. There is less chance to do this with 2 players — and there are less cards visible.


  3. For your information, Hanabi alone will be re-published by Cocktail Games this year :
    Great review :)

  4. Justus P. says:

    I just played the game at my friday night game night from 2-5 players. It worked very well for all numbers. Each of the different numbers have slightly different dynamics due to how the clue mechanism works. Highly, highly recommended. And no, I don’t think 2P is much (if any) easier than 5P.

Leave a Reply