Designer: Muneyuki Yokouchi (横内宗幸)
Publisher: Ayatsurare Ningyoukan
Players: 3-4
Playing Time: 30-45 minutes
Times Played: 5 on a purchased copy
Cat in the Box is a trick-taking game that continues the “tradition” of untangling suits and ranks from their natural connection. When we think about card games – trick-taking or not – we go to rank and suits: games played with a traditional deck of cards have it (e.g. rummy, poker, blackjack). Uno has it.
Cards have one number and one suit.

As best I can tell, this changed in 2005 with Joe Huber’s small release Transportation Tricks. Players are dealt a hand of a fixed amount of cards that show only a number and a variable amount that show only a suit. The suits have a fixed hierarchy, and after using one of their number cards to set the value of the suits for the round, players will play two cards to each trick: one number and one suit.
This twist doesn’t seem to show up in another published trick-taking design until dois’ first release in 2014. Rather than playing one suit and one number to each trick, dois’ has one of the two persist. That is, if you play a 7 (rank) and a Crosswalk (suit) to the first trick, you’ll replace either the rank or the suit for the second trick, but the other will remain – you will be playing either a 7 or a Crosswalk card to the second trick.
In these first two examples, an unstated symptom is that there’s no lever controlling the distribution. A hand may see many copies of the same “card” played, and others not played at all. (This is something that also happens in dice-based trick-taking games.) Taiki Shinzawa, the designer of dois, followed that design up the next year with mantel, a game that puts us closer to the Cat in the Box setup, as it ensures each person will play the ranks 1 through 12 exactly once each in a hand. In almost the converse setup of Cat in the Box, players are each given cards 1 through 12, suitless, and face up on the table, but are “dealt” a hand of colored cubes that will represent the suits. In each trick, choose a cube from your hand and place it on a rank; that’s the only time you’ll play that rank this hand. This ensures that each of 1 through 12 is played 3 times, but the distribution of the 4 suits is still left to wander.
Cat in the Box adds additional structure to the available card distribution by fixing the play space to exactly one of each card in the distribution, such as 1 through 8 in each of four suits. It doesn’t do this perfectly, but puts the onus on you to make sure it works out. And if you mess it up – welcome to Paradox City! The hand ends, adversely for you, and closer to fine for the other players.

The game includes a grid showing the potential cards that can be played, and players have tokens that they’ll place in the grid to show the cards they have played. That is, if you play an 8 from your hand and declare it to be yellow, no player will be able to play a yellow 8 later.
While the game includes 4 suits in the grid, there are 5 of each number. This is one of the ways the game tempts you with the paradox. All of the cards are dealt out, but two cards will go unplayed – one is discarded by each player at the start of a hand, and the other, ideally, is left in your hand unplayed.
The game play is the usual “must-follow” affair, with the highest card of the lead suit winning, unless trump (red) was played, in which case the highest trump card wins. The game does use the “trump must be broken” rule, meaning that a red card cannot be lead to a trick until it has been played by a player who could not follow suit.
….but in a game where the cards are suitless, what does not following suit look like? Um, whatever you want! In a design choice that harkens back to the Willi / Meinz’ rule that allows players to win a trick simply by declaring they want to on their turn, in Cat in the Box, you can just…be out of a color! Dale leads a blue 6 and you don’t want to play blue? Simply declare yourself out of blue. (Each player has a card where they track what they’ve declared they are out of.)

The game is played over a number of rounds where players will earn points for the tricks they’ve taken, a simple 1 point per trick. Additionally, as in dois and mantel, Cat in the Box features bidding where players must predict how many tricks they’ll win. It uses a “softer” bidding where the penalty for not hitting your bid is absent, and the “bonus” for hitting it is the core to why the design works so well – though when I first read the rules, it felt superfluous.
This bonus centers around the patterns players have created with their tokens on the central board. There’s no penalty for missing your bid, but if you hit it exactly, you’ll earn 1 extra point per token you have in your largest orthogonal grouping on the board. It’s this mini-game which drives the larger arc of the game and makes the game a treasure, and not simply a novelty.
Let us step back to an initial deal. We’ve been given three 5’s, one 4, two 3’s, one 7, one 8, one 2, and one 6. Your “Turn Zero” analysis here is balancing how many tricks you can win and what sort of pattern you can create. The bids you can make in the game are limited – maxing out at 3 or 4. If you think you have a hand that can win 6 or 7 tricks, you may simply be best off winning as many tricks as possible and forgoing attempts at the bonus – not only will it lock in your points as there’s no penalty, but this sort of strategic zag may throw the other players off their game as they may have been expecting you to only win the number of tricks you bid, say…1, perhaps.
If we try to make a go of the orthogonal pattern, what would that look like? The 6/7/8 can only connect in a line, that is, in a single suit. The trio of 5’s will help us spread out, so that connecting our single 4 and our single 6 won’t mean needing to play 4/5/6 in a single suit. So we pick something to discard, hoping to connect our 4/5/5/5/6, and if we’re lucky, one of the 3s and maybe the 7.
Let’s pause this discussion and talk about what will happen when you cause a paradox, or, fingers crossed, the player to your right does. If you need to play a card that can’t exist, you’ve done it. It could be that you’re down to a 5 and a 7, but there are no places left to play them because the players kept all five of each in circulation. It could be that the only remaining 5 and 7 slots are in suits you said you were out of 4 tricks ago.
When it happens, the trick ends immediately, so if the player to your right triggers it, the scoring rules are none-the-wiser if you were also about to do it. Whoever causes it to happen will lose a point for each trick they’ve taken, and the other players will score normally. (Well, it’s not exactly the same, as you may have been waiting to win one of those final tricks in order to hit your bid, and if the round ends early, you could miss out on the bonus.)
Back to our example – perhaps it’s not wise to keep the three 5s? Are we tempting a paradoxical fate in the 5 column? What if Dale lead’s the yellow 5 to the first trick – do we bail on yellow in order to start claiming the other 5s?
I love the way the pressure to not cause a short-suit paradox in the late game leads to a single suit filling up in the early game; this makes it difficult to get intra-suit runs, like our 6/7/8 we were hoping for. This fear causes you to want to keep the sets (like the three 5s), but as we discussed, that strategy has its own risks.

Sometimes a game is so clearly brilliant to me that I forget to write a review for it! Let’s Make a Bus Route!, SCOUT!, etc. This is one of those, and I don’t think Muneyuki Yokouchi gets enough credit for the virtuosity they exhibit in their designs – especially the trick-taking ones, like 7 Symbols, and 7 Nations (Yokai Septet).
The design here is audacious, as are your tactics and strategies. We’ve touched on winning many tricks, optimizing grid placement, and trying to avoid paradoxes, but we haven’t gotten to defensive plays where you interrupt other players’ groupings; strategies around which color to open with as they have different adjacencies and real estate futures; and which numbers may be more available for large grouping strategies.
There is so much to explore – in a single game and in the next game. With the group you’re playing with today, and the one you’ll play with tomorrow. It also does it with minimal rules overhead.
Cat in the Box’s….box, that takes the bifurcated suits and ranks and corrals them, is what enables everything else to flourish: the grid placement, the short suit declarations, and the paradoxes.
This game is a treasure, and I hope you have a chance to play it. There should be a new edition out in April from Hobby Japan with slight rules tweaks and a new player count.
best,
JN
Thoughts from the Opinionated Gamers
Joe Huber (1 play) – I agree entirely that Cat in the Box is a clever game – it’s just not one that struck home for me as much as I had hoped. (And, since Transportation Tricks was mentioned, I’ll say – I think Cat in the Box is clearly the more clever design.) I found that there wasn’t as much opportunity for smart play as I would have preferred – there are choices, but they can be limited by the cards one is dealt. I also found the lack of wrap-around on the board very limiting – the game would be more interesting, in my opinion, if wrap-around was allowed. All that said, it’s still a game I’d be willing to play more, and appreciate even if I’m not so enamored of it as others.
Rand L (7 plays) – “Virtuosity”, “audacious”, “clever”. All words that sit well when describing this design. Though James Nathan opened the article with talk about “tradition,” it is clear that this game is an outlier that bends tradition to a near breaking point. And if it didn’t quite break tradition, Cat in the Box certainly broke my brain and the brains of everyone I’ve introduced it to. That initial jarring “you play a card HOW?!” fades away rather quickly, though, and we’ve found ourselves joking with our colorless cards that “Oh, green? I don’t know how I wasn’t dealt any green,” which of course is a daring gambit. When a hand is played to perfection, the bonuses from the contiguous area come flowing in, but if you’ve hit your bid well without managing the contiguous area in a similar manner, you likely would have been better off going for max amount of tricks. I still have not played the reverse side of the board that features shifted areas for the four suits, and that is because the base is so strong without fooling with the areas that I’ve never even felt the temptation to change it.
Dale Y: (about 5 plays) – I’m on the fence whether I love this one or not. The idea of being able to declare yourself out of a suit whenever you want is so out there – it really F**** with your head. In one sense, you have to wrap your head around a wholly different paradigm of trick taking game – it’s hard to know which tricks you can win, and heck, sometimes it’s hard to know which tricks you WANT to win. But, there are definitely times when the cards that you are dealt leave you no option but to try to spoil other player’s board positions as they simply don’t leave you much opportunity to do anything interesting on the board. Who cares if you double a group of 2 on the board? That being said, I’d never say no to this game, and I predict that I’ll end up picking up a copy of the new version this summer when it is available to me.
Dan B. (3 plays):I like the idea a lot, and it’s worked well enough so far… but I can’t shake the feeling that it’s solvable in some sense, i.e. that with more experience a player with a certain hand distribution can dictate the outcome. But I could well be wrong. And it’s definitely better than dois, which I am much more certain has issues.
Ratings from the Opinionated Gamers
- I love it! James Nathan, John P, Rand L, Chris Wray
- I like it. Dale Y, Dan B. (for now)
- Neutral. Joe H.
- Not for me…