Beyond Poker reviews games which are not poker, but which scratch the same itch or exercise similar strategic muscles. In this series you’ll find games which involve hidden information and bluffing, probabilistic thinking, risk-reward estimation, or which simply apply poker concepts like the hand rankings in a novel context.

So far, in Beyond Poker, I’ve been looking at single-player digital games which make use of poker hands and probabilities as part of their mechanics. Now, I’d like to do something different and look at a physical, multi-player card game which bears very little superficial resemblance to poker, yet exercises some of the same parts of the brain. The game in question is Welcome to the Dungeon and the specific resemblance I see between it and poker is that it requires “hand reading,” that is, interpreting opponents’ decisions to make inferences about which cards they’ve seen.

Welcome to the Dungeon is a 2013 release from IELLO Games, a company probably best known for the dice game King of Tokyo. It’s a translation and re-release of a Japanese language game called Dungeon of Mandom, designed by Masato Uesugi and published by Oink Games; I assume the change of name has something to do with the fact that Dungeon of Mandom sounds, to the Western ear, like a gay S&M porn site… but maybe that’s just me.

In any case, IELLO Games has done a great job with their version; like King of Tokyo and their other games, they’ve taken something simple and applied really high production values to it to produce a gorgeous product at a very affordable price (US $14.99 MSRP, $11.99 via Funagain Games). It comes in a small box and consists of a couple dozen thick cardboard tokens, a tiny deck of cards, and not much else, so it’s very portable; if you wanted to, you could easily repack the entire game into a standard two-deck playing card box.


The basic concept of Welcome to the Dungeon is simple, and may even seem too simple at first glance. This is Oink Games’s modus operandi, however; all their titles make use of a minimal number of components and rules in order to produce a game that nonetheless taxes the brain, usually by leveraging the complicating power of hidden information.

In each round, there’s a single adventurer who is going to go into a dungeon, and he’ll either kill everything inside or be killed himself. The players are essentially playing a game of chicken with one another, making the challenge progressively harder for the adventurer, while testing their belief about whether or not he can succeed. Ultimately, one player will take the adventurer’s side after everyone else has bailed out, and either share in his success or suffer the consequences of his failure.

There are four different adventurers, each with a certain number of health points (2, 3 or 4) and his or her own unique set of six equipment tokens. Most of these either give the adventurer additional health points or grant immunity to one or more specific monster types, but some have more complex and subtle powers. For the first round, the adventurer is chosen randomly, after which the player who accompanied the last adventurer into the dungeon chooses which one is up next.

The monsters, meanwhile, are represented by a deck of 13 numbered cards, with each value corresponding to a certain monster type, with higher values being scarier monsters. There are two copies of each monster numbered 1 through 5, and then one each of 6, 7 and 9 (there is no 8).

Each player’s turn consists of just two decisions, each quite simple: First, the player must choose between backing out of the round or drawing a monster card (provided any are left in the deck). Second, if the player drew a monster card, she looks at it, then decides either to add the monster to the “Dungeon” (a face-down pile of cards in the middle of the table) or remove it and one of the adventurer’s equipment tokens from play. In the latter case, the player puts the card face down in front of herself, with the equipment token on top of it, to remind everyone in case they forget who removed what. Under no circumstances, though, do the other players know which monster was added to the dungeon or removed from play.

When a player elects to back out (or is forced out because no cards remain in the deck), he’s done for the round. His turn will be skipped, and he no longer has anything to gain or lose from the round’s outcome. The round is resolved when only one player remains, at which point he or she will accompany the adventurer into the dungeon.

Starting with the top card of the dungeon and proceeding one card at a time, the monsters in the dungeon are revealed. Monsters which can be defeated with one of the adventurer’s remaining pieces of equipment are set to one side, while those that cannot go in another pile. The values of undefeated monsters are subtracted from the adventurer’s health point total (which includes health point bonuses from equipment). If, after all monsters have been revealed, the adventurer’s remaining health is zero or less, the adventurer loses, otherwise he or she succeeds.

The player who ended up siding with the adventurer then gets a card indicating whether he received a success or a failure. A player with two successes wins the game immediately, while anyone with two failures is eliminated from the game. This means that there are two ways to win, either by achieving the two successes necessary, or by surviving until all opponents have failed twice and been eliminated.


Much of the strategy of the game centers on the equipment tokens, as which ones have been removed is one of the only two pieces of public information available – the other being the number of cards that have been put into the Dungeon so far. To be able to discuss strategy in clear terms, then, I’ll need to provide the details of the simplest of the four adventurers – the Warrior – and his equipment set.

The Warrior has 3 health points intrinsically, plus boosts of +3 and +5 from his shield and armor, for a total of 11 to start. His other four pieces of equipment are as follows:

  • Torch: Automatically defeat all monsters valued 3 and less.
  • Holy Grail: Automatically defeat all even-valued monsters.
  • Dragon Spear: Automatically defeat the Dragon (value 9).
  • Vorpal Sword: Pick a number before beginning to flip cards. Defeat all monsters of that value.

As you can see, the Warrior could easily survive all monsters in the deck if he was allowed into the Dungeon with his full set of equipment; the only ones he is vulnerable to from the start are the two 5s (“Golems”) and the 7 (“the Demon”), and he can take one of those two out with the Vorpal Sword and be left with enough health to survive the other(s). This means it’s important to remember that if there are no cards in the deck on a player’s turn, she’s forced to back out; if no equipment were removed, the last player to add a card would end up winning the round by default after the others back out, so at least one of the other players must remove a piece of equipment at some point to prevent this from happening.

You can also see just how much things can change with the removal of a single piece of equipment. Imagine, for instance, that you’ve already taken away the Shield, leaving the Warrior with only 8 health. If I now remove the Dragon Spear, whoever ends up accompanying him into the Dungeon either has to hope that no one put the Dragon into the Dungeon, or else hedge by choosing to “slay 9” with the Vorpal Sword but leave himself vulnerable to Golems and/or the Demon as a result.

As with poker, there are levels of thinking in Welcome to the Dungeon. Level one is that you want to maximize your own information advantage by making your moves so as either to maximize or minimize the chances of the adventurer surviving. Moves which produce a more “average” difficulty Dungeon are of little use to you, as you will have much less idea of whether to stay in or run away on your next turn.

Therefore, if you’re going to remove the Dragon Spear, you’re probably going to want to know where the Dragon is; if you haven’t drawn the Dragon yourself, someone else may have, so removing the Spear would be risky in that it could give an opponent more information than it gives you.

That brings us to the second level of thinking, however, which is interpreting the actions of the other players. Since you want to know where the Dragon is before removing the Dragon Spear, you can deduce that a player removing the Spear has probably either put the Dragon into the Dungeon on a previous turn, or is removing it along with the Spear. Assuming that the player is balancing between those two strategies, you still don’t know yourself whether the Dragon is in or out, but the knowledge that he’s probably seen it can help you make other inferences, such as what monsters other players are likely to have seen based on their actions, and whether to use the Vorpal Sword to protect against the Dragon if that player backs out on his next turn. From there, you can of course go to all the higher levels of thinking present in poker; bluffing, double-bluffing, establishing a risk-taking or conservative image to exploit in later rounds, and so on.

Even more interestingly, though, once you get beyond the first round, the various outcomes aren’t necessarily of equal value to all players. A player who has already suffered one failure should be looking to back out early and avoid elimination, while a player with a success and no failure may be hoping to be sent into the Dungeon for her shot at a second victory and an instant win. Thus, you can guess that most of the time, the former player will be looking to increase the Dungeon’s difficulty prior to bailing out, while the latter is more likely to be keeping it easy with the intention of staying in. This isn’t exactly analogous to ICM considerations in poker, but it’s not entirely different either, in that the game dynamics force some players to emphasize survival and others to take a more aggressive approach.


As I said in the introduction, Welcome to the Dungeon has exceptional production values for its price tag and is super-compact. The number of components in play at any time is also quite low, so you can play it almost anywhere – on a plane, in a coffee shop, lying in bed, etc. – and the playtime is very short, anywhere between five and 30 minutes depending on the number of players and how things play out. It’s very convenient in those ways.

In terms of skill factor, I think the skill-to-luck ratio is very high for beginners, but I’m more skeptical about the game’s long-term skill ceiling. I have yet to lose a game, and imagine that I will continue to win a very high percentage of the time against inexperienced players. The game tree is quite small however, so I suspect that after a few dozen plays and a bit of study, a sufficiently smart player will be limited more by the ability of the human mind to randomize properly in executing a mixed strategy than by the intricacies of the game itself. Certainly, it seems like the sort of game that could be solved completely and in a short period of time by the sort of machine learning algorithms currently being used to attack poker.

Played with more than two players, it’s also highly dependent on seating order and vulnerable to collusion. This is true of the vast majority of multiplayer games, however, so I wouldn’t count it as much of a strike against Welcome to the Dungeon, but mention it mainly because this is a poker site and feel I should caution you against playing it for any significant amount of money, unless you’re playing heads-up.

So, in the pro column, we’ve got elegance of design, affordability, portability, convenience, short play time and some interesting strategic puzzles to work out when you’re learning the game. For cons, it’s mostly just a lack of depth or potential for long-term serious play. All of those factors combine to make it a perfect game for travel, and for poker players who want to dabble in something else but aren’t looking to commit very much time or money to a new hobby. You can grab it for cheap, toss it into a backpack or suitcase without needing to make much room for it, and it will definitely keep you and your friends entertained through a plane flight or while killing time at the hotel, whereas it might eventually get a bit stale if you played more frequently than that.

Alex Weldon (@benefactumgames) is a freelance writer, game designer and semipro poker player from Montreal, Quebec, Canada.