Decisions Based on Incomplete Information
Chris Russell
One of the most common compliments I hear from L5R players is that it’s “the chess of CCGs.” These people are referring to the game’s strategic depth, one of its main selling points. Maybe it’s the humanities major in me, but when I hear that, I start thinking about ways in which chess and L5R are more different than alike, and the difference that immediately comes to mind is complete versus incomplete information.
Complete information is a term used in game theory to describe a game in which all players know at all times exactly what all other players can do. Chess is one of the classic examples of a game of complete information; no matter what you do, you cannot keep a secret from your opponent in chess. “But wait!” you might be saying. “What about promotion? I can trade in my pawn for a queen, or a knight, or even a rook or bishop!” Really, this is only a case of having multiple strategies available to you and is no different from choosing to move your rook forward instead of backward.
Like every other CCG I’ve played, L5R is almost never a game of complete information. Cards in one player’s fate hand are concealed from the other player, the specific cards in one player’s deck may be unknown, and the order of cards in any deck is unknown to every player. Many cards are additional sources of incomplete information, from High House of Light to face-down Ninjutsu personalities. In the most likely game state in which all players have complete information, neither player will have a fate hand, face-down cards in provinces, a fate deck, a dynasty deck, or a way to replenish either deck. In short, when L5R becomes a game of complete information, you probably won’t want to play anymore.
The impact of incomplete information on L5R is not obvious – or rather, it is so obvious that most players do not consider it. How many times have you wanted to attack your opponent but chosen not to do so because it seemed risky? He had four more cards in hand than you did, and none of your cards was very useful at the time, so you just bought personalities and hoped to draw into something good. Think of all the battle actions he could have had! The battle could have been a disaster!
“Could have” are the key words there, of course. A hand of three First and Final Strike is every bit as intimidating as a hand of three Focus. Stepping back, you will see that your opponent only knows that you do have three cards and that you do not have your cards in play and in your discard pile. Your other actions (whether you attacked last turn, which personalities you kept unbowed, etc.) may give your opponent hints about what you have, but he is missing a powerful weapon: certainty. With certainty, a smart player will rarely walk into trouble.
Emperor’s Under-hand was one of the best cards in the Gold arc precisely because it provided certainty. If you used it to look at your opponent’s hand and he could not draw any cards (card draw was rare in Gold), you had complete information for one turn. At one tournament at the end of the Gold arc, I watched a game in which Conrad Jackson was down to two provinces against a large but mostly bowed-out army. He used the Under-hand to study his opponent’s seven-card hand and the board for a few minutes, then give his opponent back his cards. He followed it up by killing the largest of his opponent’s three unbowed personalities and launching an all-out attack at four provinces. What would have been a very risky play before using the Under-hand was now the only correct one; Conrad had seen that there was no way his opponent could keep him from winning the game on that attack. To reinforce an earlier point: without seeing Conrad’s hand, the other player couldn’t be certain he was going to lose the game that turn, but Conrad’s choice not to discard a card before attacking must have been a powerful hint.
Emperor’s Under-hand is gone now, thanks to a post-GE design directive that I call “thou shalt not peek.” Even so, smart players can exploit smaller information imbalances when the opportunity presents itself. This is especially meaningful when playing a deck such as Shizuka Toshi honor, which forces its opponent to make difficult choices at least once per turn. To look at the relevance of information gaps here, take the situation at its extremes. If the Crane player has no hand, the opponent only has to consider cards in play to make his decision; a full hand means the Crane player could potentially do all kinds of unpleasant things to a dishonored personality.
Aside from that extreme circumstance, even small tidbits of information can be valuable if they come at the right time or if they influence a large number of choices. Let’s say that you control two copies of Doji Ayano. You have a four-card hand, including The Bitter Shadow of Shame, which you want to play on your opponent’s personality, but you have no way of dishonoring him outright. You hope that your opponent chooses dishonor when you use your stronghold, but first you want to draw a card with Shrine to Fukurokujin. You bow the Shrine, and you flip… The Bitter Shadow of Shame.
By using Fukurokujin too early, you ceded an advantage; your opponent now knows with certainty that you play Bitter Shadow and that you have it in your hand. He can use that information – which, before using the Shrine, only you had – to make better decisions for the rest of the game. It is now much less likely that he will make the choice you want him to make when you use Shizuka Toshi this turn. If you knew from the start that you were going to use Shizuka Toshi, there was no reason to use Fukurokujin first and risk making your opponent's decision easier. As a general rule, you want to be careful to share information with your opponent as late and as rarely as possible.
Of course, it isn’t only Shrine to Fukurokujin that reveals hidden information. Sometimes it’s the card that was discarded, the reaction that was played, or the decision not to lobby. Let’s continue with the last example. You Palm Strike your forehead, but you can still get three honor out of your box. You go ahead and use it, and your opponent immediately turns his personality upside down.
What now? He is telling you that, based on his knowledge, he would rather have you play Bitter Shadow on the personality you targeted than give you three honor. Does he have death negation in hand? Is the personality not as valuable as you thought he was? Does he want you to bow an extra personality and/or pay the four gold? Is he bluffing? Just not paying attention? Those are questions you have to answer based on what you know about the current game state – and there’s no Under-hand to help you out.
In any given game, you should be looking for ways to use hidden information to your advantage – essentially, to make your opponent do what you want him to do. In that vein, here’s a little puzzle for you:
Brash Hero
You are playing Crane out of Kyuden Kyotei. You are at 38 honor and one province.
It is your opponent’s action phase. He is playing Vigilance Keep and has one card in hand, a Sneak Attack he drew with Shrine to Fukurokujin. His only personalities are two Matsu Yoshifumi (3F) and Matsu Benika Exp. (4F). The last card in his dynasty deck, Akodo Terumoto, is face up in a province. He cannot possibly reach 40 honor.
The only personalities you control are Doji Ayano and Doji Tsubakita (whom you made a courtier). Two Kabuki Theater Troupes will put you to 40 honor next turn, no matter what. You also control one bowed Governor’s Court, which you just bought.
Your last dynasty card is face down in your province. You know it is a holding, but your opponent does not.
You have one card in hand, Strength in Certainty. You have already played your other two copies. Thanks to Temples of Gisei Toshi, you know that your draw will be useless next turn.
Assume that I have given you all the relevant information. Which of your personalities do you bow to use Governor’s Court, and more importantly, why? There may not be an objectively right answer to the question, but I know what I would do. Post your answers (along with anything else you'd like to say) in the comments.
