Teach Your Kids Bridge
A post recently made the rounds on hacker news claiming that you should teach your kids poker, not chess. The comments on that post go through a lot of the reasons why poker is a bad game to teach your children, but I felt that I was well suited to opine on this topic, and explain why duplicate bridge is the best game for practicing the life skills involved in business and programming, compared to all of the alternatives.
I am a passionate player of strategy games, but never a professional player at any. I played chess as a child, ultimately studying to the point where I can play at ~1500 play strength today. In high school, I was a passionate Magic: the Gathering player. I played a lot at the competitive level and also became a level 2 Magic Judge for a brief time. In college, I started to seriously play bridge and poker. When I turned 21, I made money playing Texas Hold’em and Pot Limit Omaha (the latter of which I enjoyed more, but had fewer players). In college, I played bridge with the college club twice a week, and became decent at the game. Today, I am mostly a bridge player, but I still play chess, Magic, and poker once in a while. I believe that bridge is the best game for learning life skills, and that it is a great game to play as an amateur.
The Game of Bridge
Bridge is unfortunately associated with an aging population of players and is a lot less “cool” than chess or poker. However, there is a small community of young players that are both friendly and competitive, hopefully the benginning of a renaissance for the game.
Bridge is played with four players, two teams of two, using a standard deck of cards. Each player sits across from their teammate. On each hand, the entire deck is dealt out, and then the game goes through two phases: bidding and playing.
During the bidding phase, players call out public bids to determine the contract, exchanging information about their hands through the limited information of a number and a suit (“one spade” is a bridge bid). Information exchange occurs during the bidding phase using agreements about the meaning of each bid called bidding conventions. Partnerships can come up with their own agreement or use one of a few common bidding conventions, like Standard American, 2-over-1, and ACOL. Learning bidding conventions is like learning chess openings: you need to spend time learning one of them.
During the playing phase, one player from the partnership that won the bidding places their hand face up on the table, and the other player from that partnership plays a trick-taking game (similar to hearts or spades) against the opposing pair to try to take as many tricks as possible. The person playing the contract is called the “declarer,” while the other two players are “defenders” (the player whose hand is face-up is the “dummy”). The playing phase is then scored based on whether the declarer has made the number of tricks promised in the contract. Taking as many tricks as promised (or more) is called “making the contract,” and offers rewards that depend on the level of the contract and the number of tricks taken, while taking fewer tricks is “going down.” The defending side is rewarded for making you go down. Playing the hand is often subtle and challenging, regardless of whether you are defending or declaring, and every trick matters regardless of whether you are expecting to take 10 tricks or 2.
The Benefits of Bridge
Information management is central to bridge. During the bidding phase, you and your partner determine what is important to share with each other to determine a contract. During the playing phase, you have to work with the fact that two other hands are not publicly known. Probability comes into play here: understanding distributions of cards and possible hands that opponents can have will allow you to figure out how to play for the maximum number of tricks. The bidding and playing phases are all about deriving the maximum value you can from the limited information you have, much like life.
Planning a Line
Like chess, bridge play often involves thinking several tricks ahead and planning your line of play. However, you have to plan for contingencies and plan to mitigate the effects of the unknown information. Because you know that the entire deck is dealt out, there is often an understandable amount of missing information. There are 13 tricks in a hand, which is long enough to plan a line based on a few critical decisions, but short enough that you can reasonably work with the probability space.
There is a tremendous amount of depth to bridge play strategy: For example, squeezes and endplays are examples of ways to play a bridge hand based on giving an opponent options for how they want to give up a trick to you. To execute an endplay or a squeeze, you have to plan for a lot of possible hands and outcomes.
Additionally, there is a lot of strategy on the defensive side, including strategy on how to make an opening lead and strategy around how to give information to you partner.
Asymmetry, not Randmoness
The cards that are distributed around the table are not distributed fairly. The game of bridge teaches you to deal with the cards you are dealt and maximize the number of tricks your partnership takes in every circumstance. Due to duplication and the contract scoring system, one trick often matters a lot, and playing weak hands well matters as much as playing strong hands well.
Bridge games are set up so that the aspect of “who has the good cards” is completely mitigated. Bridge is often played “duplicated”: many tables play with the same sets of hands, and your result is compared against other pairs (at other tables) with the same hands, not the other pair at the same table. In this way, playing weak hands well is rewarded, while playing strong hands poorly is punished. The randomness of “who gets the cards” is mitigated in the scoring. You can still lose a hand to someone who takes a different (possibly worse) line of play and happens to catch the cards in the right places, but the skill difference tends to average out over 20-30 hands, which is normal for a bridge tournament.
Bridge is a partnership game, and you are going to have to learn to work with your partner. This means learning your partner’s tendencies, learning how to work with others, and learing to become a good communicator. Like a team game, your individual skill is only half of your team’s success. Bridge will also involve miscommunications and mistakes, and so you will also learn to ask for forgiveness and to forgive others for their mistakes.
Partners and teams will often have a postmortem after a bridge game where they review the hands and learn from them. The postmortem will help you learn to give and take criticism constructively (although postmortems are not always constructive).
Learning to work with bidding conventions and learning to recognize certain types of tactics during the play of the hand are important for being a good bridge player. This kind of preparation rewards you the same way that learning chess openings rewards you. Crafting new bidding conventions or modifying the ones you use with a partner are also important for maximizing your chances of winning. Many different kinds of bidding conventions are out there, and you can learn a lot from learning about them and thinking about them.
Comparison to the Alternatives
In comparison to bridge, there are several alternative games on offer, including poker, chess, trading card games, and other board games. I am going to be categorizing these on a few axes:
- Imperfect vs perfect information: In a perfect information game, both players have the same information about the state of the game.
- Asymmetric vs symmetric: In a symmetric game, both players are playing with the same objectives and set of pieces. Asymmetric games have different objectives or different pieces.
- Luck vs skill: In a luck game, some information about the game state that is hidden from all players (eg with dice rolls or facedown cards).
By this categorization, bridge is an imperfect information asymmetric skill game. It is very unusual to have imperfect information but no luck in a game, since imperfect infomration is often derived from luck. Bridge derives 100% of its information gaming from asymmetry, and mitigates luck through the way it is scored. In my opinion, it is hard to have a richer strategic space or a better learning environment than you can find in an imperfect information skill game.
Because we are talking about games for kids, I will also be talking a little bit about the communities around each game, as I have observed them. One point against bridge is that the bridge community as a whole is not the friendliest or the most welcoming. However, you can find players, particularly younger players, who are happy to play with new players who are learning the game.
Imperfect information symmetric luck game
Poker is a game of information, but poker is also primarily a gambling game: money and gambling are inseparable from poker. People will play poker differently when they are playing for large amounts of money than when they are playing for small amounts of money: they are a lot more conservative when a lot of cash is on the line. Due to the nature of pot sizes in poker, people in small games often find themselves making big-money decisions and vice versa. However, if you are playing for no money, that dynamic disappears and the game changes a lot. Unless you want to introduce gambling to your children, this fact alone disqualifies poker.
A lot of people think that high-level poker is a lot more about “tells” and “reading your opponent” than doing math. This is not the case. Playing pot odds poker (straight math) will often earn you money at low-stakes games because people look too hard for tells. High-level poker players will talk and think a lot about “ranges” and probabilities. A “range” is a set of hands that would act in a particular way, and high-level players understand that betting at a poker table is more about your range and your opponents’ ranges, and how you stack up.
Since poker involves luck, when you lose or win a hand, you have the task of separating out your skill from your luck. In all luck games, people are biased in favor of attributing your wins to skill and your losses to luck, and it takes a lot of discipline to avoid this fallacy. However, poker does involve a lot more skill than most luck games.
Perfect information symmetric skill game
If you want to become a professional at a strategy game as a child, I would personally suggest chess. There is a tremendous amount of strategy to study around chess, and at the high levels it actually has a lot of the “information gaming” aspects of poker or bridge. High-level chess involves a lot of learning and preparation, and you can gain an advantage against another player by playing an opening or aiming for a position that you have studied and they have not.
However, at lower levels, chess games are mostly determined by who makes fewer blunders. There isn’t much of an information game when you sometimes leave pieces open for your opponent to caputre using basic tactics and combinations. It takes a tremendous amount of thought and attention to play the game to even this level. High-level chess is an imperfect information bluffing game, but lower-level chess is about avoiding mistakes with perfect information.
Finally, there is a vibrant community of serious chess players everywhere, and there are a tremendous number of resources for people who want to go from “serious casual” to “professional.”
Perfect information symmetric skill game
I don’t know much about how to play go, but I understand that it is similar to chess: the low-level game is about avoiding mistakes, while the high-level game has components of strategy and preparation.
Trading Card Games
Imperfect information asymmetric luck games
Trading card games like Magic: the Gathering and Pokemon are often marketed as being “combinations of poker and chess.” I have generally found that not to be the case. In my experience, randomness determines the outcomes of games more than skill or information asymmetry. Generally, players start with a hand of cards and draw one card per turn from the top of their deck. This trickle of resources both creates luck and limits the asymmetry of information: you only get one new card per turn to bluff with, and you can find yourself seeing a good mix of resources or being starved of one resource or another.
Good magic players have some ability to limit the effects of randomness, but much less than in other luck games on this list. The top ranked players by the MTG Elo Project have win rates below 66%, and their “GP records” (games played in “Grand Prix” tournaments, where anyone can enter) generally have a win rate below 70%. These highly skilled players cannot crack a 75% win rate over random players! Furthermore, these are match wins: their game win rate would be closer to 50% than to 75%, since best-of-three matches have the effect of decreasing the effects of luck.
Futhermore, trading card games are expensive to play, with a single competitive Magic deck costing $300-2000. In an environment where “the best” deck changes every 3 months, there is a tremendous amout of money required to play seriously.
“Limited” format tournaments, where the cards are provided for the tournament, are also fairly expensive because buying packs of cards is expensive, and packs don’t hold their value once they are opened. For example, a sealed Grand Prix usually has a $100 entry fee, and on average you will walk out with <$10 of new cards. A casual draft may have a $25 or less entry fee, though, and I find these to be the best way to play magic.
One final note: competitive Magic players are often a lot less friendly than compeititve players of other games. I don’t know why this is. Depending on who you ask, they may also be less honest: compeititive Magic has had a lot of cheating scandals in its recent history.
Imperfect information symmetric luck game
Scrabble is a great casual game that can teach your child a trememdnous amount of vocabulary, particularly words that are two and seven letters. I learned such words as “aa,” “cwm,” and “aorists” playing Scrabble with my grandmother.
Like trading card games, Scrabble is an imperfect information game played between two players with an aspect of randomness. However, high-level Scrabble players are really good at removing variance from the game by opening up fewer spaces where new words can be played. The fact that the dictionary of words can be completely memorized also helps. Like poker, high-level Scrabble players know how to mitigate luck to the point where it is almost a skill game.
I have never played competitive scrabble, but I have heard that the competitive scene is the most welcoming of any game.
Give Bridge a Try
Compared to the alternatives, bridge offers a rewarding experience that combines asymemtric gameplay and imperfect information while avoiding the effects of luck. This makes bridge a perfect game to teach a growing young mind. They will learn to deal with the asymmetric circumstances of life and others with hidden information, while also teaching about planning lines of play and the value of teamwork.
A lot of the alternatives are great games, but if you are interested in teaching and learning useful life skills through the platform of a game, there are few alternatives that offer the richness of contract bridge.