Hacker News new | past | comments | ask | show | jobs | submit login

It's a shame that Bridge's popularity is waning because it is a very interesting and skilful game. It suffers from having a relatively high barrier to entry. A lot of comments here are from people who clearly have never played the game.

In bidding there are 48 calls you can make (1-7 of each suit plus notrumps, X ie double, XX ie redouble and pass). The bids (1-7 of each suit and no trump) are ordered meaning once someone bids 2D you have to bid 2H or higher so you've lost the 1 bids and 2C). The two teams are interleaved and are trying to convey information to your partner but your opponents have access to that information too.

It's kind of amazing how much information you can convey within this simple system. It's not even just about what bids were made but what bids weren't made. Bidding is a deep, deep topic.

Additionally you can lie to your partner (within limits) and in competition play there are rules around what you can and can't do. This can confuse your opponents but will also confuse your partner (if it doesn't it's often ruled as "illegal information").

After all this and the opening lead the dummy hand is placed on the table. Every player can see it. The defending team has the auction to inform the defense. So it's not just a simple trick-taking game. The auction and the play are inextricably intertwined.

The defenders can convey information to each other through which cards they play. The declarer can see this too but has less context on the meaning.

In both the auction and the play Bayesian reasoning comes into play. What didn't happen? How does that affect the probability space?

Building partnerships takes time although experienced players can sit down with a new player, agree on some fairly standard conventions (eg 2/1, upside down carding) and be at a reasonable point.

But the high barrier to entry, fixed numbers of players and the reliance on partnerships make it a tough game to play casually.




It sounds like if two players on a team have a lot of experience playing together they probably develop an intuition of each other, and I would guess this is viewed as an important part of the most successful teams. It also sounds like if those two players attempted to codify a communication pattern, like "I will play cards this way, which means this" it would be ruled as illegal? It sounds like a game of cards, and game of communication, and a game about pushing the rules and the moderators, like me and my partner need to learn to communicate illegal information, but we cannot do so in a way that is detectable, and if we're people of integrity, we wont make communicating illegal information our explicit goal, but our practice together is always helping us be a little better at it, but we maintain deniability and integrity because we never explicitly planned it.

Am I way off here? I'm guessing all these subtleties make online play difficult as well?


Carding conventions like bidding conventions are public. Let me give you some examples.

When defending a trump contract and you, as a defender, lead a side suit, what card you lead describes your holding. An example is you tend to lead top from an honor (AKQJ are honors) sequence. So if you have QJxx you'll tend to lead the Q. Doing so denies the K (as KQxx has a different lead). It may or may not deny the A depending on what you've agreed upon. So with AQJxx you might lead Q ow low depending on your convention.

Likewise your partner's card tells you something. For one you need to know if they're carding count (how many of that suit they have) or attitude (whether they want you to continue that suit or shift). So upside down attitude means you respond with a low card if you like it (standard ir high as a positive signal but most consider that an inferior treatment as you may waste valuable high cards). The leader can infer the meaning of your signal based on their holding and what's in dummy. Like the leader might have 32 and partner plays the 4. Leader knows this is a positive signal. Declarer who cannot see your 32 might not know this.

This is an example of imperfect information mentioned in the article.

Declarer knows what conventions defenders are playing. Hidden understandings are illegal. You can lie to your partner but if, say, you discourage a lead and your partner continues it and that's the winning play then you may get penalized for that in competitive play.


Pinochle has these characteristics, and team play definitely improves with understanding your team mate better.

One time, for a bit of fun, my team mate and I were looking for a bit of an edge. 4 suits = 2 bits = two eyes open or closed, one combination for each suit.

We triggered reading the eye lid states as particular points in the game were happening. A query was triggered by a shared set of words.

The opposing players were quite adept and we had to take measures or we would be caught!

We confessed it after a modest but fun string of great games and the opposition, who are good friends, thought it clever and pretty funny for us to end up wanting wins badly enough to develop the scheme!


I used to be a bridge tournament director, and have spent a bunch of time working out interesting ways of cheating (which is quite easy, given that very little information needs to be passed for a good player to have a massive advantage).

My favourite 'famous' method of cheating reveals the number of Hearts in hand (mod 4) by the number of fingers visible in front of the fan of cards.

My favourite 'own' method relies on colour differences between similar packs of cards. I discovered a couple of packs of the club's cards had been dropped and reassembled into complete packs but making the subtle difference in the blue colour on the back obvious. If you did that deliberately, with all the hearts swapped between the two packs, you'd have a complete read of heart length in all four hands, which would make life much easier. Wouldn't even need a confederate partner, and if it was discovered nobody would know it was you...


Nice!

The color difference is ultra sneaky! Well done. I like everything about it, minus the need for acute color perception. But not me. Mine is solid. And perhaps it can depend on other artifacts.

I love these kinds of exercises. They are stimulating and good over all skill builders to keep a person sharp.

Thanks for sharing that.


If your partner has revealed information to you, you are required to alert your opponent to that fact.

In practice having better communication is still a big advantage, but it’s not because of an information asymmetry (for honest players).


>But the high barrier to entry, fixed numbers of players and the reliance on partnerships make it a tough game to play casually.

At various times, I've played a lot of different games. And Bridge always seemed like one that was more of a commitment than I felt like making at a given moment.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: