Board Games and Markov Chains | Hacker News Reader

Board Games and Markov Chains

kevingal.com

82 pointssigna1128 comments

> which involve no skill or decision-making or social engineering or anything that a reasonable human might associate with the word "fun"

What's wrong with playing games that aren't mentally exhausting? It's about having fun, not about trying to be good at having fun. It's impressive what some people can do in more complex games like Factorio but for me that niche is filled with coding. Can be fun as well and in a different, more fulfilling way, but it needs attention and energy.

Some people seem to get offended when others have fun the ‘wrong’ way.

I think the first paragraph fundamentally misunderstands why young people play those games. The challenge is in learning how to follow rules and take turns. And it’s plenty fun for my 3yo and 5yo, who, like any human of any age, find something they’re not yet trivially good at a satisfying challenge.

Also such games are a kind of automatic "story" generators. You're in a non-linear, non-predictable race, that will automatically create exciting emotinal moments, where one is ahead, but dropping back, or someone's overtaking finally etc etc.

These techniques are also useful for parts of more complicated games, like battles in Risk.

That said, I agree with the author that it's weird how little social engineering people put into their games. I refuse to play monopoly or Risk if the house rules prohibit entering complicated agreements with other players like forward contracts, lease-to-purchase agreements, variable-interest rate loans, swaps, and what have you.

I still haven't gotten anyone to play with me under free banking rules where everyone is free to print their own currency, but I'm really excited about the idea.

I have literally lost a friend over a game if Monopoly 10 years ago. I can't imagine going through the same ordeal with those rules! Although since I work in Finance the thought excites me!

You cannot do this with people who take the game too personally, obviously!

Since a large part of it relies on trust (a loan without collateral puts you at huge counterparty risk) it sort of only works if you play repeatedly with sort of the same people, so they have an incentive not to screw you over.

I usually play Monopoly with Credit Default Swaps, and the ability to repackage properties into CDO's. When I play Risk, there's an added element of forex which makes the game more interesting. But all Risk games inevitably end up with someone saying "I'm not beaten yet. I still have armies in the Ukraine!"

I refuse to play monopoly because it’s a horrible game.

It was literally designed to be a horrible game: it was originally a teaching tool for Georgist economics, attempting to illustrate that rent extraction made one player rich and everyone else broke. Then you're supposed to flip over the board and play "Prosperity", the alternative economic system in which all players prosper.

https://en.wikipedia.org/wiki/History_of_Monopoly

And many older people don't like boardgames because it is the only one they know. Fortunately, game designers like Teubner (Catan) knew how to make fun games.

What have you tried to do to make it more interesting? I find that when a group wants to play a game I don't like, it's significantly easier to swap in some fun house rules than swap out the entire game.

How would printing your own money work in Monopoly? Since the objective is to bankrupt all opponents, there is never an incentive to accept their currency as payment.

They will probably offer their currency at a discount to incentivise to take it anyway!

Of course, that promise is only credible if they aren't near bankruptcy anyway, so I suspect that specific rule won't change much in how the game is played.

(Things like selling shares in one's properties, on the other hand, makes it possible for the game to go on for a really long time. Some games turn into complicated interlocked arrangements where nobody wants anyone else to fail lest the whole thing comes tumbling down -- much like in the real world. At that point the game is less about winning and more about finessing your way around a financial system.)

The fun of those games is playing together as peers.

And never being told you're doing it wrong.

And a level playing field where you might actually win.

The pleasure is an absence of common pains.

YMMV.

I distinguish between "family games" and "gamer games" on whether you have gaps of time where you can ignore the game and interact socially with the other players or need to concentrate on the game. The former group is mostly an excuse to socialize, so game mechanics are largely irrelevant.

Scotland Yard and Pandemic are the two best cooperative board games I know of

Pandemic suffered from the orchestrator effect for us, where the best player just ended up telling everyone what to do.

Lots of co-op games struggle with this; some address it by simply adding so many details and complications that no one individual could possible track and orchestrate everything (the one that comes to mind being "Spirit Island"). Some games instead use a strict limit on communication ("Hanabi", "Regicide"), though that may defeat the social element for some.

Yeah what worked for us was limiting the kind of requests you could make - no “show me all Your cards” type and more “we need to defend X who wants to do that” type.

It takes a lot of people to shine, but Captain Sonar has great team co-op mechanics AND an extremely clever competitive mechanic in the radio operator.

Low skill board games offer you a chance to win when you’re low skill. It’s a pretty rare thing. The possibility to beat everyone else due to good fortune is pretty appealing. We see the same thing with high variance barrel royale video games. People love gunning for those low odds to win everything.

Related: nice visualisations of game graphs https://writings.stephenwolfram.com/2022/06/games-and-puzzle...

If you enjoyed this, you might like some of the literature around boardgame research that uses Markov Chains. Prime examples are RISK and Monopoly, but someone wrote a thesis on "Analysis of 'The Settlers of Catan' Using Markov Chains" as well. I maintain a nice list at https://github.com/captn3m0/boardgame-research (grep for Markov).

The DeepMind course on Reinforcement Learning covers Markov Decision Processes (MDPs) quite well in this lecture: https://www.youtube.com/watch?v=lfHX2hHRMVQ

A more detailed analysis with python and math:

https://jaketae.github.io/study/markov-chain/

This post explains Markov Chains using an example 3 state system whose only resemblance to a board game is that state transitions are made from a single die roll.

>The only pleasure I can imagine being derived from such games is the trickle of dopamine that comes from winning and the schadenfreude that comes from seeing a snake devour your opponent, which in my considered opinion is setting children up to become inveterate gamblers and psychopaths, respectively.

winning through nothing but pure chance is a good life lesson: why do others win? because they are lucky

snakes are fun slides, not devouring monsters. players are not killed or defecated. another good life lesson: if you fall down, get back up

the rules are simple - you cannot play monopoly with infants

we play these games with children because, aside from the hidden life lessons, they are fun and promote bonding

the author is lacking in many respects

Another nice post in the same direction by Jake VanderPlas:

https://jakevdp.github.io/blog/2017/12/18/simulating-chutes-...

Reply on news.ycombinator.com