next »

[battlerager]

Hello people!
I started a new series of game design articles over at http://www.dopterra.com

The first article is an introduction on the broad topic of randomness in games (and why I think it's important to think about it) focusing mostly on card and trading card games.

Randomness, Part 1 – Luck of the Draw



This is my first game design article, so it has its flaws. Sorry! Any (honest and brutal) critique, discussion and feedback is super welcomed. If you liked it, stay on the lookout for more. I'll update this thread as new articles are done.

[valrus]

Thanks, this was interesting.  I don't know this genre well so it was all new to me  

The topic is one I've been musing about, as well.  Specifically, I've been chewing on the principle "For each source of random outcomes, where the outcome possibilities aren't balanced, there should be an associated choice that could mitigate the outcome."  (Put another way, when the player can fail due to the luck of the draw, there should also be a choice that wouldn't have made that a failure.)  This led to the question "What kinds of mitigation mechanisms exist?"  Your article helped me generalize some of my ideas:

• Mechanisms that allow additional "rolls" to increase the chances of success (Inspiration) or let the player "reroll" (mulligans).
• Letting the player simply decide the outcome (Rampant Growth).
• Alternatives to the randomness, like "fixed minimum" options (Dark Ritual)

Another possible mechanism, that MtG doesn't use but another game could have:

• simply stacking the deck, so that players can't get into fail states at the outset, or get overall card distributions that are unwinnable.  (Pandemic, for example, uses a "seed the deck" procedure where the four really bad cards are shuffled into *quarters* of the deck, which is then recombined; it makes getting two-in-a-row very unlikely and three-in-a-row impossible.)  This is easier to do in video games, of course, where you can do this behind-the-scenes.  (I've considered doing this in my game, making sure the initial state is such that a new player won't immediately become stuck.)

[jgrams]

Have you seen the article Volatility in Game Design by James Ernest of Cheapass Games? He argues that "biased" randomness, like you normally see, is to some extent a bad thing, and that mitigation attempts can't "fix" the problems entirely, and that it is better, when possible, to have "fair" randomness which gives different options of roughly equal value, but which require different strategy or whatever.

And he points to a couple of interesting examples, like Bobby Fischer's "Chess 960", which randomizes the starting position of chess to attempt to make memorization infeasible.

It's not terrifically long, and well worth the read. If nothing else, it's an interesting way of thinking about the problem.

[Graham-]

I think it is straight-up harder to make something "really" random. Starcraft for example struggles with issues like these. You learn to play that game, then learn the "correct" way to handle a scenario, and the meaningful play becomes rote memorization, then a battle of execution: fun to be sure but less fun than "talent-only" games (which of course don't exist).

edit:

So I've given the OP's article an actual read now. Your writing is clear and to the point. Randomness is good. The observation that the MTG mulligan is a band-aid is strong. I agree with it completely. Designing MTG in such a way so that land droughts don't affect you, or that over-landing (whatever the term is), doesn't affect you would be worthwhile.

An obvious solution would be to pull the lands out of the core deck. Then specialty cards can be played to influence how you pull from your "land-only" deck. So the control of land distribution is still there but is made more interesting. 2 decks is better than one! Add a layer of abstraction!

There would be some rules that control how quickly land makes it out onto the board. Now the designers are in a nice position: they can tune deck balancing as they did before but with more freedom. You could go even further and have 3 decks - uh oh. Too abnormal? Life is shifting?

A clever observer will note that any balancing can be done with 2 decks as well as it can be done with 1. The balancing just has to be done differently. Not only does the mulligan only partially solve a randomness issue, the problem it hides also blinds players from understanding the relationship between their build and how a game went, because too much mental statistical math is involved. This is why band-aids are bad!

Good article. I'm impressed.

[OnslaughtMike]

Thanks for the article it was a great read and something I will keep in mind since I have pieces of randomness that i need to use for my game.

[s0]

i liked the article and i think you're correct that randomness (at least the type of randomness you're talking about) is basically an "accessibility" feature.

anyway, you might wanna consider adding a left margin to your site layout. having the text stuck to the edge of the screen kinda makes it annoying to read imo.

[Snail_Man]

Nice article. It was an interesting read.

Personally, I've never played MtG, so some of the terminology was a bit hard to get my head around, but you explained it pretty well.

Anyway, good work. I'm looking forward to your next one!

[battlerager]

Personally, I've never played MtG, so some of the terminology was a bit hard to get my head around, but you explained it pretty well.

I'm glad you could follow it somewhat; there's certainly room for improvement, even with my explanation. I'll try to do a better job from now on!

anyway, you might wanna consider adding a left margin to your site layout. having the text stuck to the edge of the screen kinda makes it annoying to read imo.

Thanks, I'll  try that.

Volatility in Game Design by James Ernest of Cheapass Games

Thanks for the article, I've read it now (after writing my article). It covers many different, interesting topics. I think he offers a somewhat naive view on the "fair" worth of cards in TCGs - there are some rare cards that are not really "niche" and just better / mandatory.

The "Grit" concept he mentions is something I've seen in the "Robinson Crusoe" board game, too - fail a task, you get 2 'determination' points or something to use your character's special abilities with. It's pretty interesting but hard to get right - the player will usually need / expect a certain result and be in a worse position with other results, even given a consolation price like that.

He seems to have a hate for catchup mechanics. I agree with him to the extent of  "enforced" catch up mechanics (think bonus speed for racers in the back in most racing games, better items for racers in the back in mario kart). There's game systems where an advantage for one player makes him vulnerable though, basically "soft" comeback mechanics. (In Dota and the like, pushing a wave and destroying a tower means overextending far into enemy territory, for example). Comeback / rubberband mechanics is totally a topic I wanna cover in the future.

Thanks, this was interesting.  I don't know this genre well so it was all new to me  

You're welcome! Awesome that you got something out of my article. Seems like it helped you with some design issues. Great!

So I've given the OP's article an actual read now. Your writing is clear and to the point. Randomness is good. The observation that the MTG mulligan is a band-aid is strong. I agree with it completely. Designing MTG in such a way so that land droughts don't affect you, or that over-landing (whatever the term is), doesn't affect you would be worthwhile.

An obvious solution would be to pull the lands out of the core deck. Then specialty cards can be played to influence how you pull from your "land-only" deck. So the control of land distribution is still there but is made more interesting. 2 decks is better than one! Add a layer of abstraction!

There would be some rules that control how quickly land makes it out onto the board. Now the designers are in a nice position: they can tune deck balancing as they did before but with more freedom. You could go even further and have 3 decks - uh oh. Too abnormal? Life is shifting?

A clever observer will note that any balancing can be done with 2 decks as well as it can be done with 1. The balancing just has to be done differently. Not only does the mulligan only partially solve a randomness issue, the problem it hides also blinds players from understanding the relationship between their build and how a game went, because too much mental statistical math is involved. This is why band-aids are bad!

Good article. I'm impressed.

Thanks for the kind words! I'll cover other card games and their approaches to the "land problem" with my next article that should be out this weekend. The idea of splitting land and other cards into 2 piles is something that other people have brought up before (like everything in this world, lol) and I agree it is an interesting approach. The problem with MtG is how huge and steeped in tradition it is. Changing something this core to the rules now is near impossible, so the problem will probably exist forever. Good thing we can just design fresh (collectible/trading) card games that are not plagued by that problem (and run into different challenges, ahah!).


Thanks for reading and being so positive, everyone. I'm gonna finish up part 2 this weekend. Preview topics: Digital Trading Card Games (Hearthstone and Scrolls)

[Graham-]

Magic is about money, like anything else. Prove that a system is substantially better than another one, and that players will prefer it, greatly, then the change will be made.

The proof is hard to find of course.