Friday, March 7, 2008

1-2-(3)-Infinity

Nowadays I'm big on the 7+/-2 rule of game design construction. (Again, a psychological rule-of-thumb for how many things people can store in short term memory at once -- and therefore, I think, the maximum number of options anyone should have at one step in a character development process.)

http://en.wikipedia.org/wiki/Seven_plus_or_minus_two

Now, there's something else that just burbled up at me: I'll call it a rule of "Max 3 repetitions" until I come up with something snappier.

Here's the basic idea, again arising from some psychological experiments. It seems as if things repeating 3 times takes up the same mental space as things repeating infinitely. For example, there's a polling experiment referred to as a "Beauty-Contest Experiment" where people pick a number, and the winner is the one who guesses two-thirds the average. People who don't think about it at all pick a random number (average 50). People who realize that's what happens randomly pick two-thirds of that (33). People who can frame that line of reasoning as common, pick two-thirds of that (22).

And then the next most common selection group is to pick zero (0). That is, if people take a third step in this process, they assume the reasoning continues infinitely, reducing their guess to the limit of zero.

http://ideas.repec.org/p/upf/upfgen/438.html

Here's some other places this pops up. Quite a number of more primitive languages have words for "one, two, many". That is, anything beyond 2 they just lump all together with a word that means "many" (3 and infinity holding the same mental space for them). Also, you many notice that all professional jokes are written in a pattern of 1-2-(pause)-3. I think that's because after 2 repetition beats, our minds fall into the pattern, assume it goes on regularly and indefinitely, and are then kicked out of the repetition with the pattern-breaking punch line (on the 3). End result: laughter.

Okay, so what's got to do with game play? Well, don't do the same thing more than 3 times because then we're all bored with it. I might recommend not having the same monster or kind of encounter happen more than 3 times (I picked up on this in an old Gygax module today -- there's one single place in the adventure where the same basic encounter might happen two dozen times in sequence. Whoops.) Also, don't let someone use a "special ability" more than 3 times in any day because then it's inherently not "special" or "magical" anymore. So with that I'm thinking that 3E spontaneous healing, or 4E healing surges (6/day by everyone) are superfluous. More than 3 times per day is for mundane stuff (sword blows, arrow shots), no more than 3 times for day for anything you want to feel magical (spells & supernatural powers).

One thing you'll find in any piece of literature (book, comic book, etc.) is that practically no one uses the same power or move more than once in any fight. Certainly not more than 3 times, then it gets boring for everyone. That basically happened automatically in OD&D and 1E with the very limited number of spell slots that wizards & clerics had. Being able to duplicate spell slots (e.g., fill all spell slots of one level with the same spell) would be an exploit that would really break the game experience, I think (and fortunately I never saw it in play). Spontaneous casting in the d20 System falls down the same hole pretty badly.

4 comments:

  1. Reminds me of something a design teacher taught me...
    "Incidence, Coincidence, Precidence"
    In other words, if something happens once or even twice, it can be attributed to random chance, but if it happens three times a pattern is beginning to emerge. I imagine there's something in our brain architecture that flags a third occurrence for filing in the "get used to it, it's gonna keep happening" folder.

    ReplyDelete
  2. There's a another huge area where 3E massively fails the 5-9 yardstick: number of modifiers to common rolls, which I think is absolutely critical to good game design.

    Consider attack roll vs. armor class, which is after all what the DM must deal with often and quickly. Generally for older D&D, you have a (1) stat bonus, often (2) a magic weapon bonus. Then there is (3) the THAC0/BAB conversion. On the other side for monster's there's just (4) AC. For monters attacking PCs it's much the same (single number vs. stat + magic + equipment AC).

    That leaves enough room to throw in another bonus or three on the fly or for situational things, racial weapon bonuses, proficiencies, bless, etc. Still, even these numbers are pretty static even in AD&D and 2E. You can write a big 5 on that shield on your character sheet and you're good to go. Who knows the DM might have enough brain power left to actually add some flavor instead of just crunching numbers.

    But consider 3E. You have (1) your BAB, (2) your stat bonus, though it is very often enhanced with (3) a magic item or buff (which duration needs to be tracked carefully per-PC!), plus (4) magic/masterwork weapon, plus (5) extremely common situation bonuses like flanking and charging, vs. (6) monster's base AC, but often having to worry about (7) natural AC (8) dodge/dex AC bonuses separately because of surprise, touch attacks, etc. We've failed already but even the average low-medium level party will have more modifiers than this floating around every combat.. someone's got two-weapon fighting, or point blank shot, or weapon focus, is doing flurry of blows, etc.

    The headspace required for a single attack roll is so big that the DM is tempted to no longer even think about the PCs' scores and just focus on the monsters, especially as levels go up and things start getting crazy (God forbid if the monsters have PC levels with buffs and magic items or suffer stat damage...). The player rolls D20, announces "I hit AC 23", and the DM will either have to trust it or do a spot check of the player's math, which often enough will be wrong, or combat will end and someone will remember "oh wait we forgot to apply the -4 for shooting into melee!"

    Modifier insanity is why I stopped playing 3E. I appreciate what they tried to do breaking down AC for example to make touch attacks work (though for Bull's Strength, I want to wring their necks). But it seems far too often what the did was say "all right, we've fixed this one edge case of AD&D rules, now let's try to make it come up all the time by adding feats and items and spells that use it".

    ReplyDelete
  3. Wow guys, great posts.

    Cheepicus, I love the observation that 3E got so complicated that DMs couldn't audit player rolls on the fly, I hadn't put it that way before -- and what that wrecks is the ability for a new player to get into the game with no mechanics knowledge ("I shoot my bow, what happens?").

    And I totally agree with the observation that advancing D&D tends to fix one thing and then complicate 6 other things at the same time.

    ReplyDelete
  4. Mathematicians also usually count 1-2-many. Sometimes zero is a special case that needs cosideration, too.

    ReplyDelete