I just want to note that I used to be in the anti-dice crowd (which seems distinct from the “anti-luck” crowd) and I think it might have been Twilight Struggle that won me over, helping me to realise that luck from cards, hidden information, simultaneous action selection, etc. isn’t better, it’s just different.
I still do generally gravitate towards games that don’t have dice-rolled outcomes, but these days I’ll tend to prefer a dice-heavy game over a low interaction game.
I think we are ready to spill forward.
First, we haven’t explicitly talked about this but I think we’ve touched on it and discussion won’t expand on it:
Luck provides two things: unpredictable outcomes, which excite us as humans and level the playing field, and unexpected situations, which provide novelty and allow us to test our mental reflexes in handling novel challenges or adapting to problems.
Human interaction or unmasterable difficulty can provide these effects similarly to randomness, perhaps with a slightly different flavor.
Can we explore the different flavors and kinds of luck, including hidden information, RNG, opponent action and inaction? How might these be categorized and/or scaled?
(The last part will be what you like and where you’ve seen all this well or poorly implemented. What’s your flavor.)
That was a bit vague and verbose. Let me rephrase:
If someone were to ask you to create a rubric explaining the different flavors and functions of luck (which could include non random effects that share the same function), what might you propose?
At a minimum
First the “why is luck bad” question.
I sometimes get a feeling that these are the same people who prefer strategic planning over tactical reactions. Longterm plans work better when you have less luck involved–except of course that the entropy caused by other players sometimes acts in the same way as luck namely by destroying your plans.
Games with a high amount of luck can be swingy and give players a feeling that they lack agency. It doesn’t matter what you do, the luck will decide the outcome. Games with a lot of dice rolling can be the prime example given for these cases.
Except good games will include ways to mitigate and play around the luck-based sub-systems just like you need to mitigate what the other players do.
And because people realized at some point that swingy games can be frustrating, the next thing we know is games that attempt to reduce luck (I am not talking about abstract 2 player games with no hidden information like the Gipf Project) sometimes so much that through the reduction even player entropy aka interaction is removed.
I hope we’re past all that in an age where game design walks the path of compromise and thoughtful use of random elements to give us the best of both worlds.
I’ll try to get to the next question later.
I always feel like games with a LOT of dice rolling follow probability distributions and are pretty dependable. Risk. Catan (I know, sometimes Catan goes wonky and the order of rolls matters as well, but it evens out).
Where even I get a little uncomfortable are games with a little dice rolling. I do like Polis but when you roll the die 10 times in a game, and the outcomes can destroy your chances, man that’s a whack. The 2e rules fix this a bit. Learning how to play (don’t roll the dice early!) helped.
Quantum was good but also had this, where a few dice rolls during the game determined major outcomes. While I like Quantum, I’ll admit I’ve put time into thinking about a better die mechanic.
Cyclades gets around it, the 0-1-1-2-2-3 die makes rolls less volatile and the “lose just one piece” consequence brings it back into the fun zone, lending excitement rather than frustration.
Maybe that’s it, the more volatile the probability, the less severe the consequence can be. Not inconsequential, but bearable.
I am not anti dice, don’t get me wrong. They aren’t my favorite randomizer but they are pretty decent especially if well implemented with a good mathmatical model (say hi to the good Doctor from me). And lots of them… I agree tend to do better on the probability curves.
I just wanted to say that dice are the most well known and obvious entropy machine in games and dice are so easy to blame.
Luck flavors:
I guess that is not the aim of the question. I’ll try again and find which ones come to mind or have previously appeared in this thread. Luck can be generated by:
Dice and card draw are just two classic variants: Dice roll independent of previous “draws” while card draw is not leading to completely different models. Card draw seems less chaotic usually. Although increasing numbers aka rolling more dice and increasing the stack of cards (aka the TM effect) moves the perception of how much entropy they generate closer together.
Opponents might be the least chaotic and least “fair” because some people can just play/manipulate and read their opponents better giving them an advantage that the other variants do not have.
Just some first thoughts (workbreaks …)
My hackles rise when I see a game say “roll a d6, get that many whatevers”. Too much variation, especially if you’re only doing it a few times in the game.
But “roll a d6, 5+ to succeed”? Fine.