Categorizing Choices

The other day Tesh caused all sorts of problems by asking questions. When I am dictator I will not allow people like him. But forget I said that.

Or else.

"meaningful risks and decisions"

Two different things... and how are we defining "meaningful" anyway? Is it in the deciding, the challenge, the potential loss or the potential reward? Maybe some calculus involving all of the above? How squishy is that calculus? Could we drop risk entirely and still have meaning for decisions?

So anyway, the questions were about choice. That's one of those fun words, and by fun I mean overused and often meaningless, not because it has no meaning, but because it has so much potential meaning that we don't know which one, and then thing we're arguing past each other because we're using the same word to say different things and don't even realize it.

Choice.

What do players mean when they say they want choices? Who knows? I mean, maybe they do, but I don't, and you don't, and I'm convinced that no dev has the slightest clue. Problem is, choice means a lot of things and thus far no one has done a player-by-player survey of what exactly they mean when they say they want choices.

I want to fix that. I will likely fail due to a lack of publicity, which I may rectify by tweeting photos of my Anthony. But I will try!

Slicing and Dicing Choices

Time
How long do you have to make this choice? If you have all the time in the world, then the choice may seem trivial, even pointless, since you can take the time to look it up. For example: gemming. But slow choices need not be trivially over-optimized by the netosphere. Think of a turn-based game, where yes, there may be best options, optimal options, most efficient, but you don't quite know what they are and it's not because you're a lazy idiot. Which forest should you chop down first in that game of civilization? Who should you invade first? These are slow choices, but they are not trivial (maybe the forest has very very little impact)

Importance: Personal or Objective?
Is the choice aesthetic or does it impact the game? Do I want a blue or red cloak? Do I want a strength or an agility gem? Note that these can be hazy, such as when people have different objectives, so objective and personal can overlap. Flavor differences may also come into play, such as different specs of a DPS class; you may like how one plays, but maybe it isn't the highest damage.

Magnitude of Consequences
What are the consequences? If you pick wrong do you lose the game? Or are you mildly inconvenienced? Compare "soloing Garrosh" with "stacking spirit as a hunter." The good and bad consequences do not need to be equal in magnitude. In the gem example, switching to agility would be better than spirit, but per-gem it's nearly trivial, and there is no cost beside the opportunity cost of the correct gem in the slot. Not soloing Garrosh won't give you anything, but it will save a death. Or for non-obviously stupid choices, try raiding or questing: each one denies you the other, but neither is stupid (I'm assuming the person soloing Garrosh is not doing it for the fun of silly deaths, in which case not soloing him would be the stupid choice).

Keep this one in mind, since the magnitude is often what turns people away, either due to lack or excess. Death in WoW is trivial. Less so in EVE. Guess which one has more subscribers. Guess which one has more people whining about how everything is too easy.

Obviousness
Is it a choice if there is one right answer with significant negative or positive consequences and we know the answer ahead of time? As an extreme example, imagine if a game was found to be too easy, due to having a Win button, which caused anything between winning the current NPC battle to winning the entire game. The devs decide to make it harder, and add choice, by adding a Lose button. Sure, you now have the choice to press the Lose button, but are you? Maybe once, just to see what happens. But then it's right back to the Win button. Or quitting, because that would be a terrible game.

Let's try some choices
To simplify things, let us assume that our goal is victory. If our goal is instead to have a peaceful empire in a Starcraft melee match, that will require a much different approach.

Where should I attack the enemy base? First off, we have some information. From scouting we have some idea of their buildings. We know what units we have (unless you're like me and someone lose dropships). We can remove some of those obvious bad choices, such as rushing a tank-bunker combo with marines and no medivacs. But what is the good choice? Maybe we don't need to worry about those tanks at all and could instead focus on harassing workers. Or we know they're just turtling forever, so we can ignore them for now and build up units and technology. Or there don't have many detectors and we're pretty sure they don't use sensor sweeps much, so a cloaked banshee attack could wipe them out without them even firing back.

This isn't obvious. We don't know all of what the enemy has, or how he will respond, or whether we can both attack and deal with his secret fleet of battlecruisers, against which our banshees will be useless.

This also isn't fast. But it isn't slow either. We have time, but not enough to go read an essay on counter-terran-turtle tactics. We can think, but not too long.

The magnitude can be significant, or not. We could attack with a few units as a sort of test run, but if we don't bring enough we increase the chance of failure, even while keeping down the magnitude. We could rush with everything, but if that fails, we're in pretty big trouble. What is the gain, anyway? Maybe they're defending nothing at all. Or a gold mineral field. You could even end up doing him a favor, psychologically, by freeing him from feeling he needs to hold that ground. Losing a few tanks and uselessly trapped marines is a small price to pay for freedom to act.

Since we set the goal as victory, and since Starcraft doesn't have many aesthetic choices anyway, we can say this is an objective choice. But what is the actual choice to pick? We don't even know yet if we are even going to attack, let alone how, where, when. These all have some sort of measurable (after the fact) benefit or loss, but how we think still matters. And how they think.

Note that while I used a PvP example, this still applies in PvE. Possibly even the specific scenario, partially thanks to the very vague setting.

Moving on to WoW
Tobold has a strange habit of complaining about raids not having many choices. He has a point, in some ways. Raids often have a correct, intended strategy, and going against that can result in the fight being changed specifically to keep people in that box. This means that the overall choice of the raid is one with significant difference in magnitude of consequence: wiping vs. epics. The choice of strategy itself may not be obvious, but thanks to the ability to reload the fight exactly as it was, and also that internet thing, it can become obvious. Then there is time: we have a lot of time to figure out strats, since they never change.

I want to note that this so far sounds like a really bad game: high magnitude difference, high obviousness, and lots of time. But this isn't what any individual players actually do. Instead they are making much quicker choices, which may still have a high magnitude difference, death/wipe vs. epics, but which will not always be obvious. Of course afterward we can see that that player was obviously retarded, but at the time, they had very little time to make a high magnitude choice, which given the limited information, maybe have had a total lack of obviousness. Low time, high magnitude, low obviousness: that's more or less what salespeople try to push you into, "Buy now! If you don't terrible things might happen! Don't think about it or research it, buy now, before the sale ends!" I think any moral, sane person can agree that sales is an evil thing. Alternatively, this player could be better informed and aware, changing it into a fast, high, obvious choice, which is often takes the form of a twitch test or dance routine.

See how the individual aspects of choices, fast or slow, obvious or not, high or low consequence, are not objectively good or bad. But certain mixes of them can be profoundly unfun, subjectively.

If you're making a turn-based game, your players probably aren't looking for fast choices, so adding them is likely to be unpopular. But if you're making a fast game, asking them to ponder philosophy may not be popular either. These aren't hard rules either, so a game may benefit from a mix of both. Such as Starcraft, where micromanagement may get a lot of attention (check the APM, it's off the charts, I click so fast, you can't beat the rush) #notintendedtobearhymingstatement but the best tactics won't make up for a bad overall strategy, which is a rather slow decision.

Choice vs. Decision
After writing this post I realized that recent events in life might be relevant, and add another level of confusion. I have a bachelor's degree in psychology, which started as an attempt at whatever one gets in chemical engineering. This fall I am heading north to get a master's degree in public affairs (no guest speakers announced yet). One last bit of information: I am the least educated person in my family. My father has a PhD, mother has a master's, brother has a master's (working on PhD now), other brother is a lawyer. Under these circumstances I was at the absolute very least going to college.

College was the path of progression. It was going to happen. I did have choices: school and major. The choices had high magnitude consequences, low obviousness, and were very slow (and yet somehow, still too fast to get it right).

In contrast, going for my master's degree was a decision. This may be a distinction in my mind only, but many things are and that has never stopped me from sharing them. There was no path that said I should get it. It was not the next step or part of any guide. It was something entirely novel, initiated by me.

Can games include decisions, in addition to choices? I think so. After all, while there was no breadcrumb trail to lead me where I am headed, it is not as if I had to invent everything from nothing. Arguably what is going on here is still just a choice, just not a choice that is obvious or expected. From that perspective a game could replicate this: just give so many choices that we cannot possibly evaluate all of them at once, and without obvious best paths, so that we instead must decide on an entire path, rather than just taking one or another road on the same route.