News Shop
Events Chat

[Game Design] Interesting Decisions

One of the things I most loved about the old Sirlin forums was that we could often get into quite good conversations about games and game design. It’s been a good place for me to test my thinking and learn over the years. Since this is a new forum and a fresh start I thought it might be fun to start right over at the very beginning with just about the most basic question you can ask that’s still useful.

Games are about decisions, in other forms of media the consumer’s only agency is the decision to stop reading/watching/listening/whatever. But games are defined by interactivity, they’re defined by asking the consumer to make choices. Since decisions are right at the heart of the experience they should be interesting decisions. So here’s my question to you:

What makes an interesting decision?


[quote=“Fenrir, post:1, topic:399”]
Games are about decisions
[/quote]Some games are about decisions! Some games, however, are more about challenging execution tests. Some of those aren’t even physical execution tests. Let’s not accidentally pretend that only a subset of things generally understood to be games are “real” games. That way lies madness.

[quote=“Fenrir, post:1, topic:399”]
What makes an interesting decision?
[/quote]In general, the decisions I find interesting are the ones were I can’t trivially compute a “best” choice from the information currently available to me. This means that decisions can be interesting for multiple reasons - difficulty of computation, hidden information, or comparably valuable options, for instance.


What makes an interesting decision? For me, ambiguity. Uncertainty that it’s the “best” decision given the information that I have available to me.

But that has to be coupled with the ability to look back, after the game, and judge my decisions with a little more clarity. Basically, I need to be able to learn from my mistakes.


An interesting decision usually has multiple effects and costs. It also needs to make assumptions about hidden information.

What games aren’t about decisions? I’d posit that, even games that are “more about” execution tests are also about decisions. Fenrir never did say the two are mutually exclusive.


I would say that an interesting decision is one where there’s reasons to choose either way. And that largely depends on the context of the decision.

If I ask you “do you want chocolate or vanilla ice cream?”, and you absolutely loathe chocolate ice cream, that’s not an interesting decision. If you are fanatically pleased by vanilla ice cream, it’s not an interesting decision. But if you like both, even if you like one more than the other, it’s now an interesting decision, because you have to weigh both options and make a determination on which option is best in the current situation.

Interesting additional point: if two options are found to be equally correct choices, then it’s just as uninteresting of a decision as when one option is obviously correct. It’s the valuation that makes a choice interesting, and the challenge of figuring out how to determine which choice is correct in an unpredictable situation.


An interesting decision is one where you have more than one viable option for a given instance and could justify each option regardless of the outcome. This often includes ambiguity and uncertainty as people had mentioned before. In Yomi, I think KD can create some very interesting decisions. It’s sometimes very hard to valuate KD and you could justify either way whether converting damage or going for KD and attempting a mixup is the best decision.
In many games, balancing immediate benefit and opportunity cost is interesting, hard to valuate, and could be argued either way.


Execution is an act of doing. Choosing how to do something is a decision :kappa:

On the other hand, I would say it is a lot easier to say what would be an uninteresting decision. As Socrates say, it is a lot harder to know what is right, than to know what is wrong. Or the other way around.

Decisions are at their most basic a choice between 2 different things. But having decisions boil down to a 50/50 is not necessarily very interesting. Then again, having too many choices can equally feel just as meaningless.

OK, so one property on which there’s a clear consensus is ‘Ambiguity’.

In order for a decision to be interesting it must be ambiguous. I.e. the player must not be able to determine the definitive “correct” choice in any situation of play. How you create ambiguous decision is a question on which I have many thoughts for another day. But for now, ok, pretty much everyone agrees that interesting decisions have to be ambiguous.

What about this as a criterion? “Predictable.” Not in the sense that you know exactly what’s going to happen as a consequence of your decision. But in the sense that you need to have enough relevant information to make an informed decision.

All I’m really trying to say here is that a ‘decision’ is not a ‘guess’.


An example of decisions gone wrong:

Where your decisions don’t matter: This is where you could do any of all the decisions available to you and none of them has any influence on what the outcome will be. This ranges from actually acceptable if done in a strickly scripted one-time event where the game is (more often than not) narratively trying to give you the illusion that you are able to escape only to find that you are being caught in the end anyway. Or to the stricktly wrong approach where you know exactly what your opponent is going to do, but even with full knowledge of what is going to happen, no counter exists or no option will be able to beat whatever your opponent is going to do. This can be applied to some end game states in Yomi, most prominently displayed with Gloria. (Tho, the knowledge of them makes all the previous decisions within a game against her all the more meaningful)

About predictability, I feel that the difference between a guess and a decision is predictability. What you can’t decide, you have to guess. Decisions feels … or should feel, meaningful. Guesses do not.

Ok, what about this hypothetical situation.

You’re playing [insert name of moba here] and you have killed your opponent in lane. You move forward to put pressure on his tower and your ward expires behind you. A moment later you notice the ward has expired and move to retreat. As that happens their jungler comes out of the brush (or whatever) and attacks you. You’re far from your tower and their jungler happens to be higher level than you also is a really bad match up. No combination of button presses will save you from getting murdered in this gank. You’re dead.

Did you make any decisions?

Of course, you overextended and played unsafe. You took a risk and got punished for it. Had the higher level (or by extension the better player) that came out of the jungle not been able to punish the overextending player, that would be an example of having decisions be meaningless since the overextending player apparently can do whatever she wants without it having consequences.

Did the player who over extended make a decision though? Or did they make a guess? Like, the position of the jungler is an unknown variable… so really what they did was make a guess that the jungler was far away?

By that definition, Yomi is a lot of guesses…

I would say that they made the decision to move forward based in part on their guess that the jungler wasn’t nearby. The two aren’t mutually exclusive, and I’m sure more factors can be identified as part of the reason why they moved forward (for example, the desire to take advantage of the time frame where they know the enemy who usually guards that lane is unable to stop them).

In other words, I’m saying that a decision is often based on multiple factors, some of which might be guesses.

1 Like

A guess would be more akin to sending an area of effect spell into the jungle, hoping you would hit someone who was hiding there.

Having the knowledge of the minimap, the knowledge that there is one less character on the board, that they probably know where you are since you just ganked her, then you are making the decision whether or not you feel it is the correct call to continue forward without backup.

A guess is different from an informed decision. I would say there is some semantics going on here, where does the line go? I would suggest that the less you know about the different possible outcome, the more of a guess it is. The more surprised you are by an outcome, the more of a guess it is.

1 Like

I don’t agree with the more surprised you are by an outcome the more of a guess it is. I can play, as an example, Fire Emblem, where the enemy has a 1% Hit and 1% Critical rate, and only if a critical occurs will my character die. If that critical lands I’ll be very surprised, but I wouldn’t call that a guess in any way.

I do agree that the difference between an informed decision and a guess is based off of how informed you are, and it’s a sliding scale, but if anything I would say the more surprised by an outcome you are the less of a guess it may be. At least when I guess I tend to consider all the possible options roughly likely, so I wind up less surprised. If i make a decision about all sorts of variables and consider the least likely option incredibly unlikely I’ll be surprised when it pops up.