Open Information and Game Flow: How Game Design Incentivizes Analysis Paralysis

Chess is a perfect information game. While it’s become popular to play Chess online with very limited time limits in recent years, Classical Chess – what most of us think of when we imagine Chess – is played with extremely long time controls.

The recent World Chess Championship featured time controls of 2 hours per player to start, plus an additional hour at move #40, 15 minutes at move #60, plus a 30-second increment per move for any past move #61. It’s not uncommon to hear about Chess matches lasting 6-7 hours in this format.

FUN FACT: The longest tournament Chess game in history clocked in at just over 20 hours. The longest world championship match, over 8 hours.

The reason for this is my first sentence: Chess is a perfect information game. With a strong enough calculation (beyond even computers’ abilities at the moment) you could “solve” Chess.

If someone were to introduce an abstract game to the market that achieved its peak potential at 6-7 hours of gameplay, it would be destroyed by critics. And yet, some of them probably are best this way. The social contract of gaming simply does not allow for this 99% of the time, but the “ideal” way to interact with some games’ system would likely be to sit down for a 5-10 hour session.

I’m working toward both a thesis and some examples, highlighting strengths and problems. This isn’t a “Thing Bad” or “Thing Good” post, but rather an exploration of something I don’t see discussed much.

Thesis & Extrapolation

The thesis statement, if there is one, is this: Open and “perfect” information lends itself to intense, protracted calculation, which in a social setting can be seen as analysis paralysis (AP).

One of my arguments is that, outside of tournament settings where the expectations are different, this can be a problem.

One of my other contentions is that players who push back against claims of AP and too-long sessions are making concessions for the social environment, which is fine, but that in itself isn’t a refutation of the correlation. Whether or not this is a problem is more subjective and contextual, and depends on the group or individual we’re looking at.

Oh No, I Need to Discuss Splotters

I have an unofficial rule not to debate Splotter Spellen games, due to how polarizing they tend to be. So let’s break that rule?

Games like Horseless Carriage and Food Chain Magnate (FCM) are perfect information games, or nearly enough so that I shouldn’t need to include caveats. The length – both overall and in individual turns – as well as the “analysis paralysis” that can occur, are not things only I’ve noticed. These are, almost inarguably, features of these games. Their fans love the “brain burn” of them, which is exactly this calculation.

And it’s because so much of the information (and there’s a LOT of it) is open.

If there were to be a world championship for these, I’d assume there would need to be time controls. Because otherwise, the game really could last the better part of a full day, given all the forward-looking calculations that could be run.

This is made more apparent when players are encouraged to lean into the interactive elements. In Horseless, you could (and some advanced players likely do) piece together the potential board states of opponents, in order to design yours against them. Crafting one’s board state alone can cause lengthy turns, let alone accounting for everyone else AND attempting to look ahead several rounds to parse for potential opportunities.

So like, the anecdotes we have of people playing 6-8 hour sessions of FCM, to me, are simply where players adopt the mentality of world Chess championship players. I have this time, they think to themselves, and I’m going to use it to make more informed moves. This is how they best absorb and engage with the design.

For clarity, Splotter games aren’t the only examples I could name here. But they’re the most convenient.

The Social Dilemma

If everyone’s happy doing this, there’s no problem. But of course I don’t have to condescend to you and explain how a 6-hour session of a game (billed as 2-4) might be boring for some. It’s self-evident.

So I’ve had discussions with fans of the games mentioned above, and several others, and they don’t have issues with AP in the groups they play with, and can play FCM, for example, in two hours. And that’s awesome!

I’d argue, though, that this says more about them than the design, since the staggering amount of open information, and considerations for pivoting your own play to match it, does lend itself to this type of intense and lengthy calculation.

They’re willing to play approximately rather than precisely in many instances, which is good enough to feel the strategic brain burn, but without creating onerous interpersonal issues. The experience is better, even if their strategic play isn’t as precise. But at times it means they’re ignoring aspects of the system presented to them.

Why This Isn’t Always Bad

I do actually have one sub-group of friends who are ok with long sessions. I played Sidereal Confluence with them, for example. Generally a ~3-hour experience for me, and it was between 5-6 hours.

But it wasn’t dull. And it was because everyone was more than willing to explore the entirety of the experience for all it could deliver. Table talk was off the charts. They weren’t simply trying to race to the next game. They were invested in the moment.

Maybe it was a drag for someone at the table and they simply didn’t say so. But I’d also argue that there’s something here that many gamers could learn from.

The team variant for Concordia Venus is, by most modern definitions, clunkier, and the length bloats to match it. Part of this is that there’s a lot of open information in the game, and suddenly with two of you on the same team, strategic considerations become a lot more intricate. I’ll end up having Chess-like discussions with my partner about what we do if {Team A} does {X} as opposed to {Y}, and there are a lot more of these moments than in the base game.

So in the comparatively streamlined base game, my head is down in my own world a lot more often. In the team game, I’m constantly interacting with the other players. It might be twice as long as the individual game, but it may also be better.

I do personally draw a distinction between adding length as a result of table talk vs. a result of inward, silent calculation. The latter is not as fulfilling to me. But for others, they’re functionally equivalent, so long as some type of engagement is present.

Missing the Forest for the Trees

Gamers sometimes tend to think incorrectly about this. You’ll hear a lot of takes like this: “There’s a ton of AP in this game because there are so many options on a turn.”

This can be true to an extent. But options on their own don’t necessarily lead to analysis paralysis.

Rather, it’s the amount of open information, and calculable eventualities. Adding copious numbers of action options can make these things worse, yes. But when I truly dig into the games that induce AP – and where the criticism seems to follow them around – it’s invariably due to how much is laid bare to analyze, as opposed to obfuscated behind hidden or randomized elements. The number of action options available is secondary.

“It’s a Player Problem, Not a Game Problem”

You’ll hear this in defense of games that can induce AP. And I’ve never been able to fully agree with them, but also have struggled to refute the claim.

And it’s because it depends. Which is the wishy-washiest opinion to have. But also often the correct one.

It depends because I truly believe design can incentivize AP, inducing higher levels of it by rewarding this investment of time, which can be onerous for others at the table. Yet I also believe that the social meta of a gaming situation should come first, and “reading the table” in order to play quickly enough should be prioritized over optimizing one’s strategy.

So I cringe a bit when this is dismissed “only” as a player problem. The critics are onto something in this regard; it just won’t affect all play groups equally.

The Conclusion

My point, then, isn’t “AP Bad” or “Open Information = Bad” but to truly acknowledge that open, perfect (or near-perfect) information does lend itself to calculation-driven downtime.

The more you give your players to chew on that is empirically verifiable, not behind randomness, player screens, blind draws, or unknowable player actions, the more you incentivize personal calculation.

If your design is bloating in length relative to your aspirations, for example, you might obfuscate certain elements to demand more “gut” plays rather than precisely calculated ones.

I feel like we sometimes deny this fact in order to defend our favorite games that feature these elements. To acknowledge it isn’t to acknowledge a flaw, since whether or not it is a flaw is subjective and contextual. But to deny it is to ignore what creates the dreaded analysis paralysis, or the less-dreaded brain burn that might look identical but actually be more satisfying to some.