Design Thoughts: Managing Expectations
Managing Expectations
Part 1 - Intro and the "Marketing Trapdoor"
A friend of mine, a Dungeon Master, was chatting with me recently about the fact that two of his player groups with no overlap ended up making the identical decision to start a kobold uprising, which more or less upended the module he was running. I'm a player in one of these groups who made this decision, and I thought about why we thought of inciting a kobold rebellion. After asking some questions, I determined that this situation probably occurred independently for both groups because both groups encountered a kobold who changed sides earlier in the campaign, an expectation was set for both groups that kobolds can defect to the players' side. So when faced with a large number of kobold forces as part of the enemy army, both groups separately arrived at the conclusion that they could potentially incite a kobold revolution.
This got me thinking about how games/game designers set expectations for players and deal with player expectations that come from outside the game, and how that affects player experiences.
I remember hearing a story that I can't confirm is true I found a GDC talk confirming a story I heard about how Rocket League's physics were adjusted to be a lot more predictable. This was a decision that ultimately came down to player expectations. In a game like Rocket League, if your player speeds their car into the ball from a direction and the ball reacts by flying in a direction, the next time the player speeds their car into the ball from roughly the same direction, the player expects the ball to fly in the same direction, whether consciously or subconsciously. Rocket League using a simplified rectangular block collider is a great way to make sure that player expectations are consistently met. Human minds have a tendency to simplify the world they perceive. Similar to how first impressions are important for people, they're important for a game, too: the first interaction sets an expectation for future interactions, and if those future expectations fail to meet the set expectation, the player is left confused or with a feeling of "wrongness." A sort of vague feedback a player might provide like "this feels wrong" or "this feels off" usually means some kind of expectation is not being met properly. This can work for or against a given design. And expectations can, fortunately or unfortunately, be set outside the bounds of the games itself. On the other hand, if a game exceeds, subverts, or plays with player expectations in the right way, it can delight players.
A recent example from a more business-y perspective: the high-profile "failure" of Cyberpunk 2077. Cyberpunk 2077 is a tragedy of mismatched expectations. CD Projekt Red marketing drummed up enormous amounts of hype, which inflated player expectations to a degree that the game was more or less doomed to disappoint. Especially considering the choice to launch on previous generation consoles whose hardware simply could not support the vision behind the game, and to restrict the availability of console footage of the game pre-launch. Having played the game myself on a mid-range gaming PC setup, I think it's just fine. From what I understand, the console version launched in a state that was "borderline unplayable," while high end PCs got really close to the visual experience originally advertised, though not necessarily the game experience promised. "Just fine" and "borderline unplayable" fail so hard to meet the Everest-high expectations drummed up by marketing that the launch of the game is seen by many as a complete failure, even if the game didn't necessarily fail for some players, and even if it wasn't ultimately a financial failure for CD Projekt Red.
I would say the type of failure that Cyberpunk 2077 experienced is a pretty common one for bigger game releases.Anthem experienced something similar a year or two before. It has to do with expectations set by business and/or marketing and/or vertical slices and what basically amount to "campaign promises." I would wager that smaller games run into this less often because it's harder for smaller studios to invest as much in marketing and in polish for a vertical slice that's too far removed from the actual game.
If I were to draw a curve for player expectations over time for a game like Cyberpunk 2077, it would have a few steep-ish jumps with slow builds in-between from the initial announcement, to the next showcase where they showed the game, to the next news drop, maybe some slow falling after a longer period of no information, and then in the year or so leading up to release, some exponential building in expectations with a steady information stream and news outlets getting to play vertical slices and releasing impressions. Delays do a weird thing where even though they upset some impatient players, they still end up building expectations upward overall, unless they are accompanied by information famines, in which case expectations fall as players begin to suspect vaporware conditions. For CP2077, I would say the delays increased expectations. As far as I can tell, players are becoming more aware of reasons behind delays and more understanding of delays themselves, leading them to expect a more finished and polished product after a delay or two, which means the expectation curve moves up. Then, upon release, you would see the expectation curve sharply plummet as player experiences fail to meet their expectations. I'll call this the "Marketing Trapdoor" as the curve takes a sudden shift almost straight down.
I think it's okay to build expectations and an upward trending expectation curve is often a good thing, but building expectations is a bit like building debt: eventually it needs to be paid off or one must suffer the consequences.
Part 2 - Genre Expectations
Expectations can also be set by genre. Expectations can be especially strongly set by games considered "genre-defining." One of the games I worked on, Vanguards, ran into this early on in development. Vanguards is, at its base, a MOBA, which is a genre largely defined by DOTA, League of Legends, and a couple of other prominent examples. Early on in the development of Vanguards, we experimented with a control scheme that was more like a twin-stick shooter (or whatever is the PC equivalent thereof), with the player pawn movement controlled by WASD and skills actively aimed and fired by the mouse and mouse buttons. What I didn't understand then was that calling a game a MOBA sets a number of expectations that extend throughout a game's mechanics, including input methods. We got feedback fairly quickly from some industry professionals who came in to play the game that subverting the input expectations the way we were doing would mostly serve to push away people who enjoyed playing other games in the genre. Essentially, it would become a hill for our game to live or die on, and the question was whether we really wanted to die on that hill. We decided that we didn't.
Marvel's Spider-Man for PS4 does a really good job with managing genre expectations. A staple or standard for open-world games in the past decade or so is the presence of activity icons spread out across the map, showing players where they can do various things. This serves as a checklist, which motivates players to complete things, but I've found as a player that they can often be overwhelming. Something many games do to mitigate this is having "watchtower" activities that reveal sections of the map upon completion, reducing how many icons a player can be overwhelmed with at once. This sometimes serves as an antipattern for players like me, who like to travel to all the "watchtower" locations to reveal the whole darn map before making much progress in the game at all. Something that Spider-Man PS4 does in addition to watchtowers is having activity types that are revealed only after progressing to certain points in the story. This is, in my opinion, a great way to deal with a number of different problems that open-world games have, one of which is the aforementioned overwhelming icons issue. This also serves to help adjust the incongruity between "the world is ending in a few days" and "let's go do this sidequest," which is another genre problem open world games often have. Spider-Man PS4 does a slew of other things right by me in the game design space, so expect to see it again in future posts.
Sometimes genre expectations can come from just a single other game in the genre. For example, I recommend people who have played neither The Legend of Zelda: Breath of the Wild nor Horizon: Zero Dawn play Horizon: Zero Dawn first. And the reason I do so can be boiled down to a single mechanic present in Breath of the Wild: climbing. In Breath of the Wild, you can climb just about everything. Playing it first sets an expectation for exploration, and playing H:ZD too soon afterward will create an expectation gap. Not being able to climb everyting, and not having the world designed in such away that climbing and exploring every nook and cranny will yield little rewards will "feel wrong" for many people. Reverse the order in which one plays these two games, and the expectation curve moves upward, rather than downward, which is a better overall experience.
Part 3 - "Taking Toys Away"
Speaking of expectation curves moving downward, "Taking Toys Away" is a type of expectation management issue that I've seen fairly often. The quick summary is basically: once players have gotten attached to something and put effort into something, taking that thing away is usually a bad idea. I say "usually" because I think it's been done correctly in some cases, but those cases are rare. Sometimes, you -have- to take toys away, and if so, it's better to do it sooner rather than later.
For the former situation, players had built up years of equipment, abilities, and identity for their characters in D1, which in moving from the original game to the sequel, was for all intents and purposes erased. You could port over your character's facial appearance, and the game remembered some achievements you'd accomplished, which altered pieces of dialogue and an opening cutscene, but beyond these primarily cosmetic bits, nothing else remained of the time spent in D1 when moving to D2.
The Pokemon Company faced this issue with the recent release of Pokemon Sword and Shield. For the five previous generations of mainline Pokemon games, players had been able to transfer over just about any Pokemon they wanted to from game to game to game, moving forward with each generation. But, for Pokemon Sword and Shield, it was announced that not all Pokemon would be able to be transferred to/used in the game due to, again, the problems that arise from an indefinitely growing set of options for players. This caused an uproar in the community. The expectation had been set by the five previous generations of Pokemon games that everything would be able to move forward and be used in each new game.
Part 4 - Steepness and Conclusions
Now that I've typed through several examples, a pattern I'm seeing emerge is basically that the sharpness of the expectation curve is what is most memorable and to a player. "Taking Toys Away" and the "Marketing Cliff" are examples of sharp drops in the expectation curve. That makes these situations very visible and they often feel very negative to players and leave long-lasting impressions. On the other hand, sharp jumps upward are positive and can have similarly visible and long-lastion impressions.
No Man's Sky experienced the "Marketing Trapdoor," but was able to recover to a great degree because it was able to release a large update that created a steep positive surge on the expectation curve. Final Fantasy XIV did more or less the same thing with its failed initial release, followed by a complete overhaul in A Realm Reborn creating a steep rise on the expectation curve. The sharpness of the positive jumps in the expectation curve arguably saved these two games.
You can see the opposite in both Cyberpunk 2077 and Anthem. Both games have failed to recover from the Marketing Trapdoor. Both games struggled with incremental updates and small fixes after launch that were too little too late. Without sharp jumps in the expectation curve, they can't recover. Anthem's planned steep jump update was cancelled, and Cyberpunk 2077 appears to intend to remain in a holding pattern. It's possible that CD Projekt Red is intentionally keeping expectations low so that when they do release a big update, the expectation curve spikes enough to save the game's reputation and their own. Time will tell.
These ideas apply at a smaller scale as well. If you release, say, a new playable character in your game, your marketing of the upcoming character will likely generate an upward trend in the expectation curve. If you let that rise too high and the character releases underpowered, you create a microcosm of the Marketing Trapdoor. It's important to note that If you then power them up too slowly, you fail to memorably change player opinion of the character and the character is likely to be underused for a long while. If, however, you spike their effectiveness to make them a bit overpowered and then tune them back to "balanced" slowly, you'll achieve a memorable positive turnaround, which will positively shift player opinion. Depending on how low the expectation curve got before the turnaround, the amount the character needs to be buffed will vary. If it's low enough, bringing the character up to "balanced" might be enough of a shift to make it memorable, but the longer the character is considered undertuned (up to a certain point, more on that in another post, perhaps), the harder it will be to shift player perception.
I've highlighted a lot of negative examples here, but I want to end on a positive one. I think Legend of Zelda: Breath of the Wild did a pretty great job of setting and meeting expectations. A concern I heard people fairly often voice leading up to the release of Breath of Wild was that "sure it'll be big, but what if it's mostly empty?" That was an expectation that was set by the initial teaser and subsequent marketing communications. The first teaser trailer was a vast, wide-open, largely-empty field. It was reported before launch that the map was twice the size of the Skyrim map. What did players get when Breath of the Wild launch? wide-open fields, a map twice the size of Skyrm. Was it mostly empty? In some ways, yes. But there was a level of attention to detail in a bunch of minute interactions in gameplay mechanics and content in the fields was arranged in such a way and the setting set up in such a way that the world doesn't feel that empty, and when it does, it makes sense. There's an intentional lonely emptiness to parts of the Breath of the Wild world. So the expectation was set, even with some negative cast to it, but then Nintendo took those expectations and met them and ran with them, leaned into them, and exceeded them in meaningful ways. Botw is a fantastic game, and an incredible example of setting and meeting/exceeding expectations.
P.S.
Speaking of managing expectations, I have no idea how frequently I'll post. I'll try not to be slower than once a month, but that's super-slow by prevailing internet-content standards, so I'll try to be faster when I can be.
Comments
Post a Comment