Ambition is a difficult thing to assess in game development. Witness the above sentence: There's always a qualifier. It's difficult to find a review of an indie game, for example, that doesn't contain some variation on a sentiment like "for such a small team, this game is great." Conversely, there's an assumption that the BioWares and Rockstars of the world, with their infinite resources, should be able to crank out expansive 100-hour works of genius with zero bugs while innovating on every aspect of the medium.
Okay, maybe that's putting it a little strongly. But the fact remains that we typically judge a game's success relative to the resources that went into creating it. I don't think there's anything wrong with that approach per se - more resources do often translate to more opportunities for creativity and polish - but it can blur our understanding of what we mean by "ambition."
Ambition, to me, does not necessarily entail innovation. Just because a game doesn't do anything new with the medium doesn't mean it's not ambitious. Minecraft is an innovative project, combining the randomly-generated worlds of a roguelike with the freedom of a map editor or LEGO set. Super Meat Boy, on the other hand, is not especially innovative - it's pretty much a platformer - but it is ambitious in its approach. By ratcheting up the difficulty and tightening the precision of the controls, Team Meat established their unique vision of what the genre can be and challenged players to come along for the ride.
That almost antagonistic "let's see what you got!" attitude toward the player, I think, is what defines Super Meat Boy's ambition. It's a risk. Especially at a time when games seem to be getting progressively easier, it's ballsy to build a game around the idea of punishing difficulty. Yet Super Meat Boy ended up dominating the platformer conversation in a year in which Super Mario Galaxy 2 was released. Clearly, the ambition paid off.
But ambition - which I'll define as risk-taking with a purpose - does not always have to be fulfilled to be noteworthy. Failure can often be more instructive than success in this regard. (Listen to Three Moves Ahead's dissection of what went wrong with the spectacular bomb Elemental for proof.)
Aside from the York/Zach conceit, I'm not sure Deadly Premonition does anything especially innovative. In fact, there's a lot about the game that is pretty baldly derivative. The Other World conceit is cribbed from Silent Hill; the stand-and-shoot combat from Resident Evil; the labyrinthine sidequest structure from any number of JRPGs; the open-world exploration (and attendant ephemera) from any number of sandbox games; and the surreal atmosphere from David Lynch. And in every case, the source material is better implemented.
Where I think Deadly Premonition is especially ambitious - and successful in fulfilling its ambition - is in its world-building.
Given the challenges facing the development team, it's remarkable Greenvale is as engaging a setting as it is. In the post-mortem on the game in Game Developer, SWERY even admits his project - "replicating an entire rural American town from inside Japan" - was "rash." But it's not for the reason you might expect. As game designer Chris Pruett notes in his terrific analysis, "chalking Deadly Premonition's strangeness up to the cultural divide is a vast oversimplification." In other words, whatever discordant notes Deadly Premonition hits are not due to, as Chris says, "normal 'Japanese insanity.'" They are, by and large, intentional, meant to evoke a surreal feeling.
SWERY again: "For the story, I kept as much sense of reality as possible while embellishing it with a somewhat dreamlike milieu. As a result, I believe the sections depicted as reality have a dreamlike feeling, and likewise, the actual dreams are given a sense of realism - we were able to smear the boundary line. There are very few video games so particular about depicting such things, and I think Deadly Premonition may have become a unique example among them."
While the textures and animations in the game are often ugly, there is a different type of detail present in Deadly Premonition that's more important. As SWERY explains in the Game Developer post-mortem, each NPC has a 24-hour pattern of activities, specifically designed to reveal characterization. Peer into Thomas' apartment in the morning to observe him at his morning rituals. Follow Nick and Olivia around and you may uncover a possibly adulterous soap opera in progress. Head to Emily's house after hours and help her become a better cook.
What works particularly well here is that because Greenvale's cast is small and memorable enough for us to become interested in each character - and because this is a mystery, where our protagonist is expected to investigate - observing NPC behavior leads to meaningful conclusions. NPC activity patterns are constructed in service of not only a more realistic setting, but also of layering the story with additional complexity. (By contrast, it's debatable whether the same could be said about NPCs in Red Dead Redemption.) Examining this aspect of the game, Danny Weissenberger goes so far as to claim Greenvale is "the best-realized location in the history of video games."
Even ignoring the constraints of budget and technical expertise, it's quite a risk for a developer to spend so much energy and time fleshing out NPCs, especially when the bulk of the player's interactions with them are optional. Yet it's all in service of the game's vision of its world. SWERY: "I was insistent that the content of our side missions always draw upon aspects of the NPCs' personalities, thus adding further depth to the world of the game." And there's the key to an ambitious game: The risk-taking has to have a purpose that furthers the larger goals of the game. (To some extent you could argue that the loyalty missions in Mass Effect 2 operate on a similar principle, but not only did those have tangible rewards, they also never felt truly optional. They weren't all that risky.)
Of course, the entire Deadly Premonition project was a giant risk. Here was a detective game that was part talky adventure, part survival horror, part open-world, and all kinds of Lynchian crazy. How to sell this thing? The North American box cover is a misleading mess, billing the game as much more of a horror title than it truly is; one wonders if that was a marketing tactic meant to sell to the Silent Hill crowd. It's ironic that the other big measure the publisher took to mitigate risk - compelling the dev team to insert combat - ended up being one of the game's major turnoffs. And yet, even there, SWERY's ambition is evident: Although the "hold your breath to avoid zombies" mechanic is kind of broken and unnecessary, it was an interesting and risky attempt.
My other GOTY candidates, Mass Effect 2 and Red Dead Redemption, are both obviously bigger in scope and far more polished than Deadly Premonition. But with respect to the outstanding efforts of those development teams, neither game was as ambitious, as risky, as SWERY's. A sprawling space opera with intricate dialogue and moral choice systems? Kind of a BioWare specialty. A massive open-world shooter with incredible detail and freedom? Right up Rockstar's alley. Both games were based on existing, established IPs. Neither deviated too far afield from a proven formula. Both had millions of dollars of marketing capital to burn, and both were created by large studios with hundreds of experienced developers. By its nature, a videogame is a risky product, but these particular games were always going to be safer bets than most others.
So sure, there may be some element of rooting for the underdog at play here. But I think Deadly Premonition's ambition speaks to the thing SWERY has his team nailed: "we poured our love into the game." Due respect to other games, but the love that went into Deadly Premonition is more tangible than in most AAA titles.