What Games Are: The Scientism Delusion

Editor’s note: Tadhg Kelly is a veteran game designer, creator of leading game design blog What Games Are and creative director of Jawfish Games. You can follow him on Twitter here.

Making a game is rarely as simple as having an idea, coding it and shipping it out the door. It looks as though it should be, and every year there are one or two media stories of studios who do exactly that, but in reality it’s not that common. Usually there’s a lot of playtesting, blind alleys, a need to rework or rethink how the game works and to iterate. This has always been the case, though perhaps in this more connected age it’s easier to iterate against the market than behind closed doors.

Bad games die in obscurity. They vanish into the great unwashed of the App Store’s 150,000+ games, never to be seen again. Their failure is commonly equated with poor execution and iteration, but it also happens to many well-executed games. A game may be diligently developed, iterated, analysed for user behaviour and whatnot, and yet they never go anywhere. And conversely there are many examples of games that you might not consider any good, which sell big.

So it’s reasonable to conclude that gaming success is not just about delivery. There’s another factor at play, one that you might call hype or marketing. By which you probably mean a combination of advertising in channels like ChartBoost, publicity and trying to please Apple so that you get featured in the App Store. It all seems to be a part of that same execution equation, but wider. Iteration+execution+marketing=success.

And yet it still doesn’t work.

Formulaic thinking is incredibly common in the games industry. It comes largely from an engineering mindset, which is unsurprising given how closely developing games and software have always been linked. In software every application is essentially trying to solve a problem by providing features, and over time an application builds into a suite of features that users like. And sometimes the resulting applications become so over-complicated that a new developer reinvents them for simplicity, and the cycle begins again. It almost seems scientific.

To many game makers it’s similar. Games seem to solve entertainment problems, and game developers provide features to do so. Since all problems boil down to tests and verification, like a science, the method should be clear. So the strategy becomes about trying to find corners, niches, better problems or out-solve the competition on existing problems. This is why most games tend to be conservative in their ambition, and why genre thinking is highly prevalent.

From a developer side, genre thinking largely views games as a series of problem categories. A new form of slot machine on your iPad or a game where an avatar runs along an endless path emerges – and suddenly everyone is talking about the “social slots genre” or the “endless runner genre.” They talk about how they need to be “in that space” with “their play.” They search for external validation through observing what the competition has done, and use that to justify why they should take that risk themselves and add their secret sauce (which leads to some amusing places, such as the awkwardly named Running With Friends).

But, again, this approach is usually far more unsuccessful than successful. Temple Run and Jetpack Joyride may have hit on something, but for the dozens of other developers who then made an endless runner, not so much. Zynga and Supercell may have hit the mother lode with a farming and a strategy game, but most of the other studios who tried to replicate this success failed. Their successes were not just about iteration, execution and marketing.

Iterate-execute-market thinkers don’t tend to make the logical leap and conclude that their approach is wrong. They almost never seem to say to themselves “Maybe the problem is that we were too conservative, too timid, too risk-averse.” Instead they conclude that they just didn’t iterate, execute and market hard enough. They just didn’t get the combination of ingredients quite right to make the secret sauce sing. But there must be a secret sauce right? I mean, look at Supercell, they say. They got the sauce right.

The reality is that this whole way of thinking may look like science, but it is not science. It’s scientism, which is the axiomatic belief in the empirical and that that which seems science-like is probably right. Furthermore, in the absence of empirical validation, the assumption that there is always a formula out there somewhere. And that that method is above reproach.

There are some corners of the gaming world where formulaic thinking hold true. Sports, casinos and various others show regular evergreen patterns over long periods of time, but even there the patterns that scientism-believers think they see are overblown. What is often forgotten in the rush to confirm a bias for success is that entertainment industries do not solve problems, even if they are software-based. Toy Story does not solve a 3D-animated toy movie problem, and the “Life of Pi” does not solve an inspiring book problem. They do something else, which is why those entertainment industries are always full of big surprises from bold makers, and legions of follow-on publishers trying to understand Yann Martell’s secret sauce.

What I’m talking about is understanding the difference between the act of creation and the process of creation. Of course you should be iterating. Of course you should be evaluating and playtesting and trying to figure out what will make your game fun. Of course you should be looking at the market and trying to find the right approach to get where you want to be. But the thing you should not be doing is following others’ successes to the letter on the cod-understanding that that constitutes a scientific approach. All it is is a fearful approach.

Just as with movies and books, breakout games start from the place of having an idea of your own. On the App Store there are thousands of apps all working those same promotional channels, yet the majority go on to fail. If scientism really was a proven way to succeed then companies like Zynga would not find itself in continuing difficulty. EA wouldn’t be shedding jobs a-go-go and THQ’s debts wouldn’t be so crippling. These are (were in THQ’s case) all companies whose primary asset is their ability to market and promote, to push product and dominate attention, to apply formula thinking. And yet their failures far outnumber their successes.

Rather than applying mystical formulas to self-selected confirmation biases, and assuming that secret sauces are only the right pinch of salt away, the more successful approach to games usually revolves around what Seth Godin calls tribal marketing. Rather than say “We have an idea that fits a genre, let’s make that idea, then when we make that idea let’s market that idea, and if we just do that well enough then we will succeed,” the really successful game maker usually starts with “I have an obsession. Do other people have the same obsession? Let’s see if there’s a game there.”

In other words, rather than relying on the law of averages to stumble you into success in a made-up genre invented by the industry, think like Shigeru Miyamoto – who turned his own fascination with weighing himself into the Wii Fit. Of course you might say that Miyamoto has Nintendo behind him, so he can afford to be more ambitious than you or I, but this is not true. He can afford to spend more money on his ambitions than you or I, but even Miyamoto started somewhere much smaller. Today’s Nimblebit was  yesterday’s Nintendo, which is to say that the right creative approach can scale.

So what to do? The answer lies in two terms: “alignment” and “marketing story“.

The first means getting aligned with your people. The scientism-fan is often somewhat like the analyst, the intellectual or the craftsman, used to holding the market in cool regard and wondering what key will unlock it. He’s like like the French politician Alexandre Auguste Ledru-Rollin, who apocryphally said “There go my people, I must found out where they are going so I can lead them.”

Successful game designers are often obsessives in one kind of game, whether it be simulations, driving or fighting games. The aligned game maker is connected very deeply to her audience because she is essentially one of them, and the very successful ones are so because they are aligned to large-enough audiences which make their obsession viable. Will Wright, Todd Howard, Jane McGonigal, David Cage and many many others are all true believers who have found large groups of like-minded people.

And the marketing story? All the most successful products, from burgers to $500,000 cars, resonate with their intended audience. Whether that product satisfies their conception of what it should be or represents leadership toward where it goes next, successful products tell the people the story in which they already believe. Skyrim means something as much to roleplayers, more than being just what it is. So too FIFA Soccer and Gran Turismo. So do most other gaming successes (the main exceptions being outright novelties and platform wedges).

But of course getting alignment and marketing story right is easy to say but in reality very difficult to do because it requires a lot of self-examination. You can’t be aligned and also cooly detached at the same time. You can’t espouse a marketing story unless you actually believe in it. You’ll just get confused, and come across as such. It means getting past the scientism while at the same time not losing sight of due diligence.

As game makers we’re  too used to thinking of games as a business rather than a passion, and so our on-the-street sense of connection turns into scientism. We often forget the innate joy of our own obsessions and our flights of fantasy. We often stop asking the question “What if you made a game that…?” and instead say “What genres do we need to tap?” We fall too deeply into thinking of players as users, to be measured in daily-active and monthly-active cohorts and broken down as revenue generators and customer segments. We forget that we’re entertainers, not engineers, and that genres are mostly bullshit to an audience that just wants to have fun and feel the fantasy of the worlds that we create.

Sometimes scientism works really well, but more often than not the true source of its success are rooted in good timing. Zynga may have appeared to be a company of game-scientists solving problems, but the reality is that it had the smarts to be risky with first-mover advantage back when everyone else was waiting for validation. It’s a similar story for GREE, BigPoint and many others following essentially the same path. The mysticism of their methods is mostly well-understood and usually starts with someone taking a huge risk before they started to apply scientism to their thinking.

We intellectuals, analysts and craftsmen like to talk of secret sauces and whatnot because they make us feel reassured. All we are doing is buying into a marketing story of our own, the one that says games will one day be solved, in the future, because we are uncomfortable with uncertainty. Nono, we think. It can’t be as dumb as simple luck and timing. That Notch guy? Maybe he just got lucky. But that Pincus guy? He’s got the method, the secret sauce and the plan.

Nope. He just had the guts to gamble huge before anyone else. All the rest is mostly fiction.