This year marks my thirteenth in the industry, and I have created two games (Civ3 and Civ4) which I consider roundly successful. Unfortunately, both games came from my first five years; projects from the following eight years were all either executed poorly or cancelled outright. How did I lose almost a decade of my professional life?

To answer that, I need to start in October 2005, with the release Civilization 4. The game was critically praised (94 Metacritic, highest ever for a Firaxis game) and hugely profitable, selling over 3 million copies on a modest budget. It won Game of the Year awards. The soundtrack, which I selected and edited, was acclaimed. The theme song, “Baba Yetu,” won a Grammy, the first ever for a video game. One mod, Fall from Heaven, developed a large following of its own. Civ4 was that rare project in which everything that could go right did go right.

I started the project from scratch, wrote every line of game and AI code, grew the team over two-and-a-half years, and shipped the game two weeks ahead of schedule. I gave everything I had to give to that game; my only regret was that I did not have the stamina left to contribute meaningfully to the expansions.

Six months later, after the patching process finished and my energy level returned, I needed to decide what to do next. Firaxis (or rather, the new owners, Take-Two) offered me the chance to lead Civ5, which I declined as I couldn’t make the radical changes necessary to justify a new version. In contrast, I was overflowing with ideas when Civ4 began as Civ3 had been an incredible learning experience. Most of these ideas were now explored, so I didn’t have much left to give the series.

I did, however, have plenty of ideas for new strategy games, which I was very excited to make. I had proven myself as a designer with Civ4, and it was time to make a game wholly of my own. I pitched my favorite idea, and it was rejected. I floated a few other ones, and they were declined as well. The company couldn’t afford to put resources into a new project with Railroads! in full development, Revolution beginning to ramp up, and Civ5 looming on the horizon.

Ultimately, Firaxis was in a bind. Because developments costs were rising and the window for PC retail games was shrinking, new IP had become increasingly risky. Furthermore, the company had a wealth of proven IP from Sid’s back catalog to develop, so the opportunity cost of producing new IP was huge. (In fact, between 1997’s Gettysburg and 2013’s Haunted Hollow, Firaxis released no new IP, with the possible exception of SimGolf, which was certainly original but also traded on the Sim brand.)

My pitches had all been for smaller projects, with budgets between one and two million dollars. The problem was that, at the time, no distribution method supported games of that scale. We only needed to sell a few hundred thousand copies to break even – a very reasonable goal with the company’s reputation – but the retail channel didn’t support such projects. PC games had to either sell millions in a $50 box, which was only viable with a large budget, or sell in a $10 jewel case, which was the shovelware market. Steam had just begun reaching out to third-party publishers – by 2007, only id, Capcom, and Eidos were on the service – so digital distribution was not an option.

Today, of course, things are much different as a market exists for games of all prices, from free to $60, and of all budgets, from less than $1m to more than $100m. Digital distribution, microtransactions, and platform diversity have altered the landscape of the industry, and it is likely that if I was pitching a game inside Firaxis under today’s conditions, we could have made it work. At the time, however, my only option was to hang on as a creative director, giving advice to the active teams while prototyping games which might never come out.

In fact, if I had known then what the next six years of my career would be like, I would have likely stayed at Firaxis and assumed that something good would come of it. I loved working there and still love the company, but I am only human; I felt that my work on Civ4 had earned me the right make a game of my own. Being denied that hurt, and I made a perhaps hasty decision to go.

I interviewed at the companies I respected most – Blizzard, Ensemble, Valve – and settled on joining Maxis to work on Spore. Will Wright had amazed developers and journalists with the surprise reveal at GDC 2005, and joining the team meant working on one of the highest-profile games in the industry. I have compiled my thoughts on Spore in a previous post, and despite the game’s flaws, I can’t say I regret working on it. The team was inspiring and immensely talented, and I wanted at least to ship something before too much time had passed. I joined to finish the project, and the game was done 18 months later.

The other reason I joined Maxis was that they wanted to support my future projects; if I proved myself with Maxis, some interesting opportunities existed post-Spore. Unfortunately, the game underperformed, and EA’s stock cratered shortly afterwards. (The two events, of course, were not entirely unrelated.) The company laid off a chunk of its workforce and retreated from new, risky IP towards fewer, safer titles. The chances of me pitching a new, innovative strategy game inside of EA, one which I could commit to fully and protect from compromise, dropped to zero.

I was at a crossroads again, and I didn’t know how to make a game my way inside of EA. I actually spent the months following Spore’s release pitching a browser-based strategy gaming company to various venture capitalists in nearby Silicon Valley. At the time, asynchronous and free-to-play games were hot investments, and the best way to control my next project would be to found a company to build it. Unfortunately, my vision was too niche for the VC’s – I wanted to make core strategy games that would grow from player modding – and I couldn’t find funding.

Instead, I found refuge at EA2D, a browser-based gaming studio at the Redwood Shores campus. Their main team was building Dragon Age Journeys, a Flash-based spin-off with tactical, turn-based combat. Mark Spenner, the studio’s GM, gave me the opportunity to prototype the web-based Strategy Station for a year, which was essentially the same project that I had unsuccessfully pitched to the VC’s. I built three different moddable strategy games that could be played online asynchronously, using the Google Web Toolkit as my browser engine.

I released the games with little fanfare; in fact, I never once mentioned the site on this blog although I did talk about it on one episode of Three Moves Ahead. In some ways, I was afraid of publicity or success; I didn’t know how to make the site viable, either scalable technologically or profitable financially, but I was sure that few decision makers inside EA would share my vision. I decided to make as much progress as I could on my own and hope for the best. (I rationalized that they couldn’t kill a project without a development team.)

The site never grew beyond a few thousand users although it developed a dedicated audience in Japan, with some players finishing thousands of games. (Here’s a popular Japanese blog dedicated to the site, and here’s a video of Kingdoms, the most popular game, being played with the Japanese language and art mod, which replaced the human soldiers with bunnies for some reason.) I didn’t know how to justify asking for resources from EA for such an odd project, and when it became clear that EA2D needed a success to justify its existence, I preemptively killed the project myself. (I did attempt, unsuccessfully, to take the project outside of EA, so that the Japanese community could continue playing. An independent Strategy Station could have served an enthusiastic niche audience, but I had created a personal pet project inside of EA, one inappropriate for a company of that size.)

The potential success EA2D needed became Dragon Age Legends, a loose sequel to Journeys, built within Facebook. Social games were hot, hot, hot in 2010, and I wanted to see if I could make one that respected core gamers yet took advantage of the new format. The results were mixed. The game actually tested quite well within the company, especially among the executives – CEO John Riccitiello and Games Label President Frank Gibeau both had very high-level characters and spent not a little money on the game. Nonetheless, the friction of the energy model, the core gamer hostility to free-to-play, and the mismatch with the Facebook audience ultimately doomed the title.

For myself, I genuinely pursued the project as an interesting experiment, but the game was clearly not what I would have developed if I controlled my own destiny. Sadly, I did not even try to make the games I wanted to make within EA. I was unwilling to engage in the politics necessary to pitch them, doubtful they could be approved anyway, and afraid of how they would be handled if they were approved. I was, essentially, giving up before even trying.

The summer of 2011 was probably my lowest point in the industry. One of my favorite sites, Rock Paper Shotgun, lambasted Legends for its business model, and the game’s audience had dwindled down to 20,000 daily active users. The game was not the success EA2D needed to support future projects, and the group became BioWare Social and began to bleed talent. I had no idea what I should do with myself inside EA.

Enter Zynga, or – rather – enter Zynga East. Brian Reynolds and various other refugees from Big Huge Games had founded a Zynga studio in Baltimore to make social games, resulting in the hit 2010 game Frontierville, which included a number of important genre innovations, such as the energy bar and story-based quests. Zynga was flush with cash, and Tim Train, the studio’s GM (and Brian’s old BHG business partner), recruited me with the promise of developing a browser-based game on my own terms. They wanted to carve out a protected space in their Baltimore studio in which I could prototype safely.

I worked at Zynga for less than 18 months, and it was, needless to say, an interesting experience. I was indeed given the freedom to work on the game of my choosing; it was playable within a few months and was quite popular around the office. In some ways, however, I had too much freedom. Since the game had little oversight outside of Baltimore, the game had no real political support. I did not push the game through the greenlight process as I was afraid of executive interference, so it lingered on as a mystery project, free from both the negatives and the positives of the company’s attention. Thus, when Zynga East wound down after CityVille 2 performed poorly, the game was easy to cancel.

Ultimately, I was given incredible freedom at Zynga, but the project was likely doomed from the start. However, the only person to blame is myself. When leaving EA, Zynga was the easy option for me to take – the pay was good, the personal risk was low, and I was making the game that I wanted to make. The problem is that the game I most want to make is one that actually ships, and excuses about external forces are just excuses. I joined Zynga knowing that I would not have control over my game – at any moment, it could be altered drastically or cancelled outright.

Looking back at my post-Civ career, I compromised the games I wanted to make with what my employers were willing to fund. With Spore, that compromise meant finishing someone else’s game. With Strategy Station, that compromise meant working without a team. With Dragon Age Legends, that compromise meant turning an RPG into a social game. With Zynga, that compromise meant making my game under the shadow of indifferent management. I was giving up before I had even begun.

Well, I am giving up on giving up. Only one option exists if I care about making games my way, one which will demand much more of my time, my energy, and my security. I have a backlog of game ideas, more than I will ever be able to make in one lifetime, which means that I am already running late.

It is time for a change.

It is time to go independent.

Follow the story at mohawkgames.com.