Spry Fox Games took a huge undertaking last year by releasing Realm of the Mad God. Not only was it an MMO which is already hard to manage and constantly develop, it was also a bullet hell shooter which requires constant balance updates. How did Spry Fox Games handle this enormous task? They didn’t listen to their players.

Gamasutra is reporting that Spry Fox Games CEO David Edery delivered a post-mortem on their MMO shmup darling that pissed off players for the changes they made, even those if those changes made the game better.

Edery references Star Wars Galaxies as the text book example of “if it ain’t broke, don’t fix it” philosophy of game design, but he argues that there’s always the exception. in the case of Realm of the Mad God, their changes to the core game saw their player retention rate go from 12.7 percent to 37.6 percent.

What changes were made to the game? The developer found that players, like myself when I play a shmup, never actually look directly at the boss. Players were firing in the boss’ general direction while avoiding the bullet hell that was heading their way. They even found that some players were just looking at the mini-map and firing at the boss instead of using the actual game screen.

To fix this, they developed a system that would allow the team to make changes directly to the game. This allowed them to change the behavior of the enemies based on a variety of factors. They then made the players’ projectiles have a shorter range and slowed down the enemies’ projectile attacks. This forced players to get in close and pay attention to the game.

In regular bullet hell shooters, these changes would break the game. While it may appeal to some players, most fans are advanced players. So why change what has already been established as the mode of play for a game like this?

Experimentation is key, even in a post-release environment. While most developers and gamers would be shocked to see the amount of tweaking and redacting that happeend to Realm of the Mad God post launch, the developer sees it as a win for them. It allowed to see real player reaction to their changes and if it really did make the game worse, they would change it back.

Using this method of quickly deploying changes to the game, they were able to design, test and deploy updates in under a month. While the changes and the awkward phase in between testing may have pissed off some players, it helped them gain more players than ever before.

The moral of the story is that it’s alright to piss off your players if the end result is a better game. Using players to test changes made directly to the game without external testing might not sit well with some developers, but Spry Fox Games have proven that it is possible. MMO developers should especially take note to the lessons learned here.