Sponsored By
Jon Shafer, Blogger

May 7, 2012

5 Min Read

You can read more of Jon's thoughts on design and project management at his website. You can also find him on Twitter.


I’m hopping off of strategy gaming for one more article in order to talk about a problem that’s prevalent in RPGs, my other favorite genre. Namely: bosses.

Okay, so maybe saying all boss battles are “broken” is a bit of a stretch. But for the most part you could say they’re… misused. This really shouldn’t be much of a surprise, as the model for boss fights that you find in most RPGs has evolved little since its crude introduction in the 1980s. At the end of every major level you’ll run into a monster just like all the others, except it has 5 times more health and does 5 times more damage with each attack. Maybe they have a special weakness, or a particularly devastating attack that you have to figure out. Yay! But once that’s done there’s no depth or strategy to be found under the surface.

More troubling still, even if the fights themselves were interesting, the fact that they exist at all hurts the overall experience of the game. Why? Well, let’s explore this topic in more detail.

Strategy… or Pattern-Matching?

One of the reasons why many fighting games and shooters are fun to play is because you never know exactly how your opponent will react. You’re familiar with what abilities or weapons are available, but that doesn’t mean your foe will attack the same way in every engagement. Sure, his character has a really devastating Strong Punch, but I know that, and he knows I know that – maybe he’ll try to get crafty and hit me with a Low Kick instead? This sort of strategic thinking, also known as “yomi” is the reason why these types of games can be infinitely enjoyable and replayable (a concept explored by David Sirlin in his excellent article).

Nearly every RPG boss fight completely lacks this strategic element. Instead, they ask the player to simply replay each battle several times, or patiently watch their enemy for as much time as necessary in order to identify its attack patterns and weaknesses. Once the formula is unraveled, the player knows exactly what to do and victory requires nothing more than spamming the ‘correct’ attack and keeping one’s health from dipping too low.

Now, achieving yomi with computer opponents isn’t really feasible. They are, after all, just lines of code being executed and not actual, strategizing humans with nuance and subtlety. But players can still be required to develop strategies and make trade-offs, even against AI enemies. How likely is it that the enemy uses Ability X versus Ability Y? How long until Ability Y can be used again? Am I willing to take some risks and perform an all-out attack this turn, or should I play more defensively? This sort of thinking requires players to actually know what an enemy is capable of, instead of make battles interesting by ‘surprising’ them. This is really no different from a single-player strategy game where the human must develop a plan for how to best proceed, identifying risks and opportunities, protecting weaknesses, and so on.

Another way bosses could be made more interesting is by imposing limits on them. Maybe they have a mana meter which can, you know, actually deplete instead of being effectively infinite. Players might know that a monster can shoot off a deadly ability one or two times per battle, but they don’t know when it’s coming. If one is too patient, the boss can whittle down the good guys with just basic attacks. Too aggressive and the enemy fires a blast that wipes the board. Varying AI personalities can give hints as to the likely outcome, but it’s still up to the player to develop a plan and weigh risk versus reward.

Bad Behavior

“I’m not going to spend my mana or potions because I know there’s a boss coming up.”

I’m pretty sure I’m not the only one who’s had this thought run through their mind. The last thing we designers should want is to actively discourage players from using the cool systems we create. Nearly every game benefits from providing long-term versus short-term trade-offs, but if players can set their watches by the frequency of boss fights then you’re not really offering players a choice – you’re just encouraging them to be conservative. Nearly every RPG I play I find myself with a full inventory at the end of the game. Is this really what the designers wanted? I obviously can’t say for sure, but I doubt that it is. It certainly wouldn’t be one of my goals.

This is a problem that owes more to the predictability of boss fights than their mere existence. If players know that every level has a midboss and an endboss, and that they’ll be roughly X% stronger than the last one they faced… they will be planning around that. Some people will still naturally hoard items and mana when they don’t know what’s coming, but this is as an issue that can be addressed in a number of ways. On the other hand, always planting bosses in the exact same place guarantees this kind of behavior.


The last problem I’ll talk about is less discrete. Games are meant to be unique experiences, ways to escape the monotony of everyday life. Players like to explore and make cool discoveries along the way. Regular and predictable boss battles tend to provide theopposite.

You don’t want your game to become rote. Players are much less engaged and excited when they always know what’s around the next corner. The best stories are always those with unexpected twists and turns. Pacing in a game is no different.

There’s nothing wrong with sprinkling in some tougher fights to keep players paying attention and to raise the stakes. One of the best features of open-world RPGs is that you never know when you’re going to run into something really nasty while you’re just wandering around. This approah could easily be adopted by more linear games. What if some levels thrust the player into a boss battle only a minute in? What if some areas hadno boss at all? This would keep players on their toes, constantly wondering what’s around the next corner. As designers, our goal should be to prevent game experiences from turning into yet another exercise of ‘going through the motions’. Adaptation and discovery are key – not just with boss fights but also every other aspect of a game.

- Jon

Read more about:

Featured Blogs

About the Author(s)

Jon Shafer


Jon Shafer is a lead designer at Stardock Entertainment, currently directing an unannounced project. Prior to working at Stardock, He was the lead designer and principal gameplay programmer for Civilization 5 while at Firaxis Games. Jon lives in Michigan and (when he's not fighting-the-good-fight on behalf of PC games) spends his free time cooking, listening to podcasts and playing strategy games/RPGs.

Daily news, dev blogs, and stories from Game Developer straight to your inbox

You May Also Like