Balancing game difficulty against player progression

It seems that the current climate of games seems to cater to an obvious progression of player power, whether that means getting a bigger, more explosive gun in Halo, leveling up in an RPG, or unlocking new options in Command and Conquer 4.

Yet this concept is not exclusive to video or computer games -- even in Dungeons and Dragons players can strive to acquire a +2 sword to replace the +1 weapon they've been using.

Yet as a systems designer, the concept of player progression is giving me headache after headache. Should I balance around the players exact capabilities and give up on a simple linear progression? (I think ESIV:Oblivion is a good example of this) Is it better to throw the players into an "arms race" with their opponents, where if the players don't progress in an orderly manner, it is only a matter of time until gameplay is unbearably difficult? (4th Edition DnD strikes me as a good example of this) Perhaps it would make most sense to untether the core gameplay mechanics from progression at all -- give them flashier, more interesting (but not more powerful!) ways to grow?

This seems like more of a philosophical question with no single "right" answer.

A lot of players enjoy feeling like they're progressing and getting stronger, so removing progression from your game is not an automatic solution. For games in genres built on progression mechanics (like RPGs), removing that core would make for a relatively unique game... but one that you'd have to be careful about since you'd be treading in unfamiliar territory.

Having enemies scale (as with Oblivion) is one way to prevent the game from getting too easy; the danger here is that sometimes players can feel as if they can never get ahead: the enemies level up with you, so it doesn't feel like progression.

The more traditional way around this is simply to divide enemies into areas, where moving to the next area means harder enemies, and it's up to the player whether to move forward now, or grind for a bit in the previous area to "level up" to make the next area easier.

An important thing to realize here is that the player's perceived difficulty is going to be a function of both their power level, and that of the enemies. If you give the player a huge power boost but give the enemies an even greater boost, the game will get harder in spite of the player's power increase. So it's a matter of what you want the difficulty curve to look like in your game. Do you want it to get progressively harder, or easier? Do you want to time the hardest points in the game to coincide with the most dramatic parts of your story arc? You can do it any which way and there's no "right" or "wrong" answer here, it comes down to your design goals and then finding the best difficulty progression to give the player the play experience you're after.

I think it heavily depends on the kind of experience you want to give to the player.

As a player, I didn't like the way Oblivion handled it. One of the positive feelings as a player gaining power and abilities is the fact that you can now dominate what used to be difficult. With Oblivion and it's constantly tuned-to-your-level difficulty system, you lost the feeling that certain areas were safe and other areas were dangerous. Sure, you were fighting harder monsters, but why they showed up felt unnatural.

Personally I prefer the more traditional mechanisms. Areas have predefined difficulties, and as the player progressed areas that were previously impossible now opened up. And areas that used to be hard are now a cakewalk. It gives you a better sense of progression.

That's not to say that all dynamic difficulty systems are bad. I just think that they need to be more subtle. On Sin Episodes we did a very complicated system that didn't turn out too terrible. The gist of it was that there were a lot of little factors that played into what knobs would be adjusted as the player goes through the game. If the player got a lot of headshots, more enemies would spawn with helmets (which meant you had to shoot them in the head twice). If you were proceeding through the game quicker than our par time, the difficulty knobs got turned up. If your average health was lower than what we would like health packs would drop more often.

Unfortunately the curve wasn't linear. One of the switches we implemented was to level up the kind of guys we would spawn for difficulty. The last jump (assault rifle to machine gun) was a lot more of a jump than some of the earlier ones (pistol to assault rifle), and you could get into a situation where the designers would place two assault rifle guys and the game could spawn three machine gun guys (which is a much, much more difficult kind of fight). But compared to Oblivion, it felt like there was still progress through the course of the game as the designers were mostly in control. The dynamic difficulty system itself accounted for only adjusting it up or down a certain amount based off player performance.

I concur with Tetrad; On the one hand I am a huge fan of JRPGs and I can be a total stat whole - collecting all the best equipment making sure my uber spells are totally uber etc.. but that is a grind and the grind has to be fun. FF series does a good job at making grinding and leveling fun because they pace the level progression with the storyline very well.

However some other game types really do make it a grind and it becomes extra boring. Then there is the problem which happens constantly in FPS shooters is that unless you give the bots auto-aim headshots the game is a breeze. The player is usually way over powered - has much too much ammo - and a gigantic assortment of heavy weaponry.

One of my fav games is Guild Wars, you can only get so powerful and that's the end; but what makes it really fun is the skills system - you can only take 10 skills with you into battle - so picking out the best combination is very entertaining and testing skills builds is very fun as well. So even though there is a level cap you still have tuns to explore and new skills to unlock. The Player is never over powering and the only grind comes finding those special skills or looking for collectables and appearance oriented content.

I'd start with reading this series on balance by Sirlin, so we know what we're talking about when we say that a game is "balanced". Go ahead, i'll wait!

Done? good! if you didn't, then at least read part 1, that has the bare minimum you need.

A multiplayer game is balanced if a reasonably large number of options available to the player are viable--especially, but not limited to, during high-level play by expert players.

--Sirlin, December 2001

A multiplayer game is deep if it is still strategically interesting to play after expert players have studied and practiced it for years, decades, or centuries.

--Sirlin, January 2002


CCed here for those that TL,DRed still. While that says multiplayer games, it largely applies to single-player ones as well.

I think i'd set up a system to track player metrics; for instance, in an overhead shmup, what weapon the player favors, the types of ships he's best at killing, how much health he's lost lately, etc. and mix it up so that the players that are doing well get more tough enemies and less powerups, while players that are struggling get enemies that they can fight and better tools to do so.

I'm designing such a shmup (shameless plug: the blog for it), and i plan to do that. In short, the player will have to hunt enemy ships and peril the dangers of space to gain enough energy to fight the bosses, as well as look for upgrades for their ship. The thing to remember is that players will naturally want to progress through your game, so you can use that to drive your game forward.

The advantage of a dynamic difficulty system is that all a player has to do to get the "lower-level" areas back is that they just need to play poorly; the game will think they're having trouble and give them an easier time.

I wouldn't throw the player into an arms race by any means, but if it's a campaign type game (like a stealth shooter or something) then just sitting around while you have things to do should definitely ratchet up the pressure.

• Of course, the danger of dynamic difficulty adjustment is that players purposefully play suboptimally to prevent the game from getting too hard, so you have to be careful about this. DDA isn't a no-brainer panacea by any means. – Ian Schreiber Aug 30 '10 at 15:05
• @Ian: in that case, it's sort of their fault, and they can play back through to get a more optimal experience if they want. Just as someone who's an expert in a game can pick "Easy" if they want... :) – RCIX Aug 30 '10 at 20:07

One thing you might think about here is looking at more strategy-oriented games like Civ, Starcraft and roguelikes such as Nethack and Crawl. In these games, game difficulty and player progression are not simple single-variable functions. The gameplay evolves over the course of the game.

In the early-game the goal of the player is to simply survive, progress, increase in level, get the next equipment with a +1 bonus, etc. Generally everything the player finds is better than what they currently have - whether that's a new fishing spot in Civ, or a new sword in Nethack.

The mid-game often features the player exploring the environment for gameplay options. Now, maybe they're setting up strategic forts, or collecting equipment with special effects that have situational uses, but still simply trying to survive to accumulate more stuff.

The late-game forces the player into a showdown, usually with the toughest opponents, but also with the most options at their disposal. More experienced at playing the game, now not only must the player survive, but also exploit the effective weaknesses of their opponents with the tools they've discovered and refined from mid-game forward.

You can see the same progression in Final Fantasy games - in the early game, most enemies can be killed with a simple attack, and maybe a couple healing spells as needed. Mid-game enemies begin to appear that are more difficult, but with obvious elemental or status weaknesses to exploit. Late in the game, the enemies might be practically unbeatable except for clear well-thought-out strategies based on less obvious weaknesses or AI script.

• Early game: Survival. Resource collection.
• Mid-game: Exploring the environment. Strategy building.
• Late-game: Strategy execution. Exploiting resource options.

It doesn't HAVE to be this way, but I've seen the pattern in many games from console RPGs to PC strategy games, and even handheld tower defense games.

In a game like Oblivion, the difficulty is that the player has to make choices in the strategy building in the early- and mid-game by choosing how to level up that can have drastic consequences late-game because there's no option to change that strategy later. It's how you've built your character. The same is true in Diablo 2, and many other games that allow you to select skills for your character without the ability to change them later on. A skill-based character advancement thus takes a bit more fore-thought into how individual skills can be used to build a cohesive strategy. In Diablo 2 you can see how the developers struggled with it by looking at how often they tweaked the skills over the lifetime of the game. It's easy for an inexperienced player to reach a point where their permanent strategy choices are no longer effective against the opponents they're facing. That doesn't necessarily make the game unbalanced - it simply turns the skill progression choices into part of the player's strategy, and may require several playthroughs to develop an optimal strategy.