I am writing a game, in the vein of a jRPG, in which you begin playing as the main villain. You don't know it. You think you are playing the hero and the more successful you are and the stronger you make them the harder they will be to defeat at the very end of the game.

My main concern is anyone who puts in a significant effort early game could feel punished for it. It is also troublesome if the player finds the challenge of the final boss insurmountable because of actions from very early in the game. Is there a way I can signal early on in the second act that if they over-powered the villain it may haunt them? And there is also the possibility the twist will get out and people will just game it from the beginning. It is my intent to get some replay value out of it by having people do multiple runs based on different villain playthroughs. But is it a distraction if everyone learns the twist. Or conversely is it better if I reveal it myself in-game so everyone knows what they are working at from the start?

How do I make my potentially disengaging design hold up without turning people away?

  • 1
    \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – user1430
    Commented Mar 11, 2019 at 15:40

12 Answers 12


Having the player play against their own earlier accomplishments actually seems like a viable approach to implement dynamic difficulty. The better the player, the more challenging the game will become.

But when the player becomes aware of this mechanic (and you have to assume they will find out about this before playing - because it's an unique mechanic and you should use it to promote your game), then they will likely intentionally play bad in the first act of the game so they can exploit their weakness in the second act. This might in fact be an interesting meta-mechanic which could be a challenge in itself: How do you successfully complete the first act while staying as weak as possible? This also can become a neat self-balancing difficulty mechanic. The more skillful the player, the more they can challenge themselves by trying to complete the first act with the least mechanical strength possible.

You might encourage the player to try both approaches for the game by using achievements. Give them an achievement for beating their 1st act character in an extremely short amount of time (which can in practice only be accomplished with the "1st act low level run" strategy) and another one for maxing out their 1st act character and then beating it at all.

But I am seeing a few design pitfalls here you need to be careful to avoid:

  1. Is a first act low-level run actually interesting to play or just annoying?
  2. Does the 2nd act become too easy when the player plays this way?
  3. Does the 2nd act become too hard when the player does not play this way?

How to solve problem 1 heavily depends on your game mechanics, how much they allow the player to accomplish with sheer skill but lacking mechanical strength and how satisfying this way of playing feels. Unfortunately the question does not provide enough information about the core game mechanics of the game to provide any more concrete advise in this regard.

Problems 2 and 3 can be mitigated by making sure that the performance in act 1 does not completely dominate the difficulty in the 2nd act. Make sure that the player's success in act 1 is one aspect of the difficulty of the 2nd act, but that there are also many other obstacles for the player which are completely independent from their 1st act performance.

To solve problem 3, you should make it possible for the player to gain more mechanical strength in act 2 than they could ever achieve in act 1, so everyone becomes able to out-grind their act 1 character, no matter how hard they tried. Another option could be to provide an alternative story path where the battle between the act 1 and the act 2 character doesn't happen at all. This path might only become available when the act 1 character exceeds a certain power level. This might in fact be a good opportunity for the "golden ending" where the two player-characters reconcile their differences in a peaceful way and then they face the True Final Boss together.

I am looking forward to beating myself up in your game.

  • 18
    \$\begingroup\$ I love the alternate ending idea because it combines mechanical (stronger character in final battle) with narrative rewards without subverting the premise. \$\endgroup\$ Commented Mar 8, 2019 at 13:29
  • 5
    \$\begingroup\$ Oh yeah, I'm definitely using the golden ending approach. Thanks \$\endgroup\$
    – Summer
    Commented Mar 8, 2019 at 13:37
  • 8
    \$\begingroup\$ @R.., in FF8, leveling up is so hugely counterproductive that the hardest part of a low-level run is figuring out how you can avoid gaining experience. It's not something I'd recommend looking at for inspiration. \$\endgroup\$
    – Mark
    Commented Mar 8, 2019 at 22:20
  • 2
    \$\begingroup\$ @Mark: Exactly - it shows really well what can go wrong. :-) \$\endgroup\$ Commented Mar 8, 2019 at 22:26
  • 1
    \$\begingroup\$ "Having the player play against their own earlier accomplishments actually seems like a viable approach to implement dynamic difficulty. The better the player, the more challenging the game will become." Not if this is an RPG with leveling mechanics. The strength of the player character in an RPG is proportional to time, not player skill. \$\endgroup\$
    – Jack M
    Commented Mar 10, 2019 at 10:51

1. Reward the player throughout the game for building (and later fighting) a strong villain. Depending on the genre, playstyle and target demographic, that might be "just" a higher score or mechanical rewards like more experience points/better loot etc. to smooth out the difficulty curve a little.

Typically, games that allow you to (knowingly) affect the difficulty will also reward taking the harder path in some way. In your case, to avoid frustration, you should make sure that at least some of those rewards are attainable during or even before the second half - ideally with a mechanic that you introduce before the big reveal and switch, which would also help tie the two halves of the game together a bit better.

2. Consider focusing more on "side-grades" for your villain's progress. If you find that doing really well in the early game makes the boss fight too hard, change some of the best rewards to something rare and cool, but ultimately not that poweful. Make it flashy, give it big numers and a weakness that - perhaps - the AI isn't smart enough to exploit but a clever player might be. You want them to think

Cool, so I get to fight against that? Ha, good thing I already know its weakness.

Essentially, reward them with coolness factor, unique options and replayability, which will persist throughout the second half, rather than just bigger numbers.

3. Hide some soft counters to earlier progress in the second half. Design some items/units/whatever to be especially effective against the stronger things the player can attain in the first half of the game. As in point 2, you can make these more challenging to find or use than other options, as long as they're effective enough to level the playing field a bit. A player that did exceptionally well in the early game can be expected to handle that, and the trope of "unlikely hero searches for lost artifact to be able to face nigh unbeatabe villain", while perhaps overused, has storytelling potential because it's not a macguffin to get the plot going but a consequence of the player's earlier actions.


Keep the abilities, not the power level

Let players control the direction in which the character evolves, but keep control of how powerful it is when the players eventually face it. This lets them customize the final battle to their hearts content, but prevents them from gaming the system, and allows you to add a few surprises where you can make things more interesting.

Thematically, this might even make more sense

If time has passed between the end of the first part of the game and the final battle, it is logical that the villain has continued to upgrade themselves beyond what they had under the player's control.

Mechanically, this will feel better to your players

It's not a great feeling for a player to reach the pinnacle of their abilities half way through the game, and then get reset back to basic and then have to spend the rest of the game getting back to where they were before.

Power resets can make for exciting gameplay, but they should be relatively brief, so that players can return to all the fun and powerful abilities they've grown to love.

The solution is to let the second character reach the power levels of the previous one relatively quickly, and then push onward, breaking new and interesting ground.

This can help you pull off your twist, too. If the player can see that they're only halfway towards their peak power level, they won't expect their progression to suddenly end and switch to a different character.

Of course, this doesn't work if the final boss is at the original power level they were left at. But if their power has grown off-screen, then it won't be an issue!

  • \$\begingroup\$ This is the best solution in my opinion. \$\endgroup\$
    – kettlecrab
    Commented Mar 10, 2019 at 18:09

If the game is an RPG, one possible way to deal with this would be to make it a moral action.

First, display things like the enemies being morally innocent, or displaying themes of bloodthirst and war being punished. For example, if you grind too much in certain areas, show the civilisations of the enemies you're grinding against start to wither, or cause the people in town to start mentioning how it's affecting the area in negative ways.

This way, any player that would be grinding for power and to make the game easier would have to directly ignore what characters and locations are implying. This could also lead to a "Hey, I found a way to cheat the game" situation caused by them levelling up too much despite the themes, which would then be punished by them fighting themselves, who has also levelled up a similar amount. In addition, people who choose to respect the area, obey the games wished and make the game harder would be rewarded for their nature.

One extra thing to consider if implementing this would be the external factors that would be necessary to show the difference in power, as if it's just You v Your other self, the attack/defense levels would just vary, and not actually display power. Allies and spells/abilities with set damage are necessary to actually display the difference in power.

Overall, this would make a memorable experience for the player, as the theming is a way to make the entire game lead up to a point where they fight themselves, as opposed to other solutions which may just make people see it as another point in the story.

  • 1
    \$\begingroup\$ I'm not sure if I understand the second to last paragraph correctly. Are you implying to display fake numbers to the player which don't actually reflect the stats used by the game mechanics calculations and make that combat actually work with fixed stats which are independent from how the player actually built their two characters? \$\endgroup\$
    – Philipp
    Commented Mar 8, 2019 at 13:58
  • 1
    \$\begingroup\$ @Philipp Chris is suggesting that you give the player an outside mark to gauge their (and their foil's) power. Take the simple case of damage being Attacker's Power minus Defenders Defense taken from Defender's HP; if you multiply all the numbers by 1 million, nothing actually changes - if the Villain and the Hero are relatively similar, there's no power gauge and the play feels stale from level 0-1000. If you have something to compare to outside - like an NPC that doesn't grow, or scenes where a character shows their power (wiping out a town, carrying a mountain, etc), it's more visceral. \$\endgroup\$
    – Delioth
    Commented Mar 8, 2019 at 16:14
  • \$\begingroup\$ cough Undertale cough \$\endgroup\$
    – Beefster
    Commented Mar 8, 2019 at 18:55

Make it spoiler-proof

If the game becomes popular, any secrets will be spoiled, so don't make it depend on secrets being kept from the player for it to be fun.

Dr. Jekyll, meet Mr. Hyde

Give the main character some trait that allows them to be both superhero and villain at the same time without realizing it, though somehow with two separate but inextricably linked bodies. Gradually reveal this fact with clues (perhaps revealed as a reward for good game-play) so that astute players can figure out early on and have the best possible chance against themselves. At certain points clearly reveal pieces of this info to all players (e.g. in cut scenes) so that no player is blindsided when they reach the final battle.

The greater the struggle, the greater the honor

Reward players who take the most difficult path and win with the greatest honor (points, badges, etc.) so players will not be motivated to slack off in the beginning just to win.

Why play only one side?

In fact you could have the player actually play both hero and villain simultaneously (switching back and forth between the two separate "games" representing the hero's and villain's journey to face one another), and winning the game requires that they win both halves of the game along with the final battle from the hero's perspective. After all, what's a hero without villain? So if the player makes the main character so weak that they can't get the villain to the final battle to face the hero, they lose. And likewise for the hero. They only win if they get both halves (Jekyll and Hyde) to the final battle and then the hero defeats the villain.

P.S. If you want the villain to win as the goal, then just reverse the labels "hero" and "villain" in the description above.


Limit or extend the hero's potential relative to the villain

Make it so you essentially can't truly max your hero unless you first max your villain.

The straightforward but boring way to do this would be to set a level cap at the start of the second half of the game relative to the level you reached as the villain, which would itself be capped at half of the hero's max-max level or less (and the villain would also grow relative to this until the final fight).

The more interesting way would be something like Josh Part suggested, with the villain changing the persistent world in ways that affect how the hero plays, such as unlocking additional areas, shaping the geography of the map, or, indeed, hiding away items for the hero to find later.

IMO, some blend of the two would be interesting.

  • \$\begingroup\$ Great idea. I feel like I can justify the cap with a little creativity. Nice addition. \$\endgroup\$
    – Summer
    Commented Mar 10, 2019 at 23:55

Create a system of counters.

For example if players are going to be mages, create a system where there are 3 schools of magic: fire, water and nature. Fire is powerful against nature, weak against water. Water is powerful against fire, weak against nature. Nature is powerful against water, weak against fire. Then when the player creates a powerful fire mage villain, make the hero a water mage. And so on...

Or when the villain is a stealth-based character, give the hero high perception and revealing spells. When the villain is a high-mobility character, give the hero some items that slow down and/or restrict movement. When the villain is heavily armored, give the hero spells that ignore armor (or even ones where metal armor amplifies the damage).

  • 2
    \$\begingroup\$ I'd like to see a version of this where the hero's character development choices are also in the player's hands, rather than automatically countering the villain. Then the player can choose whether they want to spec into a counter build (recommended path), or deliberately up the challenge for themselves by choosing a build over which their villain has an advantage. \$\endgroup\$
    – DMGregory
    Commented Mar 8, 2019 at 17:18
  • \$\begingroup\$ @DMGregory Oh yeah, you can do that too. Make villain weaknesses and their counters known but let the hero do what they want. \$\endgroup\$
    – user31389
    Commented Mar 8, 2019 at 18:05

Exclusive Choices

There are sources of "power" in the game (be they artifacts, allies, skills or actual supernatural/futuristic powers), but the first character can't get all of them. Getting one locks another out (perhaps it's an issue that the powers actually counter each other or you can't wear two pieces of the same kind of equipment or just there's not enough time to get all of them). Character 2 will have to go get the complementary powers to fight character 1.

Bonus points if getting too many powers causes the first character to go villain so the second character can't get all the complementary powers and has to deal with being less powerful for extra difficulty.


This actually seems like a very neat idea that can lead to some interesting mechanics.

A couple points that comes to mind to answer your question:

  1. Dont' just take into considerations level, abilities, or equipment. The player must eventually fight his own creation, but why stop there? Have events and choices made by the villain affect also the hero's quest, an example as simple as it might get: as the villain you have to choose to either destroy some magic mcguffin or put it somewhere safe; then when you play as the hero it turns out you need said mcguffin: if you destroyed it, the hero needs to find a way to repair it or make another; if you hid it, the hero needs to track down where it was hidden.
  2. Some games give the player the chance to play a second quest once the game is finished, this time controlling the villain. You can do that but having your entire first quest as the villain, and after the game ends, the player can start a new quest controlling the hero, whose mission will be to stop the very villain the player itself created on the first quest.

Both options give you a huge replay value as the choices made during the first act/play affect the rest of the game; just make sure you don't make way too many possibilities as that can become overwhelming for the player.


In Japanese style RPG's, Breath of Fire IV quite successfully pulled off what you are trying to accomplish, by having the main protagonist Ryu and antagonist Fou-Lu being split halves of the same dragon god. You play both of them in different chapters of the game, and get to explore their personalities and reasons for their moral stance. The ending changes depending on how you play the game,

Now as to how to deal with the leveling problem, this game does it very easily: it cheats. Fo-Lou already starts with relatively high levels, and his level and growth only matter in his own chapters. Ryu on the other hand starts with lower levels and his level and growth only matter in his chapters. RAt the final conftrontation, which is played with Ryu, Fou-Lu will be of a preprogrammed level and difficulty degree, that has no relation to the level he had in his chapters. This is justified by leaving some in game time between the final chapters of the two characters, in which Fou Lu supposedly powers himself up off-screen.

While cheating like this may seem unfair, it is an acceptable break from reality and it works remarkably well. Start out with the "dark self" at high levels and give him difficult opponents so the player cannot hold back, then let his counterpart the "light self" start from a low level and struggle even more to beat his "dark self". And if the level of the "dark self" is too low, just give him a few bonus levels at the final confrontation and handwave it with "I was holding back, but now I will show you the true power of darkness".


Unlock more second chapter content based on the first chapter's events

Philipp's (by now acceptated) answer is excellent, but I want to add one more mechanic that should help you provide replay value and balance at the same time.

Unless the twist is completely out of the blue, your hero/villain has likely been sliding down the "ends justify the means" scale, leaving more and more collateral damage.

The stronger and more successful the villain is in the first chapter, the more people he has impacted and the more resistance will grow. These become resources for the second chapter's hero as well as extra story elements to explore.

As an example, suppose there is a "steal the artifact" mission in the first chapter. The player could sneak/bribe their way in and out, gaining only a few XP, or slaughter every last person in the building, take all their loot and gain an extra level.

In the second chapter you could then have either a small "investigate the theft" side-quest providing low resources or a whole new henchman with an "out-for-revenge" line of quests that provides a lot more loot and a level.

This provides some measure of balance. It can also go the other way, with careful play in the first chapter leading to less help and more resistance in the second.

Replayability can also be improved if choices in the first chapter affect the story or play styles available in the second. If the villain rises up the ranks in the Mages' guild, they will be unavailable for help in the second chapter, etc.


Pure Levels are a relatively boring way to advance a character. Bigger numbers. More treadmill.

Use a variant advancement scheme that ties to the narrative.

Instead of just "gain levels", have the villian unlock power sources of some kind. You do have to spend time practicing those power sources ("leveling them up"), but each power source can grant only a certain number of "levels", and is tied to some specific portion of the character's capabilities.

Suppose you have a game with rune magic. Then power sources could be getting new runes, and you'd level up your skill in each rune as you leveled up. The character has a certain number of capabilities; and what that capability does is determined by what the rune-sentence they use and the levels of the runes in question.

Ik Ro Na Tuk might create a whip of fire that entangles foes, then leaves behind a serpent with a special ability to attack them. Your levels in each of those runes would determine accuracy, damage, serpent toughness, and the serpent's special ability.

(These special moves can be hardcoded with variations; then, which rune sentence is assigned which move could be randomized before the start of each game, so learning rune sentences is part of the game's advancement system.)

Once you have done this, the runes that the villian unlocks are the ones that the hero has an opportunity to learn, at least initially, with two exceptions (the "evil" rune for the villian, and the "good" rune for the hero).

Between the first and second part of the game, the villian automatically "levels up" all runes and uses combinations that you have pre-determined. (The villian gets to know every rune sentence, while the hero only gets the ones you have discovered.)

So now, a weak initial half of the game results in a weaker villian, but also a weaker hero. Possibly really careful choice of runes could result in a villian that is using abilities that the computer isn't good at using, but the player is, or that combine really well with "good" but poorly with the "evil" rune, but it isn't a direct "earn a level, boss is a level tougher"; it is "find a rune, both you and your foe have an extra ability to play with".

If the non-boss foes are mostly the same in the second half, doing well in the first half can make the non-boss parts of the second half easier, while at the same time making the boss harder.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .