# How to avoid players getting lost in and/or bored by the meta game?

Yesterday, I noticed something that I had never noticed consciously before, but which had "ruined" more or less all games that I have ever played:

The longer you play a game, the more you get lost in and bored by the meta game. What I mean is that the longer you play, the more you know the meta game. Once you know the meta game by heart, you basically stop playing the game and instead only play the meta game anymore.

Here are a few examples from different genres to show what I mean:

DotA: In the beginning, DotA is super fun, although you totally suck at it. You learn new heros, you are happy when you are able to buy an expensive item and so on. After being "pro" in DotA, you basically only play the meta game anymore: You pick exactly the hero that can counter the enemy and play exactly a perfect meta game for your hero (e.g., go to jungle until you are fat, time creep spawns to stack, and so on). You know basically exactly at which time you will be able to buy which item and you execute this. You no longer play the game, you just play the meta game. You execute a well-thought-out choreography, which you perfected by executing it 100 or more times. Eventually this gets boring!

Anno XXXX: In the beginning, you build an island and have to master more and more challenges to keep you citizens happy. Cool! Once you know the meta game the whole game becomes a "Oh, I need these 10 resources in these proportions, then the people are happy. I build a fleet and capture the enemy or keep on playing until my island is full".

Cities: Skylines (or any Sim City): In the beginning, it is fun to master what your people need. In the end, you know exactly what they need, so it becomes more or less a burden: E.g., for each new residential district, I need a police station, a fire station, garbage incineration plant, crematory, a hospital, I connect everything to the highway, add a subway station and add bus lines. Done. Lame. It starts feeling totally mechanic. I no longer feel like building a real city. I feel like building a mathematically simulation of a city and just responding to optimize all parameters for this simulation.

Civilizations: A super nice game to explore in the beginning! So many technologies, just wow! In the end: I know exactly which technology to develop when. I have a more or less complete ordering among the technologies. I know where exactly to place the next city and which units to build. Heck, I think I could even write an AI that does more than 95% of the things I do, since everything has become so mechanic.

I could go on forever. Almost all games have this. For some players, this is where the game gets really cool. Some players want to play the meta game! But for others, including me, it gets boring. I want to feel like playing a game, like exploring. Not optimizing a numerical simulation.

So, when designing a game, how can you avoid that this will happen? Is it possible to hide the meta game to such an extent that the player no longer feels it, even when playing the game for long.

• Good example of this effect is chess. Only there people like it and play it often multiple times in their lives. They still like it though and it doesn't become boring for some reason. – Trilarion Aug 26 '16 at 10:53
• you've defined the "meta game" to mean "the game, played optimally." this is not the standard meaning of "meta game." – dbliss Aug 27 '16 at 19:58
• If you are always going through the Civ tech tree in the same order, you might be playing on too easy of a difficulty. Beelining universities doesn't help you when your neighbor is poised to invade. Sure there are must-grab techs, but in nearly 1000 logged hours, I still wouldn't say I have a complete ordering among the technologies. – mao47 Aug 29 '16 at 17:26
• Maybe an RNG would help? – user90656 Aug 29 '16 at 20:37
• I feel like all single-player games will suffer from this at some point, because even a team of devs can't build a game faster than a dedicated player can play through it. Even with RNG some players will learn all the permutations faster than devs could ever build new ones. All games will inevitably go this way, although some take a longer time to be reduced to metagame (see: NetHack). You can avoid this with the right PvP interactions however, because you can't metagame other humans. Chess is a good example of this. Actually, I would say mafia/werewolf is another good example. – trevorKirkby Jun 4 '18 at 17:51

I apologize beforehand for the book I have written. I got carried away. I just hope it gives you some good ideas.

What you are describing seems to be a fundamental shift in the way you view and play the game, and I think the clue in making games that avoid it, is in trying to understand the shift, so I'm hoping to explain it in this post.

And that fundamental shift is going from solving problems to executing solutions.

Solving problems

When you start a game, it is completely new. Your understanding of how it works is likely limited to maybe some reviews, a trailer, or perhaps a "what is X?" video.

At this point; everything you see is a problem and everything you try is new. Even the simple question of "I wonder what happens when I click 'new game'?" can have surprising answers. It might jump into a movie, or a character generation screen, a tutorial, or straight into the game. Whichever it is, you're going to have to figure it out as it happens and it is almost certainly going to present new problems to solve. (Such as "How do I get started?" and "How do I win?")

As you become better at a game, you start gathering more and more answers to problems. At some point you know that when starting the game, you need to collect X resource and find Y person and build Z building. But new questions will start coming up; more advanced ones that you could not even have attempted to answer before. "How do I beat the Hard AI?" or "How can I beat this level?" are questions that are only of interest to a good player, but they still put you in problem solving mode; you simply don't know yet how to do it, but you're going to try and come up with some possible answers and try them out, and see if any of them work.

And in doing so, you'll learn some new things about the game. New experiences will be had, and you'll gather more knowledge which you can use to solve even harder problems. (like "How do I beat the Extreme AI?" ;) )

Executing solutions

But at some point, for some parts of the game, problems stop requiring you to solve them. You already know the best solution. What you need to learn or do, is to execute those solution to the best of your abilities.

For example, in a platformer, the first time you find a ravine that you need to jump over, it's a problem you need to solve. Can I jump far enough? Do I need a powerup? Do I need a running jump? These are problems to solve. But when you get to level 10, the "jump this ravine" problem has been solved. You know exactly how far you jump, how much of a running start you need. But this ravine is big. And the landing point is small. And if you want to make it, you need to time your jump just right, or you'll fall in.

This is executing a known solution, which can also be challenging in its own right, but is a completely different type of skill to work on.

In all your examples, you are describing the move from solving problems to executing solutions. When you know exactly which hero to pick, or which tech to research, or how to keep your people happy, all that remains is to perfect the execution. Can you reach a higher score? Can you beat a better opponent? You don't need to solve any complex problems; you mostly need to learn how to build faster or aim better.

But how do we fix that?

So you're asking for design methods to fix the shift from solving to executing. In order to be stuck in solving mode, what you need is a problem with no clear-cut solution. And, ideally, if you want to make this work for competitive games, a problem that can't be perfectly solved at all.

Procedural generation

A big one for single-player games that want to stay fresh for a long time, is procedural generation. The fact that your levels are randomly created means that all of them start with problem solving: "where am I?", "what here can I use?", "what do I need to be on the lookout for?"

Procedural generation will work you keep you in solving mode, until you intuitively figure out the restrictions on the generator and start expecting things. "Okay, the game gives me a two minute grace-period to set up. Also, there has to be a box with some weapon nearby. Let's go."

So, the best way to keep players on their toes, is to make the pattern as broad as possible. But that will play into the second point, also.

Multi-strategy games

Another one that helps a game stay fresh, is to have multiple paths to victory. Civilization has a set of different victory conditions (tech, diplomacy, conquest, etc) and each requires a different way to play. In addition, defending against each also takes a different approach. This means that a game where a strong player is going to win a diplomatic victory will go very different from one where a strong player is going to win through conquest.

This, in turn, means that while you can have a good idea what you will do in your rush to win, you won't know what you're going to defend against, and that will keep games different. At least, assuming that you actively need to prevent other players from winning and there isn't one fixed method to do so, which sadly in Civ is not the case usually.

You can get pretty extreme with this approach, too. In civilization, most races are roughly the same. But for example in Endless Legend, some races get traits like "You cannot be at peace with anyone" or "You can ban other players from trading resources at will", which force you to try even more things.

Multi-strategy games will work until the player figures out a 'certain win' strategy, or until they have tried all possible strategies. (In the latter case; congratulation game designer, you won.)

Limitless games

You can also keep a game fresh by doing away with limitations and letting players go completely wild. This is a fairly new phenomenon that has been made wildly popular by (of course) Minecraft. I'm not sure what the creators thought people would build in the game, but I'm fairly sure that "A working processor" or "All of Minas Tirith" weren't it. But there you have it.

Limitless games will keep you in problem solving mode until you get bored of something else in the game, so really they are the perfect thing you're looking for. Of course, at some point, you are no longer "playing a game", but are just rebuilding a different hobby inside a piece of software that was once a game.

Another issue with these games is often that they lack a clear goal. Goals can drive players towards something and keep them coming back, but not everyone sets their own goals. And it's very hard to set goals in a game that can go anywhere.

Evolving games

Another approach is games that change as you play them more. This type of game on its own seems to be very rare, but you can often see it implemented by continuous developer attention.

For example; collectible card games that have new cards released, MMOs with expansions, DLC in strategy games, balance tweaks, etc. These are always a delicate balance between keeping the metagame moving enough that it keeps people interested but not moving it so much that is makes investing in the game seem pointless.

Games that do this intuitively would be awesome, but it might be future music for now. Game design and balance remains an art for now, and computers are not very good at art.

Combining things

Many of the above options can be combined for greater effect. I'm going to hypothesize a game based on the 4X principle (like Civilization) that tries to go for maximum "problem solving mode". (But it will take huge developer involvement to run it, I think. So it might not actually be a very good game.) We'll make this an MMO game, because those are by far the hardest to do this way, due to all the players getting together to write down solutions.

First up; take a working chassis for a 4X game. Some setting, some tech, buildings, unit types, victory conditions, etc. Since this is an MMO type game, we don't expect players to be logged in constantly and the runtime will be fairly high; let's say one month. When you are absent, your AI advisor will play for you. (There's some games in the market like this)

Now, we create a procedural world. This is common practice for these games, but we'll take it a small step further: while we strive to keep the game balanced, we make no attempt to allow all types of victory conditions and all strategies possible from each starting location. Start without iron? You'll have a hard time fielding a strong military; might want to rethink that conquest victory. Stuck on island? Guess expansion isn't for you. We want the start position to be fair towards "Can this player reasonably win?", not towards "Can this player reasonably play a fixed strategy?"

Then, to make it even harder to have a fixed strategy before game start, we'll vary (but make it known) how hard each type of victory is on this map. Maybe this time around, conquest is made easier, but technology is more expensive. This means that the "best strategy" from the previous game is now useless if it can't bridge the gap created by your completely different starting position and the different relative complexity of the win condition. But if you understand the game well, this just means that you have been given a hard problem to solve; which is exactly what we want.

(Note that it's not automatically true that everyone will aim for the easiest objective. Since we know the relative difficulty of each objective at generation, we can give a bonus to players that are in a region which is naturally geared towards a hard objective for this map. Kinda like how some games have special "2vs1" maps where the 1 player has a defensive advantage.)

Then, let's give the game a mechanic where the players can make broad changes to the rules. Civilization's World Congress had some of those, but they were a bit tame and came in late. I've played a board game where you had rules like "nobody can build more than X warships" or "if you don't protect your colonies with armies, they revert to neutral" and a bunch more, including ones targeted at players. That's more like what we need. This means you can't plan out your strategy from the start; you need to adapt to the rules of the game changing. But still, if you know the game, you'll know roughly what you can expect, so you can still get better, but the available rules for each game will vary. You can learn the game, you can be good at the game, but you can't solve the game, and so you'll never get bogged down in the "Perform X, only slightly faster this time" metagame.

At this point, I think you've got a game that won't get boring for a long time because it cannot evolve a metagame of the sort that DotA or Civilization has. It will change too much to have a fixed metagame, while also changing very little, so that you can learn to become really good at it. It just takes a completely different skill to be good: you need to be adaptive and good at solving problems, not good at executing the best strategies the fastest.

• You should make a game like this. – wizzwizz4 Aug 23 '16 at 15:51
• An example of the last state is Deck Building games, in which the set of cards changes. You may know some good strategies and combinations, but 'solving' the game (i.e. finding the optimal set of cards to select from given your resources for a given set of cards to select from) is a very, very difficult proposition with a very large search space. – Nate Diamond Aug 23 '16 at 20:18
• That's precisely the opposite of what is asked for. Any situation where you google "what pokemon should I use?" and the answer is "since everyone currently uses X, use Y, at least until people stop using X" is a game that is not about problem solving, since you can just google the answer. – Erik Aug 24 '16 at 8:27
• @user1306322 you are of course free to post your own answer, if you have a better one :) – Erik Aug 24 '16 at 18:33
• Nice answer. I feel these two phrases are the key insight; you should make them bold :) - "what you need is a problem with no clear-cut solution ... a problem that can't be perfectly solved" – culix Aug 25 '16 at 5:10

The state you describe as "getting lost in the meta-game" is actually a state where the player has achieved a mastery of the game which makes them play it differently (and arguably much better) than they did in the beginning. But for some player this new way of playing is less interesting than the way they played the game in the beginning. But the beginner way of playing is not rewarding for the player, because it means they will be less successful. The result is frustration and abandonment of the game.

Please keep in mind that not all players react that way. Many in fact enjoy exploring and learning the deeper mechanics of a complex game in detail. The reason why DotA or League of Legends are so successful is precisely because of the long-term motivation which comes from mastering its deep meta-game. But the premise of this question is that we want explicitly to appeal to the players who do not, so let's continue based on the assumption that breaking immersion by understanding the game mechanics is not what our audience wants.

• Keep the game mechanics simple and easy to understand. That way any new players will soon discover if they like them or not and multiplayer will have less of a learning curve, because all players have roughly the same level of understanding of the meta-game. Case Study: Blobby Volley.
• Alternatively, make the game mechanics so complex that nobody really understands them. Try to hide all numbers from the game interface to prevent players from figuring out your formulas. This really only leaves intuition as a decision-making criteria for the players. Case Study: Most modern first person shooters. The more diligent explorers will try to obtain the numbers from your game files and will soon create a wiki where they are all listed in detail. The only counter-measure is to keep them server-sided.
• Don't make a competitive game. When you encourage competition, you put the players in a mindset where they feel they need to do whatever it takes to win. Create a game which focuses on creative expression, socializing and world exploration. Such games are far more rewarding to play when they are played in an immersive manner. Case Study: Minecraft, Starbound.
• I think I was not clear enough on this point: I do like learning to master the meta game! What I don't like is when you are at the point where you basically master it-more or less-perfectly. I liked learning DotA to almost perfection or designing cool Cities. But once I was more or less perfect, there was just nothing more to learn, only the same boring thing. I think this is very many players stop playing. Thus, a solution to my problem seems to be to create a meta game that is so complex that you will never master it. But I am not sure if this is possible. – gexicide Aug 23 '16 at 13:09
• @gexicide: Why do you want to? It seems to me that if what you enjoy in games is solving them, finding a strategy that works optimally, then once it's solved you should just move on to another game. Yes, many players eventually stop playing Civilization, but how long does that usually take? For players that are interested in Civ-style games, they usually go months at least before they've explored all of the various options. Do you really need one game that you can play forever? – Nicol Bolas Aug 23 '16 at 16:33
• Yeah, I also don't understand this argument. The point of a game is, typically, to master it. Once you've mastered the game, you either continue to play (because you enjoy it), or you move on to something else (because it's grown boring or you want a new challenge). This is literally the point of many types of games. It's not a problem, it's a property. – Jesse Williams Aug 24 '16 at 12:03
• "The only counter-measure is to keep them server-sided" -- hopefully at the point where they're seducing one of your ops guys in order that their friend can rapel down to the balcony from the roof, take a plasticine imprint of the key to the server cage, and escape through the garden, your players are having enough fun with the meta-game not to complain. – user44630 Aug 25 '16 at 16:00
• @NicolBolas: I am just wondering if it is possible to create a game that will make players indefinitely interested. I mean, people enjoy sports like soccer or skiing all their life, but a computer game always gets boring at some point. I wonder if this is unavoidable or can be overcome with a good design. – gexicide Aug 25 '16 at 17:38

May I suggest the book "The theory of fun", by Raph Koster?

In essence, he suggests that a game is only fun until you've grokked it and it is human nature to figure out the optimal solution to a problem and move onto the next problem.

(It's called learning, apparently. Once something is learned, the mind seeks further variety. Therefore, I guess you need to change the metagame continuously, keeping it between "Too hard, this is chaos" and "Too easy, this is boring". User-generated content seems to help, people end up making their own problems - and entertainment.)

# Reactiveness

"No plan survives contact with the enemy."

The enemy chose hero X.

So, you have selected hero Y and have started plan Y2, since that is the recommended counter to hero X.

Unfortunately the enemy guessed that, and started executing plan X-counterY2.

But, AHA, you guessed that and segued over into plan Y2-counterXcounterY2.

But what is the enemy doing? That doesn't look like X-counterY2 anymore...

And you lose.

A few games later, the enemy again chooses hero X. And you choose hero Z. Wait, what? Nobody plays Z against X! However, this means that the enemy does not know what to do!

You on the other hand have studied this matchup. You know how the enemy is likely to react, you know what to do against that, you know their counter to your counter to their counter.

And you win, that one game. Unfortunately you were streaming this and now everybody knows what you know.

The next game, you choose hero X. And somebody who saw that stream choose Z. You know you are going to win this, because you know exactly why Z-vs-X is a bad choice.

# How do you design a game like that?

Having a Rock-Paper-Scissors aspect to game helps a lot. It shouldn't be as blatant as RPS, but even small advantages makes sure there is no single "best" answer anymore.

One typical RPS trio is Growth - Defense - Attack. Growth beats Defense. Defense beats Attack, Attack beats Growth.

You can have Armour, that make some weapons ineffective. You can have Armour Penetration, which can be very useful or very useless.

And so on.

The main point is that the strategies must not be locked in from the start, the player must be able to change things around when they see what the enemy is doing. However, your early-game choices must have some consequences or they become meaningless.

Add limited information into the mix and things start getting interesting. Scouting can give you the ability to out-strategize the enemy, but it also costs resources sorely needed elsewhere.

• I'm sorry but didnt you just describe the dota metagame? – Niels Aug 24 '16 at 12:53
• Grass > Water > Fire > Grass. Pokemon is a great example for this. – David Starkey Aug 24 '16 at 14:14
• @Niels: I was mostly inspired by LoL, which was inspired by Dota 1&2. I have never played Dota, but from the OPs description, there is little reactiveness there, when you have decided on a plan, it doesn't matter what the opponent does. That is boring. In LoL you need to pay attention to what the opponent is doing and react to it. Much better. – Stig Hemmer Aug 25 '16 at 7:21

I'm surprised nobody has mentioned this, but chess is a pretty good example of a game where the meta is so hard to master that people have been playing for centuries. The only portion that people maybe have gotten bored of is openings (which is why Bobby Fischer proposed a chess-like game that had randomized starting places).

Chess actually has a small set of rules, so what can be gleaned from its success is that it's not necessarily the number of rules and stats and such that keep a game interesting. Chess remains popular because there's an exponentially large number of situations, and each of these situations may need a unique application of logic.

I think what you're experiencing is games with limited situational variety, at least on some level. "I've seen this, I know what to do here" is maybe uttered a lot, and the mastery comes more from execution. Knowing what to do in a given chess situation, on the other hand, is probably rarely straight-forward.

This probably all plays into Erik's answer about problem solving. In general there's two solutions: have a game rely heavily on strategy that builds up over many moves or actions (like chess) so each application of strategy can produce exponentially large amounts of responses (situations), or rely on randomness to create a large variety of situations (procedural generation, or multiplayer (other humans can be very unpredictable opponents)).

• Go is an even better example, in my opinion. It isn't a coincidence that computers took so much longer to get good at Go versus Chess; computers are far better at "executing solutions" than humans, but far worse at "solving problems" (per Erik's answer). In general, games that are more difficult for computers to solve are likely to have less exploitable metagames. – Malcolm Aug 26 '16 at 22:31
• @Malcom True, I did think about Go as well, but went with Chess because I'm more familiar with it :) (though I'm not great at either) – Bob Aug 26 '16 at 22:36

## TLDR: Give more power to players' actions.

EverQuest Next had some ideas that could alleviate this issue. That game was apparently cancelled (I'm not sure why as I haven't kept up with it), but if plausible, many of their ideas would make player's actions much more important.

## Persistent, destructible environment

If you let your players loose in a destructible world, then the game changes with the players. This area used to be a great place to mine, but it's been cleaned out and no one went there anymore, so now the assassins guild took it as one of their bases of operation.

## Wandering, limited AI

The forest around the town is full of goblins? Send the players to kill them all. You found out the best strategy for killing goblin bosses, hooray! Now there are no goblins, but the orcs and ogres are moving in. Maybe the goblins weren't so bad. Goblin meta-strategies aren't working here, so you're back to square 1.

## PvP incentives

The two player factions went to war over [insert reason] and now they are in shambles, allowing the other factions to take power. These new factions have raised taxes on swords and made significant improvements to bows. the archers are loving it, but there's talk of a knight uprising.

There's tons of potential with this kind of system. Maybe frost mages are rampant and it's making the world too cold and food prices skyrocket. Lavamancers aren't worrying about accuracy, now there are lava pools all over the map.

Granted, most of these examples work best in the MMO genre, but you could likely extend to other genres. For example, have FPS which has destructible maps that don't refresh for the next game (That sniper nest got destroyed!? There used to be a wall right here!).

• None of these avert the problem the OP observed. You'll form metagames like "mine out places that will have value even without their ores", "don't kill all the goblins" and "cooperate against common enemies" respectively. – user253751 Aug 25 '16 at 11:53
• @immibis The problem there is that every player would need to know the metagame and cooperate, which will never happen. Even if you try to leave some goblins, another player that gets a quest to kill them could finish them off. It also depends on each player having the same playstyle. The players that like to mine will mine, and once it's gone they'll abandon it. – David Starkey Aug 25 '16 at 14:59
• I like it, because the players introduce a source of entropy. There's (more or less) no way to predict what they'll do next, or what effects that will create, so everyone has to adapt to changes as they happen. – Joanna Marietti Aug 27 '16 at 8:00

There are a bunch of great answer so I'm not sure I'm actually adding useful information here, but one thing that works in almost every genre of games is to make up a bunch of modifiers to your game and pick one or more random modifiers each time. Examples:

• If it's a game where economy comes into play (city builders, RTS games) a simple modifier could be a multiplier to resources (gold x1.2 but everything else x0.9)
• A game featuring any kind of physics (platformers, simulators) can modify its physics (altered gravity strength/direction)
• A game with enemies (shooters, action games, RPGs) can modify them (enemy health, enemy damage, amount of enemies, enemy positions)

For literally any game you could think up a few of these and the great thing is that it's not very hard to implement in your game. If you list, say, 20 of these for your game and re-randomise them when it makes sense (every session, every level, after every x minutes) you are constantly changing the way a player has to play the game. You are forcing them to stay cautious and not get overconfident and with the amount of possible combinations it will take way more time for a player to learn all the possible strategies.

• I'm 100% in favor of this plan, but I'd say to be very, very careful with modifying physics - there's a fine line between "interesting variation" and "pointless obstacle to mastery", and having gravity randomly vary by 10-25% seems to me like the latter. A specific possibility, such as 1/2 gravity as a relatively rare modifier, seems alright, but at that point you've only added a single bit of entropy to your possible game state. A particularly interesting bit, admittedly. – Tin Wizard Aug 25 '16 at 20:55
• @Amadeus9 Yes, if you go this route you definitely have to balance it out a lot and tweak the values so it doesn't become too random. – Kevin Aug 26 '16 at 6:28
• What if it's a game where you fly around and explore various alien planets? Then it would make perfect sense for gravity to vary each time. Sounds cool to me - just another variable to keep it interesting. – Joanna Marietti Aug 27 '16 at 8:05

Note: I write this answer from the perspective of a video game player.

If you don't want to create a game where the meta-game plays as much as a role, you have a few options:

• Create a game that makes no sense. Any choice the player makes is by definition meaningless, and can be both good and bad. Basically, truly eliminating the meta-game completely is eliminating the difference between a good and a bad choice. A rand() simulator. A game that is overly complex in such a way that you can't master anything could as well be a rand() simulator, because every choice you make is more or less random; the choice makes no sense, and the outcome makes no sense either. (I personally think that is not a fun game to play. I list this option, because in my mind this is the only way you can completely eliminate the "meta-game", but I think you should not create such a game, because in return it removes any goal the player can have.)
• Create a game that is intended to be only played once. Look at a game like The Talos Principle, an atmospheric puzzle game. Everything in the game can be completed in the first playthrough and right until the end of the game you can revisit any areas if you deem it necessary. A few silly people will replay the ending for a few more achievements. There is little reason to play the game again, but that does not detract from the game itself.
• Create a game that randomizes certain attributes. In XCOM: Enemy Within and XCOM 2 (incl. Long War Toolbox) you can enable options that randomizes level up stats of soldiers, which allows for additional choices during the game; and allows players to play to the strengths and weaknesses of otherwise equal characters. Similarly, Civilization: Beyond Earth randomizes the tech tree if I remember correctly. It forces the player to make some choices in the tech tree that are different from earlier playthroughs, giving the player an edge in something different. I have not played C:BE personally, so I can't tell you how well that concept is executed in that game.
• Create many goals. A great example is the achievement list of Civilization V. Each faction has a specific goal that counts for an achievement. Some of these goals require you to play in a very specific way, probably differently than you are used to. This gives the game more longevity. The achievement list of Team Fortress 2 contains many goals to play the game in a non-standard way.

You might need to add a way to let the player know that such options exist. For example, given your description of the Civilization series, you likely did not know of these alternate goals.

• The first point is horrible advise, the second goes directly against the stated goal of creating a game with long-term motivation. – Philipp Aug 23 '16 at 19:14
• OP never mentioned playing the same game for multiple playthroughs as the design choice they want to make. They just want a player to play the game "for long". They gave several examples of games with a repeated match element, but personally I played The Talos Principle for 43 hours. That's a significant amount of time to spend on a single game. As for the first bullet point: That's kind of the point. A game is meaningless if for all choices, there is no way to determine which option you want to pick. You want to pick the "better" option, because you want to finish the game. – Sumurai8 Aug 23 '16 at 19:35
• I imagine RandomSimulator2017 will be on Steam soon. Don't forget to purchase the XKCD pack! – David Starkey Aug 23 '16 at 20:30
• @DavidStarkey And Dilbert DLC – Tobias Kienzler Aug 24 '16 at 9:41
• The Witness is another example of the second (and I've heard it's similar to The Talos Principle although I haven't played that). There's no point playing it more than once because the second time you'll know all the solutions already, but you can take as long as you want to do everything on your first playthrough. – user253751 Aug 25 '16 at 11:56

And that fundamental shift is going from solving problems to executing solutions.

How can you delay that shift? One way is to make it unclear which solution must be executed by introducing variety from game to game.

For instance, consider Master of Orion 1. Its core game mechanic is designing ships superior to your opposition, but both you and your opponents can only use the technologies you have acquired, and technology acquisition is different for every game. Specifically, there are 4 ways to acquire a technology:

• research: Each player can only research a random subset of available technologies, so the desired tech may be unavailable ...
• trade with another empire: The empire must have the tech and be friendly, and want some other tech you have ...
• espionage: The empire must have it, and you must be willing to piss them off (or so superior in computer tech that you can reliably frame another empire)
• conquest of a developed world: The empire must have it, needs temporary space and ground forces superiority, must be willing to piss them off, as well as their allies, but boost relations with their enemies.

... or the best solution may be to do without the tech ;-)

Note how all of these approaches are highly situational? In MOO1, there is no approach that's always best, even though there is always a best approach ;-)

Variety can be introduced in various ways:

• random situation (abilities, obstacles, goals, ...)
• dependency on unpredictable choices of other players (absence of a nash equilibrium, opaque or evolving metagame, ...)

There are many different players with many types of tastes, but I think I'm fairly close to your tastes, so I'll offer with what keeps my interest. I think it applies to many other gamers with similar tastes.

I noticed this sort of thing decades ago, and for me it was pretty much the same thing. At some point, I stop relating to the game as the situation it is supposedly about, and start relating to it as its artificial game mechanics.

After decades of game playing and designing, I'm convinced that for myself, the main thing that I find interesting in games, and that holds my interest, is when a game's mechanics do a satisfying job of modelling the situation, and when the situation is complex and dynamic and involves many systems. As long as the mechanics are fairly close to how I understand the situation, and the situation involves various layers and changes so that a large situation has lots of logical cause and effect at various levels, I can continue to relate to the game as being about the situation it says it's about, and not about gamey abstractions that I don't care about.

Some of the systems that tend to help a lot, too, are:

• multiple agents with different goals that are consistent with what the game is about, and that are often not just simple oppositions
• incomplete knowledge by the player and other agents about the situation
• persistent cause and effect, so actions and choices have effects that last

Games that model real interesting situations well can approach (and sustain) the interestingness of their subject matter, at least for players like myself. Abstract games tend to simplify things to the point that they can be formulaic, however, and feature more certainty and control that a real person in the situation would have. That removes elements from play and reduces the situation to something formulaic and not a real interesting series of choices.

The opposite is also very true for me. As soon as I can see the gamey mechanics and don't think they represent the situation well, I lose interest quickly.

Mod support helps too. If I can mod the parameters of a game to have it match the way I'd like to be, that can save it from me rejecting it. Because for example it will bother me if you can destroy a buttoned up tank by shooting it enough times with bullets, but if I can mod it so you need to penetrate the armor, then I can stop groaning at the non-tank-like mechanics, and relate to it as a tank whose armor needs to actually be penetrated to hurt it.

It has also inspired me to release some mods that extend the interest for other players, too.

Other games I've played for years which held a following tended to also be ones which the designer supported for years and listened to the intelligent suggestions of players (and at most provided mod support for the goofy suggestions of players).

I'm don't know a lot but what you say can be "solved" by adding new content regularly that can change/improve the meta-game. Making new content always helps players learn new stuff and don't feel so bored. Take WoW for example, if they stopped at Burning Crusade the game would be DED. but because they add new expansions every X time they keep their subscriber base pretty much constant.

New content is good, and more if it's free!!!!

• A way to ensure a steady supply of new content and mechanics without having to pay for its development is to support and encourage a modding community. – Philipp Aug 23 '16 at 14:17
• And making your game F2P with micro-transactions, that will encourage the developers the most – UzendayoNE Aug 23 '16 at 15:08
• @Philipp this comment would make a fantastic answer of its own. Games like Starbound and the original Doom have very active modding communities that continually keep the game fresh. Tired of the meta? Install a mod. Not a guaranteed solution from the dev's perspective since a modding community actually has to spring up for it to work, but if it does, you're all set. – Xaser Aug 24 '16 at 19:20

What about a learning AI? It would be like your nega-companion, learning about you after each interaction and figuring out a way to beat you accordingly.

With the current processing power and memory available for calculation, it could be a hard time to understand it and to beat it! (especially because it knows the rules perfectly).

On top of that you add a procedurally generated world, random starter boosts for each (you and the AI) and it should do!

The game becomes more intuitive and less rational, which increase the "meta".

• Sure, let me just type in makeALearningAI(); into Visual Studio... – user253751 Aug 25 '16 at 11:57
• Machine learning is not sci-fi (en.wikipedia.org/wiki/Machine_learning) and for my own experience, I once played a version of C&C where I fought an AI (mod) which where able to learn some of my moves, this greatly increase the game value. By the way, your Visual Studio command did not work for me :/ – bastienb1991 Aug 25 '16 at 14:35
• I think @immibis point is that AI isn't really a silver bullet here. It's tool in the tool box, and one whose power and shininess has almost no limit, but you can't make a bad game good by just "adding AI" to it unless the main thing lacking from it was poor AI. – Steven Lu Aug 29 '16 at 22:25
• Thank's Steven Lu for your answer. To my mind, we were already talking about a game worth playing (at a point you have learnt all the mechanics and cannot enjoy the meta anymore). And from now you think about a way to keep the meta interesting. Maybe by having an unexpected AI behaviour you force yourself into utilizing some ressources of the game you did not before, because it was not of strategic value, not part of the automation that ruined the meta. And as you get yourslef closer to automation, the AI is hunting you to make you stay away from those automation. – bastienb1991 Sep 1 '16 at 11:26

The problem you are addressing is, as you have so cogently pointed out, not one unique to any game, but rather is found in all games. This is, in and of itself, suggestive of the true answer.

Each game mentioned has some sort of hard limit, beyond which one simply cannot go. These seem to fall into two main categories: execution limitations (limitations to the hardware, software, or programmer's imagination/budget/deadlines), and story limitations (goals, playable content, story line and plot.)

Story limitations are limited only by the imagination, which again is suggestive of the true answer.

The real problem here is a people problem, whether of the one playing the game or the one designing the game. The answer is not necessarily to seek new horizons (another game), but rather to have new eyes. The goals and expectations each person sets when designing or playing a game are the actual problem, AND the solution. It remains only for one to become aware of what their current rules and limits are, and to perceive how they can continue to grow.

The only game I'm aware of that truly has no limits is expanding who and what one knows.

A Few Notes Before You Begin

• I generally play story-based RPGs, so most of the games I will cite are going to be story-based RPGs.
• I plan on talking about the ideas of games themselves, then moving on to concrete examples to explain the points (usually discussing games that I think did well in certain aspects and games I think did poorly in others), and then end with a general analysis and answer to the question.

## Theory

As far as I can tell, the six defining features of a task in a game are completion time, complexity, difficulty, frequency, interdependence, and reward. These six features are deeply interconnected, and having any of them out of proportion with will make a task monotonous or frustrating.

Completion time is how long it takes to complete the task. Generally, with all other variables held constant, the shorter the completion time, the less tedious the task (i.e. you press a button to fire a gun vs. collecting all the flags in Assassin's Creed). Just because a task is time consuming does not mean that it is necessarily boring.

Complexity is how much effort you have to put into something to figure out what you need to do. Figuring out how to open a door with a button next to it that says "Press this button to open the door." is a simple task. On the other hand, this is not. Complexity is probably most associated with the idea of a meta game. A task can only be as time consuming as it is complex. Firing a gun is a simple task, and it has a correspondingly low time consumption. Collecting all the flags in Assassin's Creed is a simple task in that all you have to do is find them all, but it takes hours. The first task is not really tedious, but the second one definitely is.

Difficulty is how well a player is able to achieve his goal. Killing a Radroach is one of the easiest things you can possibly do. If I want to kill a Radroach, I can kill a Radroach. On the other hand, a Mirelurk Queen will destroy you. You will need your most powerful guns, dodge all her attacks, and make every shot count. There are other forms of difficulty, too. For instance, I would consider a game in which you flip a coin to determine if you win or lose to be a sort of difficult game, as you can only win about half the time. As a really quick side note, a perfect player should always be able to complete a task, so he should never be randomly punished.

Frequency is a measure of how often you need to do something to achieve a goal. A very frequent task is using a weapon. Whenever you want to kill an enemy, you need to attack him. Other examples of frequent tasks include going to an in-game store or fighting a grunt. A much rarer task would be trying (and failing) to kill Sephiroth in Kingdom Hearts. This task is completely different from killing any other enemy in the game, as you have to find ways to counter his various attacks, find his weaknesses, etc. A frequent task gets annoying when it is both time-consuming and simple. Easiness makes the task mundane and tedious, hardness makes it frustrating. The interdependence doesn't really matter with this one as it is already considered in determining the frequency of a task. A constant reward has a diminishing effect on how fun a frequent task is, but an increasingly good reward can help to counter it. I cannot remember the exact name of this principle, but I think it has something to do with Diminishing Returns. Frequency and difficulty should have some sort of an inverse relationship.

Interdependence is what other tasks depend on that task. Killing enemies and hitting targets both require firing a gun or using a ranged weapon. Although firing a gun is a basic task, using it to achieve other goals removes any mundanity of the task. This really only applies to extremely basic tasks that I will call Fundamental Tasks that are necessary to play the game.

Reward (Finally on the last definition.) is what you get for doing a task. Rewards must be at least proportional to all the other variables of the task (i.e. getting a pretty good sword for a decently hard quest is a good reward, but you certainly wouldn't mind if you got a legendary UltraMegaDeathbringer of Death). An important note is that reward doesn't include the actual fun of doing the task. It is literally just the in-game reward. The actual fun you get from getting the reward is considered when determining how fun the task is.

Now, let's consider a bunch of different tasks. The first will be a Fundamental Task. This task must be quick, simple, easy, frequent, independent, and provide a minor reward. It is literally just pressing a button on the controller or a task like that. Moving, jumping, looking around, using weapons, reloading, etc. would be fundamental tasks. You can't really change any of the features of the fundamental tasks without moving into extreme absurdity (Dibs on the band name.) and mundanity. No one really minds fundamental tasks because they form the interface between the player and the game.

The next task will be defeating an enemy. The features of the task "Defeating an Enemy" vary depending on something like the level and the type of enemy. A level twenty grunt should take more time than, be harder than, and provide a better reward than a level three grunt. On interdependence, the level twenty grunt should be more common in difficult tasks than a level three grunt. The level twenty grunt may not necessarily provide a better reward (Assassin's Creed, for instance), but they generally do. Because these enemies are the same type, the complexity of fighting them shouldn't really change all that much. Enemies of different types should have different complexities or be as complex, but in different ways. Frequency is game-dependent. In this task, all the features are proportional. Now, let's consider what happens when we change these features. A higher level enemy that is easier and can be killed quicker than a lower level enemy in the exact same circumstances is just stupid. A lower level enemy that gives out better rewards than a higher level enemy leads the player to fight lower level enemies, which leads to a whole bunch of problems. A harder task must have harder enemies or else it is not really harder (excluding things such as puzzles). When you get really good at defeating enemies, this becomes like a fundamental task. You want to kill an enemy? Aim and shoot.

Note: This next section has a little rant about the combat systems of Fallout 3 vs. Skyrim. I love both games, but I had a problem with the combat system in Skyrim that I think Fallout 3 dealt with well. I'm not trying to start a flame war.

This was my main and only problem with Skyrim. Other than the combat, I loved it, but the fighting in Skyrim for most enemies was just mashing the attack button, or summoning someone to fight my battles, or using a spell, or chugging 30 health potions. The stealth was fun when it started, but once I figured it out, it was easy. Just walk up to someone and kill them or shoot them from far away. On the other hand, Fallout 3 had a great combat system. You had to run away and find cover. You had to get good at aiming, you had to use a bunch of different guns because yours just ran out of ammo. You had to use your VATS well. The closest thing Skyrim had to this were the bows, but you only had a bow. You did not have a bow with rapid fire, you did not have a bow with a scope, you did not have a bow that did energy damage, you did not have a long range bow or a short range bow, etc. Granted, you did have bows with different enchantments, but even they got old. The difference here is that once you learn how to use one bow, you learn how to use any bow, but learning how to use a shotgun is completely different from learning how to use a sniper rifle or a magnum or a hunting rifle. The same general idea goes with spells, one handed weapons, shields, and two handed weapons.

tl;dr: The weapons in Skyrim are too similar. Once you figure out a general class of weapon, you figure out them all.

Let us consider a more complex task: the infamous fetch quest. Fetch quests tend to take a pretty decent amount of time to complete, are very simple (this will be discussed later), and appear very frequently. You have to go find some dungeon, kill all the enemies, get the item, and come back. They are often boring and used to pad the length of a game. They are often too simple for the amount of time they require and they appear very frequently. The simplicity combined with the frequency make mastering fetch quests extremely easy. You either have to remove them or make them more complex. Fallout 3 did an amazing job of making fetch quests more complex. The quest line Scientific Pursuits combined with Tranquility Lane in which you have to "fetch" your dad and a story-related invention ignored all the clichés of fetch quests. Instead of just going into some vault and killing its inhabitants, you end up trying to escape a simulation of a neighborhood controlled by an insane scientist who loves torturing the inhabitants for fun. You had to either help the sadist, which led to complex problems that you would need to solve or find a failsafe, which had a complex problem in itself to solve.

Finally, I will talk about sandbox games like Minecraft. In Minecraft, a lot of the gameplay is extremely simple. Go to a crafting table to build things. Use a pickaxe to mine, a shovel to dig, an axe to chop wood, a sword or bow to defend yourself, and a hoe to plant. Eat food to not die. Use a bed to sleep at night. Attach blocks to other blocks to build objects. Etc. These are fundamental tasks to playing the game. The fun in Minecraft comes from building huge or amazing structures or machines. The players themselves come up with the challenge. They make their own quests. The amount of fun a player has is determined entirely by the player. All the quests give the exact reward the player wants, and have the difficulty, frequency, completion time, and complexity that the player wants. This allows for a great balance between all the features of a task.

The basic idea of how to keep players interested in the meta game depends almost entirely on increasing the complexity of the meta game, and then changing the other features to coincide with the complexity. Instead of having just one meta game, create several meta games inside a meta game. You want to beat an enemy? To do that, you're going to have to learn to master ten different guns. You want to go on a quest? To do that, you're going to have to learn how to defeat seven different types of enemies. Instead of letting the player control troops on a flat battlefield, give him some different terrain. Put a mountain pass on his left and a river on his right. Give him multiple troops with different advantages and disadvantages. Give him some Calvary.

Another option is to somehow constrain the player to force him to adopt new tactics without making it unnecessarily difficult. Don't take away all his weapons, just make his main one run out of ammo. Now he has to use a weapon he isn't nearly as good with, which leads to him having to think his way out of the situation. This is a big part of why I loved Bioshock Infinite. It forced me to think about whether I should keep my empty carbine and hope I find ammo later or pick up another gun and hope I find another carbine later.

I hope this helps spawn some discussion or give you some new ideas. Feel free to add any info or other ideas.

As a player, I've found myself in the exactly position of boredom, after "solving" the game.

From my perspective I find that the more competitive a game is, the more the players tend to copy other player's effective strategy, where available. This degrades to an infinite loop of 'try-work-faster' to sharp skills and precision in RTS, MOBA and FPS.

The less challenging a game is, the more boring will be; so how do you find the perfect balance?

My answer was to play ArmA with a co-op oriented mil-sim clan (not all-realism) to always shape this Sandbox game with a different approach to a different problem with no advice and based on the improvised decision and skills of the players combined.

• Can you please add some concrete advise to the answer which is applicable to game developers? – Philipp Aug 24 '16 at 14:41
1. Keep the meta advantage low % and pregame choices of limited importance. You don't want rock-paper-sciccors with a 15-30 minute delayed, "rock beats scissors" screen.

2. To accomplish the above, characters need a well rounded kit, that is standalone and does not rely on roles that are stiff, or a lot of hardcounters. Every character should have 2-3 harcounters, but keep the rest soft.

3. Another way is punishing stupid wins by quick games. Your players can create more fun games for themselves by choosing games. If a player gets dominated so hard in the first 5 minutes of the game, because 1 hero is totally broken and the enemy chose him, then the player should be able to leave. There is no point in playing a game where 1 team will win in 10 minutes after the result is already apparent. Your enemy looses or knows he will loose and should not be forced to play.
Suddenly short wins and breaking the game is less fun. People play for fun. Someone winning 12 games in 1 hour with the same hero has less fun, than someone winning 2 games in 1 hour, with well earned victories and close games that he could potentially have lost. (This also makes cheating less apealing, remember as a kid, you break the rules and noone wants to play with you ? Yeah, basically that, you cheat, you get 5 minutes or less wins and don't have fun.)
This is going to make people choose less broken strategies, because while fun once in a while, it get's stale to win quickly and be able to "earn it". A quick victory because of a broken champion isn't going to feel nice, especially if the matchmaking takes longer than the actual game.

4. Limit resources. Force players to compete for resources with their own team and also the enemy. If you have a game with (nearly) unlimited resources then the game becomes trivial, it's just get the most resources as fast as possible, the character able to gain them most efficiently and beeing able to use them most efficiently will become meta. The character list becomes ordered by those factors and suddenly you have a clear cut tierlist, with little dynamic in it.

5. Give the ability to deny resources and produce stalling in progress for both sides. While this sounds boring at first, stalling is important to some degree. Those stalls should not be easy to accomplish and not last forever, but they act as equalizers.
If both sides don't gain any resouces during a period and need to battle for them or show of their skill to gain them, this will stop snowballing. If 1 player got a small/medium advantage and can just progress to the next goal without needing to risk anything, then this is going to be the easy strat. Gain advantage, defend it pretty easily, cause defense is always easier. Wait for upgrade, overpower enemy or deny him resources more easily. Repeat.
With stalling, the advantage has to be actively defended. If I don't get closer to my upgrade because both me and my enemy deny each other resource gain, then we have to fight it out tactically, my enemy even though slightly behind has a valid strategy before I snowball. I can't snowball by just defending I need to actually get 2-3 kills to gain an item, or outskill him in cs in dota, to be able to buy an item in the early/mid game.

6. Don't make it easy. Don't give the defender a too big advantage. But the give the attacker checkpoints that put up a fight if they are attacked ahead of time. In dota every hit counts, playing defensively will barely regen my hp if I ran out of regen items, which cost gold. The tower makes it hard to get cs, loose hp. Still the tower puts up a fight and shoots ahead of time if I directly click an enemy. My enemy can comfortably stay in range of his tower at the start with medicre/full health, but can't really rely on staying there forever, cause his cs is hurting.

7. Realize that if you skill ceiling is below the maximum human capabilities, then everyone is basically at the highest level in skill in the higher ranks, the only thing to be abused then to gain rank is the meta game/strategy which can be complex to figure out perfectly without using complex math. You can't do anything about that, beside try to balance it so no hero hard counters more than 3 other heroes and that most things are softcounters. This keeps the game dynamic, the meta becomes less tight and you have more variety, even at the higher level.

8. Balance, real balance. This means that anything that doesn't softcounter each other should have a 50% winrate against each other.
Softcounters shouldn't be auto-wins and not exceed 60% winrate. They should still have a hard game and some trouble with their enemies.
Hardcounters should be able to be adressed by lane swapping, so keep hardcoutners few.

9. Everything needs a downside. Anything that is just plainly good is broken. Things good in the early game shouldn't scale. High burst dmg should have high mana cost/cooldown. Put stuff into conditions in general. Keep kits unique and have a large ability space. No hero should be comfortable just doing the same thing without beeing killed in a game, the enemy team should always have counter plays.

10. Don't go overboard with variance in mobility. Blinks and teleports, slow and high movement speed should be rare. They should have a very very good design reason. If there is to much variance it's impossible to balance in the long term.

11. Give the player tradeoff upgrades (items), which can be used to counter 1 thing, but give up/weaken another thing. Spending resources on it usually is not a tradeoff (unless it's cost inefficient, but it should not be an auto include for a dominating player).

12. Limit lategame potential to a constant. If growth in a game is unlimited snowballing will never stop, so a dominating player is never going to loose his advantage.

• 3. is not true. There are many players who just want quick, dirty, easy wins and will have them. More then there are players willing to put up with a broken game. Remember; griefers and trolls are a thing. They will abuse the hell out of it and enjoy every minute. – Erik Aug 23 '16 at 21:58
• Yes, true, but if they are very shortly lifed, especially compared to the match making time then you spend most time waiting or doing nothing. It obviously depends on the reward provided. If there is an incentive in ingame money for quick wins, then it encounters the hearthstone problem where quick easy decks are more popular, because they are quicker. – HopefullyHelpful Aug 23 '16 at 22:43
• On the otherhand such feedback can valuable for devs. If somethink is unfun and broken to play against then it's much more urgent for devs to fix than if ppl. just hate the game and surrender after the minimum time, but have no mechanism of punishing lame and boring plays. – HopefullyHelpful Aug 23 '16 at 22:46
• @HopefullyHelpful I'm still not sold on #3. Those types of players would probably love to maximize their reach by getting new victims faster. I'm not a psychologist, but I imagine getting reactions from others is part of their fun. – David Starkey Aug 24 '16 at 13:27
• This answer is just a collection of tips for how to balance a MOBA game. Most of it is not bad advise, but it isn't addressing the question which was asked here. So I unfortunately had to downvote it. – Philipp Aug 24 '16 at 14:58