We Don’t Know What We Don’t Know

The Power Of Research

When you are working on a game or a new feature there will always be unknowns. You can’t be sure how things will work out and how your game will look like at the end. Often the first decisions you make will be the most important because they will give the biggest impact on the direction of your game or feature. So these decisions shouldn’t be too arbitrary. But how can we reduce the arbitrary and improve the basis of your decisions before designing and prototyping?

The answer lays in research.

The following article will try to show why and how you should do game design related research.

Why we should do research

Deleting unknowns as soon as possible

The most important reason why we should do research is because we often don’t know what we don’t know. A lot of people think that they already know a lot about a specific topic when in fact they don’t. We can never be sure if we know enough about a specific topic, but we can raise the chances that we have all important information to make the right decisions by doing good research.

When you do research chances are high that you’ll find out that other companies may did something that you thought would be unique in your game but in fact has already been done by other companies. Now you have a lot more opportunities and information than you did before.

Raise innovation

Research can also be a huge innovation driver. You can jump right to the status quo of the latest games and improve on what is already there. Don’t miss this opportunity, otherwise your may good and innovative game isn't that good and innovative in comparison to games that are already out there and may did better than you. But be warned: Do not copy a feature just because you think it is fun in another game, games are complex systems with a lot of dynamics, so you can’t expect that a copied feature will work in your game as it did in another. Adapt a feature for your specific needs and try to improve it so your players will have the best possible experience.

Keep conventions

Another good reason to do research is due to the fact that a lot of players are used to specific thinks. When you don’t know about specific conventions you may design something that annoys a lot of players because they are constantly confused about the way how things work differently in your game. That doesn’t mean that you are not allowed to break conventions in your game design but you should have better reasons than your lack of knowledge.

More authentic experiences

Good research can give your game a more authentic experience. A lot of information that humans know about specific things are hidden deep in their  subconscious. When you have elements in your game that are related to real existing things that look and behave authentic,  your experience becomes more credible and the immersion will be increased. Players can feel if certain things are just wrong, do not try to fool them just because you didn’t know better. Again, of course you can drop authenticity whenever your desired experience requires it but not because you didn’t know better.

Building confidence and trust

The last good reason in my opinion for research is confidence and trust. The other developers should be able to trust you as a designer. So you should try to know as much as possible about the specific things that are related to your game. A good research can build the confidence you need for a clear and strong communication that is based on knowledge and not personal taste and assumptions. You will never know anything about all games that were made and are related to your game and that’s alright, but the more you know the better it is. A well prepared designer is much more respected than a game designer that makes decisions that seem to have no real foundation. It builds trust and shows that you really care.

When to research

Now that we know that research is important we have to ask ourselves of how we can increase the value that we can get out of our research and make it more efficient. So the next part will try to point out when to research to make it as efficient as possible.

Research soon enough

This may sound like a no brainer but I have seen this way too often to not mention it. Research as long as you can react on the new information. There is no use in research when there is no more time to make use of it. So do it in the beginning when you approach your design problems. For example when you have to design a new feature, make changes on an existing feature or work on a vision for a game.

Research as often as necessary

Things change. You can’t expect that all the information that were legit a few years ago are still work the same way. The market changes, the technology, the users, the platform, almost everything can change. So do your homework whenever you run in danger that your information could maybe outdated. If it’s not the better you just saved a lot of time and maybe refreshed your memories on a specific topic. Well done!

How to research

The next part is on how to research, there are different ways to help on making research more efficient.

Make it a task with a result

Research is something that has a high value in production, so your production plan should handle research as a task that has to be performed and should lead to specific results. Arrange enough time and ask for a result that can be shared with the other designers so you can increase the value of the information you researched. Make a presentation and discuss the results. Maybe some information is interpreted by another designer in a different way than you did. Discuss the results and take appropriate actions on how to involve them in your design in the best way possible.

Playing good games is useful...

Some games can work as a prototype for you. Play the top and most successful games that are related to your game. Try to find out why they are so successful. Check the forum of these games to see the player feedback on specific features. This will not necessary tell you if the feature was successful but it can lead to valuable insights.

Playing bad games too.

Take a look at games that are related to your game and failed. Why did they fail? What kind of risks can you recognise from their failure and do you run in danger to make comparable mistakes? Avoid theír mistakes and find out what you have to do to make it better.

Read Wikis

A lot successful games have some kind of wikipedia. This can save a lot of time when you want to find out about specific parts of the game that would take too much time to get to by playing it by yourself. I’m pretty sure that you will find valuable information there.

Read Blogs

It may sounds crazy but some guys write articles about game design or write analysis about specific games. They work for you for free. Great. If these guys are good it saves a lot of time. They did research and even analysed the information for you. Say thank you and take consequences.

Reflect your information

This is the by far most important aspect. Think about the information you gained.  Even if 10 other guys wrote an analysis with 100 pages about a specific game or game design topic, they could be wrong. So don’t be a lazy designer and think about it. That’s what you get paid for. Do it. But do not question the things I wrote. Don’t do that. Just kidding, do it too. Ask yourself how did a feature work out in the other games? Will your feature be exactly the same or can you think about improvements to the solutions the other games already used. Even when you know that other games did something that you also want to have in your game it may be a good idea to take a closer look to their implementation. You may will realise problems that you didn't think of before and have to make more drastical changes. There can be a lot of dynamics in video games. Research is all about erasing unknowns with the least possible effort. So use your brain and try to gain as much value out of your research. The information may save you some prototypes that would have failed now that you know how a comparable feature worked in another game. Great isn’t it? So use the gained time and go for the prototype that is closer to your final solution and improve on that.

Conclusion

Research is work, has value and can be done well or poorly. It can save you a lot of time and the information should be shared and conclusions should be drawn. Plan time in your production and reflect on the gained information. Start by sharing this article and reflect about it.

Let’s Raise the Stakes

Did you ever wonder why some games are so thrilling that they exhaust their players on a physical level? Today I want to talk about what components in your game can create such such intense experiences.

The Ingredients

When creating exciting moments it’s all about the player’s expectation of specific outcomes in certain situations of your game. The outcomes do not have to be the final result of the game to be exciting. Excitement can also emerge around every gameplay loop that your player faces in the game. The situation doesn’t have to mark the end of the game it can also be something like a boss fight, an encounter with a hostile player during a Counter Strike match or a randomized reward that the player is going to get. The possible situations can vary a lot but it will always need at least two of the following three ingredients.

1. Uncertainty (This one is always needed)
2. Investment
3. Rewards

Uncertainty

This ingredient is a must-have if you want to create exciting experiences. Without uncertainty, there is no excitement. Your players need to lack a certain amount of information that determines the outcome of the specific situation in order to be excited. Let’s take Tic Tac Toe as a counter example. This game is solved / mastered by most adult players and as soon as a game is mastered the result will always be the same (a draw). This game does not have any uncertainty because adult players know how to play in an optimal way. As long as both players play with the desire to win, the game will most likely end up in a draw. Uncertainty can emerge through hidden information, chances or situations that are so complex that the human mind can’t process all possible outcomes in the given time.

Hidden Information

One tool to provide uncertainty is to hide certain but relevant information from the player. The fog of war in strategy games is one of the best known tool where hidden information comes into play. Another famous example is Poker, where players can’t see the opponent's cards and of course in a lot of shooters the level environment is a huge factor when it comes to hidden information. Hiding information from the players is a well established tool but great games tend to give players tools to reveal hidden information. An important factor when it comes to hidden information is the amount of information that is hidden or revealed to the player. Too much hidden information gives the player too little to work with when it comes to making his decisions and they become arbitrary. You need to find the right spot between hidden and known information or at least give the player tools to reveal hidden information. This adds a strategic layer where players can influence the game’s state to get information and make more educated decisions.

Chances

Chances come in two kind of forms.

Random Chances: When there are almost no relevant information the player can use to make his decision he has to rely on his luck to make the right choice. Gambling games heavily rely on this dynamic but they don’t seem to be great games. Of course they provide excitement but their excitement comes from the possible rewards not their “gameplay”. A lot of casual and f2p games also make use of random chances to provide excitement with uncertainty. It is a well established tool and comes in almost infinite forms.

Execution Chances: Games that required certain twitch skills to execute specific tasks have this kind of chances. Of course these chances feel a lot less random than the random chances created by the software but they are still some kind of chance. The hit ratio in ego shooters is a perfect example to explain this. Let’s say a player’s hit ratio is about 30% (a lot of games track and inform the player about this information). When he encounters an enemy his chances to hit the player depending on the current situation is about 30%. Even if there is a good chance that the player may miss with his two first shots the perception varies widely from explicit random chances. The other great difference between these two chances is that the player can invest time to increase his chances by training his precision and thereby his hit ratio.

Complex Situations

In some games every player knows everything about the current game state. These games are games of perfect information. Those games take their uncertainty of the complexity in their possible outcomes. In chess both players know everything about the game’s state but there is still a lot of uncertainty about the future moves of the other players. The possibilities that both players have and the resulting consequences create a huge possibility space. It is so huge that players can’t process all possibilities and make the perfect move in the given time frame. This way the game is capable of creating uncertainty without chances or hidden information.

Investment

Players can develop a huge intrinsic motivation when they invest enough time and energy during a gameplay loop. The perceived gameplay loop by the player often consists of many small gameplay loops. To evaluate the result of a gameplay loop  and perceive it as finished the player needs a feedback that communicates the outcome of his actions and enough time to evaluate this feedback. As long as there is not enough time or clear feedback and the players keep investing through their input the excitement keeps rising. Players can celebrate a kill in a League of Legends match during the laning phase but they don’t if the kill happens in a huge teamfight that is far from being over as there are still plenty of things to do that determine the outcome of the current perceived loop (the teamfight). The more actions players have to perform and the more difficult these actions are the higher the stakes of the loop in terms of investment. Street Fighter matches need a lot of input that is hard to execute and don’t give players time to evaluate their performance until the very end of the fight. The newly added “magic pixel mechanic” that only allows winning through direct- or critical art chip damage keeps the uncertainty up until the very end (Remember: uncertainty is needed in order to create excitement). This way the amount of actions / investment keeps rising for a long time and this way the excitement also keeps rising. As soon as the perceived gameplay loop is finished the player is released and able to relax until the next exciting loop is perceived.

So if you want to increase the excitement of your players via the needed investment you have to increase the amount and difficulty of the required actions to achieve their perceived goal. But beware, of course there’s a spectrum of how many actions your players are willing to do in order to achieve their goal. Games that involve a lot of difficult challenges with no clear reward in between so players can relax and feel released are normally more suitable for a core audience that is really engaged. Less engaged players could lose their interest for your game when facing these kinds of exhausting loops.

Rewards and Punishment

As soon as the result of the situation has the potential of a meaningful reward for the player excitement can emerge. Play Rock Paper Scissor and the winner gets 100,000$. This will make the game exciting. You don’t have to invest a lot to play this game but the potential reward of receiving so much money makes it exciting. The higher the reward for the winner in comparison to the losers penelisation the more excitement can emerge through the reward of the gameplay loop. Now let’s say that you play RPS and the winner gets 100,000$ but the loser still gets 95,000$. Of course it is still pretty cool to get so much money but the excitement will fall because the difference between both outcomes is much smaller.

It is important to note that the reward doesn’t have to be the final reward at the end of the game. The gold you get when killing an enemy champion in League of Legends is also important even if it is far from securing the win of the game. The reward of a gameplay loop is a big help for players to evaluate the outcome of the situation. As soon as they perceive a reward it is much more easier to perceive the gameplay loop as finished. This way the excitement can cool down and start rising again as soon as the player encounters the next gameplay loop that has an uncertain outcome.

League of Legends vs. Starcraft 2

I don’t want to compare these games in terms of their quality I just want to compare them in terms of their gameplay loop constructions and the resulting excitement. Not every player likes the same amount of excitement. Especially when the (time) investment was big and a gameplay loop ends without the desired outcome players can feel a lot of frustration and may churn forever.

I think that League of Legends with it’s laning, last hitting and champion kills/respawn dynamics achieves something special as those elements work as small gameplay loops players can experience while being in the game’s big gameplay loop. Both actions give a very clear feedback in form of a reward (gold & xp) and can be evaluated due to the game’s pacing because as soon as the hostile champion is dead the player’s champion is mostly save. This way players can experience excitement and relaxation on a regular basis, even if the game’s whole gameplay loop itself is much bigger and longer. The perceived gameplay experience consists of a lot small gameplay loops, each of them features the possibility for excitement and release. At the same time all those loops build towards a big loop whose goal is to destroy the enemy's nexus. Here is an example for a typical League of Legends game in terms of its excitement-curve:

In Starcraft 2 even if it has a similar structure the feedback during the game isn’t nearly as clearly as in League of Legends. All your actions are much harder to evaluate in terms of their contribution towards your desired outcome of the game because the whole gamestate is much more complex and the rewards are much more volatile than in League of Legends. You may have won a fight in the middle of the map but your opponent may already build a second base without your knowledge. The perceived gameplay loops are much bigger than the small loops during the League of Legends laning phase and because of this the amount of investment and with it the excitement raises on a much higher level during a Starcraft 2 game. At the same time the stakes of each gameplay loop in Starcraft 2  are much higher. Losing one of them can cause the loss of the whole game in League of Legends you may give the opponent player some gold, XP or lose a tower. Only after the team has lost several towers it has to fear losing of the whole game. Here is an example for a typical Starcraft 2 game in terms of its excitement-curve:

There are a lot less perceived gameplay loops but each loop has a huge excitement value. This way players are thrilled for a much longer time. In my opinion this is a result when you combine high stakes with high investment loops.

I hope that this article helped you to think about your games excitement-curve.