Denna rapport har undersökt om game metrics genom heat maps kan användas för att hitta en choke point i en bana gjord till spelet Team Fortress 2. Game metrics och kvantitativa  metoder  erbjuder  ett  objektivt  och  nästan  automatiserat  alternativ  till kvalitativa metoder när det kommer till balansering. En bana har konstruerats med en choke point och har speltestats för att generera en heat map. För att undersöka om det går att hitta en choke point med hjälp av en heat map så har en enkät gjorts där respondenter   bads   hitta   choke   pointen   med   hjälp   av   den   heat   mapen   som genererades från speltestningen av banan. Alla respondenter lyckades hitta mitten av choke  pointen  med  hjälp  av  heat  mapen.  I  framtiden  skulle  arbetet  kunna  utökas genom  att  undersöka  om  användandet  av  bottar  eventuellt  skulle  helt  kunna automatisera  balanseringsprocessen.  Det  skulle  också  gå  att  undersöka  hur  olika klasser rör sig i en bana.

Game balance can be considerd complex - there are many factors that play a role in the perception of balance. This thesis examines complexity, game theory and intution in an effort to disconver more about the perception of balance. The findings are that tactical and strategical choices that are presented for a player need to be interesting, not to complex and not to obvious. The importance lies in the estimation of the result that the player does when making a choice. / Spelbalansering är komplext – det finns många faktorer som påverkar det som uppfattas som balans. I mitt verk undersöks speciellt komplexitet, spelteori och intuition i jakten på att gräva fram mer ut ämnet. Denna reflexiva rapport tar upp metod och arbetsprocessen runt verket och mitt arbete på ability-systemet på Lockpick Entertainment. Om de taktiska och strategiska val som presenteras för en spelare anses som intressanta beror både på spelaren och svårigheten i valet. Är valet komplext är det svårare att avgöra korrekt, och taktiska och strategiska avvägningar måste göras. Detta är grunden i spelbalansering, att spelaren har flera vettiga, men inte uppenbara, val att välja bland.

Game balance is the problem of determining the fairness of actions or sets of actions in competitive, multiplayer games. This problem primarily arises in the context of designing board and video games. Traditionally, balance has been achieved through large amounts of play-testing and trial-and-error on the part of the designers. In this thesis, it is our intent to lay down the beginnings of a framework for a formal and analytical solution to this problem, combining techniques from game theory and machine learning. We first develop a set of game-theoretic definitions for different forms of balance, and then introduce the concept of a strategic abstraction. We show how machine classification techniques can be used to identify high-level player strategy in games, using the two principal methods of sequence alignment and Naive Bayes classification. Bioinformatics sequence alignment, when combined with a 3-nearest neighbor classification approach, can, with only 3 exemplars of each strategy, correctly identify the strategy used in 55\% of cases using all data, and 77\% of cases on data that experts indicated actually had a strategic class. Naive Bayes classification achieves similar results, with 65\% accuracy on all data and 75\% accuracy on data rated to have an actual class. We then show how these game theoretic and machine learning techniques can be combined to automatically build matrices that can be used to analyze game balance properties.

Game balance is the problem of determining the fairness of actions or sets of actions in competitive, multiplayer games. This problem primarily arises in the context of designing board and video games. Traditionally, balance has been achieved through large amounts of play-testing and trial-and-error on the part of the designers. In this thesis, it is our intent to lay down the beginnings of a framework for a formal and analytical solution to this problem, combining techniques from game theory and machine learning. We first develop a set of game-theoretic definitions for different forms of balance, and then introduce the concept of a strategic abstraction. We show how machine classification techniques can be used to identify high-level player strategy in games, using the two principal methods of sequence alignment and Naive Bayes classification. Bioinformatics sequence alignment, when combined with a 3-nearest neighbor classification approach, can, with only 3 exemplars of each strategy, correctly identify the strategy used in 55\% of cases using all data, and 77\% of cases on data that experts indicated actually had a strategic class. Naive Bayes classification achieves similar results, with 65\% accuracy on all data and 75\% accuracy on data rated to have an actual class. We then show how these game theoretic and machine learning techniques can be combined to automatically build matrices that can be used to analyze game balance properties.

<p>Game balance can be considerd complex - there are many factors that play a role in the perception of balance. This thesis examines complexity, game theory and intution in an effort to disconver more about the perception of balance. The findings are that tactical and strategical choices that are presented for a player need to be interesting, not to complex and not to obvious. The importance lies in the estimation of the result that the player does when making a choice.</p> / <p>Spelbalansering är komplext – det finns många faktorer som påverkar det som uppfattas som balans. I mitt verk undersöks speciellt komplexitet, spelteori och intuition i jakten på att gräva fram mer ut ämnet. Denna reflexiva rapport tar upp metod och arbetsprocessen runt verket och mitt arbete på ability-systemet på Lockpick Entertainment. Om de taktiska och strategiska val som presenteras för en spelare anses som intressanta beror både på spelaren och svårigheten i valet. Är valet komplext är det svårare att avgöra korrekt, och taktiska och strategiska avvägningar måste göras. Detta är grunden i spelbalansering, att spelaren har flera vettiga, men inte uppenbara, val att välja bland.</p>

There is a lot of different kind of games which creates many different ways to balance the difficulty of games. This study will look at if damage output from a player is a good variable to scale in order to create a better balance and make the game feel more rewarding overall, based on that a game would be enjoyable if a player feels that it is rewarding to play the game. By letting both inexperienced and avid players test a part of a game with different settings for the damage output to see if the players finds the game to be more rewarding if the difficulty is set to a higher setting (lower damage output). The conclusion is that it is that damage output cannot directly affect how overall rewarding a player finds the game, but can affect other variables that in turn make the game feel more rewarding.

Achieving a state of balance is essential when developing a video game (Schell, 2019, “Game Mechanics Must Be in Balance,” para. 1). Despite this, game balancing is frequently overlooked in game development curricula (Schreiber, 2016, 00:30). This research describes the development and pilot study of a framework that junior game designers can utilize to gain valuable skills in the area of game balance. The framework produced by this research provides users with the ability to tune weapon parameters and see the effects these changes have on a first-person shooter deathmatch game in real time. Participants in the study utilized the framework to achieve three pacing and balance goals. Data regarding the weapon parameters selected by participants and information about the participants’ usage of the framework is described in detail. This study serves as the groundwork for future research focused on finding a method for teaching junior game designers about game balance.

This study aims to see if it is feasible, for the purpose of studying game balance, to put a concrete value on damage and other abilities in Real-Time Strategy (RTS) games. The main metric that was used for this analysis was the cost per DPS (Damage Per Second) for units in a RTS game. The game that was used to test this analysis was StarCraft II: Heart of the Swarm.   This topic was tested by first gathering data about the properties of units in the game and then calculating the average values of the units in the game. After that, a few units and game mechanics were chosen to be analyzed in further detail. They were analyzed by comparing their cost/DPS values to the average values and seeing if there were any other properties that might have an impact in the cost of a unit.   While it was concluded that it is feasible to put a value on damage and other properties of units, the subject needs to be approached with caution. Using the cost/DPS metric that was assumed for this thesis, some units came out as being too strong for their cost. Further studies on similar topics might be able to find better methodologies for approaching similar analyses. / Denna studie vill titta på om det är rimligt, för ändamålet att studera spelbalans, att sätta ett konkret värde på skada och andra förmågor i Realtidsstrategispel. Det huvudsakliga mätsättet som användes i denna analys var kostnaden per skada per sekund (DPS, Damage per Second på engelska) för enheter i Realtidsstrategispel. Spelet som användes för att testa denna analys var StarCraft II: Heart of the Swarm.   Studien genomfördes genom att först samla in data om enheternas egenskaper och sedan räkna ut genomsnittsvärden för enheterna i spelet. Efter det så valdas några enheter och spelmekaniker ut för att analyseras i mer detalj. De var analyserade genom att jämföra vad de hade för förhållande mellan sin kostnad och deras DPS-värden med det genomsnittliga och se om de hade andra egenskaper som kunde ha påverkat enhetens kostnad.   Även om slutsatsen som drogs var att det är rimligt att sätta ett värde på skada och andra egenskaper hos enheter så måste man vara försiktigt om man vill studera detta ämne. Genom att använda sig utav kostnad/DPS värden som användes för denna analys så kom vissa enheter att se väldigt starka ut i förhållande till vad de kostar. Vidare studier på liknande ämnen kan hitta bättre sätt att närma sig liknande analyser.

Game balance has been a challenge for game developers since the time games have become more complex. There have been a handful of proposals for game balancing processes outside the manual labor-intensive play testing methods, which most game developers often are forced to use simply due to the lack of better methods. Simple solutions, like restrictive game play, are limited because of their inability to provide insight on interdependencies among the mechanisms in the game. Complex techniques framed around the potential of AI algorithms are limited by computational budgets or cognition inability to assess human actions. In order to find a middle ground we investigate Game Theory and Mechanism Design concepts. Both have proven to be effective tools to analyse strategic situations among interacting participants, or in this case `players'. We test the feasibility of using these techniques in an Real Time Strategy (RTS) game domain to understand game balance. MicroRTS, a small and simple execution of an RTS game is employed as our model. The results provide promising insight on the effectiveness of the method in detecting imbalances and further inspection to find the cause. An additional benefit out of this technique, besides detecting for game imbalances, the approach can be leveraged to create imbalances. This is useful when the designer or player desires to do so.

Matchmaking is an essential part of many modern online multiplayer games and is used by game developers to give the players the best possible online gaming experience. However, since video games have become more complex, traditional matchmaking systems like TrueSkill have reached their limits in predicting match outcomes. An extensive literature survey on engagement and balance in video games as well as an analysis of Battlefield 4 player data showed that balance can have a significant impact on player engagement. This lays the basis for the new matchmaking approach that is presented in this thesis. It is developed for the online First Person Shooter game Battlefield 4, with the goal of increasing player engagement by balancing online multiplayer matches. The developed matchmaking system is based on regression models, which use player performance metrics to predict the balance of online multiplayer matches. The experimental evaluations of the developed models show that the quality of the prediction results are influenced by the complexity of the different game modes available in Battlefield 4. Furthermore the historical Battlefield 4 game report data, which is used for building the predictive models, shows that this complexity as well as imbalances in the game design add significant noise to balance predictions. Both evaluated regression models – Linear Regression and Multivariate Adaptive Regression Splines – showed similar prediction errors within statistically expected deviation. Additionally it is shown that both methods have significantly smaller errors than the TrueSkill system, when predicting the outcome of games in Team Death Match or Conquest mode. The features that resulted in the lowest errors are commonly used in online First Person Shooter games. Hence the findings of this thesis can not only improve the matchmaking of Battlefield 4, but also benefit other video games of the same genre. / Matchmaking är en viktig del av många onlinespel och används av spelutvecklare för att ge spelare en bättre spelupplevelse. På senare tid har spelen blivit mer komplexa och traditionella matchmakingsystem som TrueSkill klarar inte längre förutse utfall av matcher. Andra studier som avhandlar spelaktivitet och balans i matcher visar att spelare har ökad aktivitet om deras matcher har varit jämna. Det går även att dra samma slutsatser från speldata i Battlefield 4. Syftet med denna uppsats är att utveckla och utvärdera en ny modell för matchmaking i spelet Battlefield 4 som medför bättre balans i matcher och därmed ökar spelaktiviteten. Den utvecklade matchmaking-modellen baseras på regressionmodeller som använder historiska speldata för enskilda spelare för att förutse jämnheten i matcher. Utvärderingen av den utvecklade modellen visar att förmågan att förutse matcher beror på hur hur komplext spelläget i Battlefield 4 är. Det framgår också efter att ha tittat på historiska spelrapporter att förutom komplexiteten i spelläget så påverkar också obalans i speldesignen förmågan att förutse matcher. Båda utvärderade regressionsmodeller – linjär regression och Multivariate Adaptive Regression Splines – har liknande förmåga att förutse utfallet av matcher där ingen är signifikant bättre än den andra. Dessutom visas att båda modellerna förutser matchresultaten för spellägena Team Death Match och Conquest signifikant bättre än TrueSkill. De här spellägena är vanliga i denna typen av spel så resultaten i denna uppsats är inte isolerade till Battlefield 4 utan går att applicera på många spel i samma genre.