Esports has become a popular field of research, enabling advances in areas such as machine learning and environment modeling. However, complex modeling systems require complex visualisations. Despite that, visualisation of complex modeling systems within esports have been limited or fragmented, particularly when focused on the audience. Furthermore, the use of data visualisation and data-driven storytelling has been proven to be an effective and imperative method for enhancing audience experience for esport spectators. Therefore, this paper investigates data visualisation techniques within esports, and compiles design considerations for developing visualisation tools for esports broadcast. This is achieved through a case-study, in which the WARDS model was utilised in live coverage of a Dota 2 tournament and evaluated through observational data.

Esports - competitive video games watched by online audiences - are the fastest growing form of mainstream entertainment. Esports coverage is predominantly delivered via online video streaming platforms which include interactive elements. However, there is limited understanding of how audiences engage with such interactive content. This paper presents a large-scale case study of an interactive data-driven streaming extension developed for Dota 2, reaching over 300,000 people during the DreamLeague Season 15 DPC Western Europe tournament. The extension provides interactive live statistics, analysis and highlights reels of ongoing matches. This paper presents an analysis of audience telemetry collected over the course of the four week tournament, introducing a novel approach to analysing usage data delivered seamlessly in conjunction to a linear broadcast feed. The work presented advances our general understanding of the evolving consumption patterns in esports, and leverages esports as a lens to understand future challenges and opportunities in interactive viewing across sports and entertainment.

Data analytics is commonly used to enable storytelling and enhance esport coverage. One prominent use of it is win prediction, where machine learning models predict the winner of the game before its conclusion. However, predictions are most commonly results of black-box systems, forcing commentators to produce ad-hoc interpretations. Additionally, broadcasters generally rely other metrics to build narratives, limiting the impact of win prediction models for storytelling. This paper explores an alternative method to win prediction, identifying the needs of broadcasters to guide development of a novel win condition model. By focusing on existing storytelling points, the proposed win condition model can offer greater storytelling opportunities to broadcasters, focusing on the user needs identified from within the esport domain. Rather than utilising game state data to predict the winner, as it is usually done in win prediction, the proposed win condition model uses an exploration of the possible winners to predict the game state needed for each team to win. Lastly, the features identified for win condition are evaluated through a series of machine learning models, which provide a data-driven metric to test and predict win condition in the context of Dota 2, a popular esport title.

Esport is a large and still growing industry with vast audiences. Multiplayer Online Battle Arenas (MOBAs), a sub-genre of esports, possess a very complex environment, which often leads to experts missing important coverage while broadcasting live competitions. One common game event that holds significant importance for broadcasting is referred to as a team fight engagement. Professional player's own knowledge and understanding of the game may provide a solution to this problem. This paper suggests a model that predicts and detects ongoing team fights in a live scenario. This approach outlines a novel technique of deriving representations of a complex game environment by relying on player knowledge. This is done by analysing the positions of the in-game characters and their associated cameras, utilising this data to train a neural network. The proposed model is able to both assist in the production of live esport coverage as well as provide a live, expert-derived, analysis of the game without the need of relying on outside sources.

Work in AI-based explanation systems has uncovered an interesting contradiction: people prefer and learn best from why explanations but expert esports commentators primarily answer what questions when explaining complex behavior in real-time strategy games. Three possible explanations for this contradiction are: 1.) broadcast audiences are well-informed and do not need why explanations; 2.) consuming why explanations in real-time is too cognitively demanding for audiences; or 3.) producing live why explanations is too difficult for commentators. We answer this open question by investigating the effects of explanation types and presentation modalities on audience recall and cognitive load in the context of an esports broadcast. We recruit 111 Dota 2 players and split them into three groups: the first group views a Dota 2 broadcast, the second group has the addition of an interactive map that provides what explanations, and the final group receives the interactive map with detailed why explanations. We find that participants who receive short interactive text prompts that provide what explanations outperform the no explanation group on a multiple-choice recall task. We also find that participants who receive detailed why explanations submit reports of cognitive load that are higher than the no explanation group. Our evidence supports the conclusion that informed audiences benefit from explanations but do not have the cognitive resources to process why answers in real-time. It also supports the conclusion that stacked explanation interventions across different modalities, like audio, interactivity, and text, can aid real-time comprehension when attention resources are limited. Together, our results indicate that interactive multimedia interfaces can be leveraged to quickly guide attention and provide low-cost explanations to improve intelligibility when time is too scarce for cognitively demanding why explanations.

Esports are competitive videogames watched by audiences. Most esports generate detailed data for each match that are publicly available. Esports analytics research is focused on predicting match outcomes. Previous research has emphasized prematch prediction and used data from amateur games, which are more easily available than those from professional level. However, the commercial value of win prediction exists at the professional level. Furthermore, predicting real-time data is unexplored, as is its potential for informing audiences. Here, we present the first comprehensive case study on live win prediction in a professional esport. We provide a literature review for win prediction in a multiplayer online battle arena (MOBA) esport. This article evaluates the first professional-level prediction models for live DotA 2 matches, one of the most popular MOBA games, and trials it at a major international esports tournament. Using standard machine learning models, feature engineering and optimization, our model is up to 85% accurate after 5 min of gameplay. Our analyses highlight the need for algorithm evaluation and optimization. Finally, we present implications for the esports/game analytics domains, describe commercial opportunities and practical challenges, and propose a set of evaluation criteria for research on esports win prediction.

Esports (competitive videogames) have grown into a global phenomenon with over 450m viewers and a 1.5bn USD market. Esports broadcasts follow a similar structure to traditional sports. However, due to their virtual nature, a large and detailed amount data is available about in-game actions not currently accessible in traditional sport. This provides an opportunity to incorporate novel insights about complex aspects of gameplay into the audience experience – enabling more in-depth coverage for experienced viewers, and increased accessibility for newcomers. Previous research has only explored a limited range of ways data could be incorporated into esports viewing (e.g. data visualizations post-match) and only a few studies have investigated how the presentation of statistics impacts spectators’ experiences and viewing behaviors. We present Weavr, a companion app that allows audiences to consume data-driven insights during and around esports broadcasts. We report on deployments at two major tournaments, that provide ecologically valid findings about how the app’s features were experienced by audiences and their impact on viewing behavior. We discuss implications for the design of second-screen apps for live esports events, and for traditional sports as similar data becomes available for them via improved tracking technologies.

Esports have become major international sports with hundreds of millions of spectators. Esports games generate massive amounts of telemetry data. Using these to predict the outcome of esports matches has received considerable attention, but micro-predictions, which seek to predict events inside a match, is as yet unknown territory. Micro-predictions are however of perennial interest across esports commentators and audience, because they provide the ability to observe events that might otherwise be missed: esports games are highly complex with fast-moving action where the balance of a game can change in the span of seconds, and where events can happen in multiple areas of the playing field at the same time. Such events can happen rapidly, and it is easy for commentators and viewers alike to miss an event and only observe the following impact of events. In Dota 2, a player hero being killed by the opposing team is a key event of interest to commentators and audience. We present a deep learning network with shared weights which provides accurate death predictions within a five-second window. The network is trained on a vast selection of Dota 2 gameplay features and professional/semi-professional level match dataset. Even though death events are rare within a game (1% of the data), the model achieves 0.377 precision with 0.725 recall on test data when prompted to predict which of any of the 10 players of either team will die within 5 seconds. An example of the system applied to a Dota 2 match is presented. This model enables real-time micro-predictions of kills in Dota 2, one of the most played esports titles in the world, giving commentators and viewers time to move their attention to these key events.

Interactive surfaces are increasingly common in museums and other informal learning environments where they are seen as a medium for promoting social engagement. However, despite their increasing prevalence, we know very little about factors that contribute to collaboration and learning around interactive surfaces. In this paper we present analyses of visitor engagement around several multi-touch tabletop science exhibits. Observations of 629 visitors were collected through two widely used techniques: video study and shadowing. We make four contributions: 1) we present an algorithm for identifying groups within a dynamic flow of visitors through an exhibit hall; 2) we present measures of group-level engagement along with methods for statistically analyzing these measures; 3) we assess the effect of observational techniques on visitors' engagement, demonstrating that consented video studies do not necessarily reflect visitor behavior in more naturalistic circumstances; and 4) we present an analysis showing that groups of two, groups with both children and adults, and groups that take turns spend longer at the exhibits and engage more with scientific concepts.

In this paper we describe visitor interaction with an interactive tabletop exhibit on evolution that we designed for use in natural history museums. We video recorded 30 families using the exhibit at the Harvard Museum of Natural History. We also observed an additional 50 social groups interacting with the exhibit without video recording. The goal of this research is to explore ways to develop "successful" interactive tabletop exhibits for museums. To determine criteria for success in this context, we borrow the concept of Active Prolonged Engagement (APE) from the science museum literature. Research on APE sets a high standard for visitor engagement and learning, and it offers a number of useful concepts and measures for research on interactive surfaces in the wild. In this paper we adapt and expand on these measures and apply them to our tabletop exhibit. Our results show that visitor groups collaborated effectively and engaged in focused, on-topic discussion for prolonged periods of time. To understand these results, we analyze visitor conversation at the exhibit. Our analysis suggests that social practices of game play contributed substantially to visitor collaboration and engagement with the exhibit.

Multi-touch technology lends itself to collaborative crowd interaction (CI). However, common tap-operated widgets are impractical for CI, since they are susceptible to accidental touches and interference from other users. We present a novel multi-touch interface called FlowBlocks in which every UI action is invoked through a small sequence of user actions: dragging parametric UI-Blocks, and dropping them over operational UI-Docks. The FlowBlocks approach is advantageous for CI because it a) makes accidental touches inconsequential; and b) introduces design parameters for mutual awareness, concurrent input, and conflict management. FlowBlocks was successfully used on the floor of a busy natural history museum. We present the complete design space and describe a year-long iterative design and evaluation process which employed the Rapid Iterative Test and Evaluation (RITE) method in a museum setting.

In this paper, we present the DeepTree exhibit, a multi-user, multi-touch interactive visualization of the Tree of Life. We developed DeepTree to facilitate collaborative learning of evolutionary concepts. We will describe an iterative process in which a team of computer scientists, learning scientists, biologists, and museum curators worked together throughout design, development, and evaluation. We present the importance of designing the interactions and the visualization hand-in-hand in order to facilitate active learning. The outcome of this process is a fractal-based tree layout that reduces visual complexity while being able to capture all life on earth; a custom rendering and navigation engine that prioritizes visual appeal and smooth fly-through; and a multi-user interface that encourages collaborative exploration while offering guided discovery. We present an evaluation showing that the large dataset encouraged free exploration, triggers emotional responses, and facilitates visitor engagement and informal learning.

In spite of many advances in GUI workstations, the keyboard has remained limited to text entry and basic command invocation. In this work, we introduce the Touch-Display Keyboard (TDK), a novel keyboard that combines the physical-ergonomic qualities of the conventional keyboard with dynamic display and touch-sensing embedded in each key. The TDK effectively transforms the keyboard into an interactive surface that is seamlessly integrated with the interaction space of GUIs, extending graphical output, mouse interaction and three-state input to the keyboard. This gives rise to an entirely new design space of interaction across keyboard, mouse and screen, for which we provide a first systematic analysis in this paper. We illustrate the emerging design opportunities with a host of novel interaction concepts and techniques, and show how these contribute to expressiveness of GUIs, exploration and learning of keyboard interfaces, and interface customization across graphics display and physical keyboard.

VoodooSketch is a system that extends interactive surfaces with physical interface palettes on which users can dynamically deploy controls as shortcut to application functionality. The system provides physical 'plug and play' controls as well as support for sketching of controls, and allows controls to be associated with application functions via handwritten labels. The system uses a special digital pen, which writes 'real' ink on the palettes while functioning as a digital input device on the interactive surfaces. The palettes can be seamlessly integrated into existing applications, be appropriated by the user to suit different input requirements, and support new interaction styles across multiple surfaces, palettes and users.

We present a system that extends a typical workstation environment with a malleable physical interface for working with digital clips. It allows users to pick digital clips, give each its own dedicated key for direct access, and combine keys dynamically on a physical surface in a way that inherently reflects the state of an extended clipboard. The system affords copying and pasting of multiple clips each directly accessible through its own key shortcut. The keys can also be dynamically re-arranged to organize clips, and taken from workstation to another to transport clips, acting simultaneously as token and as copy-paste-interface for a digital object.