Authors’ Note: This article covers the fundamental difference between traditional and downloadable slot games and how this difference may affect the dynamic gaming floors of the future. Currently, operators’ use of downloadable games seems to be due to the ability to run a floor in the same way but with fewer slot techs, so we’ll also explore management techniques for downloadable games that will change the way we run gaming floors. We welcome Dr. Thomas as a new contributor to this series.
Almost all of the gaming machines today contain sophisticated software, a far cry from the humble spinning mechanical reels.1 The concept of centrally controlling the game configuration and the game seems like a natural next step for the technology. In fact, this step has, for years, been presented as heralding a future of increased profits. However, these profits can only be realized if operators are able to use these capabilities to change the way they run their gaming floor.
When an industry changes so much that its current best practices are no longer relevant, it becomes very important to look outside of the industry for input on how to optimize profit. Retail operators, for example, offer the best possible price on a volume basis, finding the best balance between sales volume and profits per sale (e.g., 20 sales at $1 profit per sale ($20 total profit) vs. 200 sales at $.50 profit ($100) vs. 2,000 sales at $.005 profit ($10)). To translate this into gaming, the price can be thought of as a combination of hold percentage, game play and minimum bet—all parameters of a slot game that can easily be changed with downloadable games.
Some gaming operators are sure to offer price incentives. For example, they could offer the current loosest video poker hold on the entire floor: 0.1 percent hold on every single game. In short, this may create competitive pandemonium in a market. Table games have suffered from this competitive pandemonium for years, with operators offering 100X odds on craps, giving an approximate 0.2 percent effective hold. While the example is dramatic, it illustrates that the impacts of a dynamic gaming floor are difficult to predict.
This example also illustrates how math becomes central to making money: Managing the number of 0.1 percent games, the customers’ yield of wallet and the amount of time the customers are willing to play is going to be the key component of operating the dynamic gaming floor.
The Self-Selection Model
One school of thought dictates that downloadable gaming should be presented to the player as an interactive screen where they can choose from a vast range of game themes in almost any configuration. This approach presents a fascinating challenge to gaming floor design, as the games would simply become cabinets, with the users choosing the product placement and pricing.
This style of game has been available in some formats for years, but until recently, the data collected has been an aggregate of location performance. The key questions for this style of game are:2
1) How can we get players to choose high-yield games?
2) How can we promote new games?
3) What price points should we allow for play?
4) Should we allow players to choose the hold percentage?
5) How can we avoid commoditization of the gaming product?
6) And, quite simply, how can we analyze the relative performance of games?
Under this model, quite simply, we are leasing real estate on the gaming floor. We design this real estate so that our best customers get the best experience, but there are some constraints on this design, including themed cabinets, progressive jackpots and signage.
Constraint 1: Themed Cabinets
Walking around many gaming floors, it’s hard to miss the cabinets of all shapes and sizes, often stretching nine or 10 feet into the air, with signage towering above that. Until polymorphic gaming cabinets are available, it is sensible to assume that the kind of games that can be downloaded into these cabinets is limited to games that make sense for the cabinet design. The performance numbers for custom-designed games allow manufacturers to charge a premium participation model. One does not have to look much further than some of the large wide area progressive (WAP) games to understand the importance of these games to the industry. In fact, the visual dominance of the participation products act as landmarks within the gaming floor.
While participation games have been known to create tension between operators and slot manufacturers, the nearly universal success of the products and the importance of the revenue-sharing participation model means that the manufacturers will most likely continue to make and operators mostly likely continue to deploy these heavily themed games. As manufacturers add downloadable aspects to participation games, they will be constrained by the fundamental design of the often quite exotic cabinets.
Taking themed cabinets into consideration when looking at the impacts of downloadable games brings a constraint to flexibility of the gaming floor. This constraint creates landmarks within the gaming floor, defining areas and attracting specific kinds of customers to those areas.
Constraint 2: Progressives
Progressives come in many flavors, including game-based, location-based, play-based and wide area. These progressive jackpots are often layered into the gaming floor, and operators have become very proficient at creating layouts that enable patrons to understand the progressives and how they are linked.
This operator-designed layout includes everything from huge signs outside of the property, advertising the property-wide jackpot, to huge rotating signs within the casino to numbers displayed in the top-box of the game. Sometimes the linked jackpots are designed to create a community aspect and span across different game types.
In any case, the second major constraint in laying out a property is management of the progressive jackpots. This second constraint sometimes creates common areas or possibly common themes that stretch across the gaming floor. These constraints have a major influence on customer behavior, bringing certain kinds of customers to specific areas and linking the behavior of customers across the floor.
Constraint 3: Signage
A third dominating feature of a gaming floor is that the signs are strategically placed across the floor. These signs often relate to a specific game type or game. Operators invest capital in these signs in the belief that they enhance the gaming floor’s appearance. These signs are difficult to move but create another major landmark within the gaming floor.
These signs are a third major constraint on the gaming floor. In many ways, the gaming floor design is structured around these constraints, and the success of the dynamic floors of the future will be defined by how the floor is designed around these constraints.
One area of the downloadable gaming floor that will not be constrained, however, is the data. Data volumes generated are about to explode. The question now is what will be stored and what value does this stored data bring?3
For a casino of 1,000 games, we estimate there will be 480X more data generated in the near future (one record per game play compared to the current one record per hour). As illustrated in Table 1, this takes the volume of data generated in five years from 89 GB to 43 TB.
Quite simply, this 480X growth in data creates the backbone of what is about to become a massive math problem. This dramatic increase in the volume of transaction data is accompanied with an associated dramatic increase in the dimensionality of the data. The traditional world of attributes such as game theme is now replaced with game theme analysis.
Games Themes as Apps?
Currently, the barriers to entry into the gaming market for game designers are significant, with hardware and software requirements coming along with a complex sea of necessary certifications and approvals. Software companies, on the other hand, are renowned for starting in somebody’s garage and still producing world-class products. Why can’t this be the case for downloadable slot game design? There is little doubt that the approval process for a software product is a great deal simpler than testing a whole cabinet, its peripherals and the game “framework.”
Take the iPhone, for example. The iPhone app growth is one of the hottest stories in the software industry today, and this story has shown that software engineers produce vast numbers of apps for a well-defined framework (see Table 2 and Chart 1).4
With the opening up of the casino floor to game designers, it is quite likely that a vast number of themes will become available. Imagine being able to select from hundreds or thousands of new game themes each month and deploying selected themes onto the gaming floor in seconds.
This sea of change in the industry is definitely creating an ocean of future possibilities in which the winners and losers are, without perfect foresight, hard to pick. However, we can describe some of the techniques that can be applied to benefit from these changes as they happen in the gaming world.
Metrics and Attributes
Enter the new metrics of the gaming floor. These metrics do not provide the categorization that we have relied on for years in the industry; in the future there is no game theme, no denomination and no number of lines. Instead we need to understand all the configurations available to the customer at the time the customer is on the game. The “combinatorics” behind what used to be very simple information could become staggering—a game that averages 10 available themes and five available denominations that is scheduled to change once per week, for example, will have 200 different configurations in a single month. Applied to a floor with 1,000 games, this means that instead of 1,000 data points indicating the floor configuration over the course of a month, we have 200,000! Removing these staples of gaming analysis is like saying we can no longer drive our cars; it creates a very basic and fundamental change in how we do things. It might mean that the work will take longer in the beginning, and it will certainly lead to different approaches to the work in the first place.
In addition, the gaming machine manufacturer of today is likely to disappear,5 with companies being able to deploy G2S-compliant games from any games designer, and the cabinet and the game theme will be separated. And it will not stop there—secondary gaming events such as jackpots, bonuses and gaming floor specific events can be combined from multiple providers. To illustrate this, a WMS game can exist in an IGT cabinet with an ARIA-specific jackpot. Following is a sample of some of the impacts on metrics and attributes of the future.
The Demise of Win Per Unit
Today, games are commonly selected and described on the basis of win-per-unit (WPU) and utilization, typically weighted more heavily on WPU. But in a server-based environment, there is no such thing as a game unit; there is now a game container and a game theme. In this more flexible environment, product use will be weighted solely on utilization and yield management. So, in a sense, available product analysis becomes completely disconnected from WPU and tied directly to utilization and yield.
Location becomes even more interesting in this yield management environment, as today it is essentially the job of the operator is to manage yield so the best players find the games they want at the configurations they like in the highest-yielding areas of the floor. But if full self-service downloadable games were implemented, this yield management would become ineffective. Simply put, in order to avoid a “first come, first serve” floor where low-yield customers control the game mix, operators will need to be very proactive.
To illustrate this concept, ask this question: If, as an operator, you have a game that generates 50 percent below average WPU but has people waiting to play it on the weekends, do you purchase more of these games?
Game Theme Metrics
With multiple themes on a cabinet during a time period, either through player-selection or operator control, it is humans that are controlling the games that are available at any point in time. Historically we would leave a game at a location for a number of months, and this long period of stable exposure would give a stable analytical platform.
Following are some new metrics for downloadable gaming:
• Number of game themes at a location;
• Percent of self-selection for a theme; and
• Highest performing game theme and relative theme index.
There was a time when denomination dictated the level of play, but with the rise of video slots, denomination became an attribute of play that slots systems seldom collected. Now enters the denomination metrics, which are measures of what kind of play is taking place.
Following are some denomination metrics relevant to downloadable gaming:
• Mean denomination (a simple average of the denomination is quite different from the average bet but does define how the game is being played);
• Median denomination;
• Mode denomination;
• Game theme frequency distribution; and
• A measure of diversity (e.g., Shannon’s Entropy).
Most gaming systems now contain one or more secondary windows, and these windows are full-fledged gaming devices in their own rights, although at his Gaming Technology Summit keynote address, Rich Schneider from IGT stated that his customers see the secondary games as “part of the actual game.” The secondary game brings combinatory complexity to the analytics, which along with questions like “Which secondary games should be used with this theme?” will challenge the gaming analyst of the future.
Following are some new metrics and attributes:
• Number of secondary games;
• Percent of time spent in secondary games; and
• Actual hold of secondary game.
The Dynamic Gaming Floor
In researching and writing this article, we, the authors, came to the conclusion that downloadable games are inevitable and will enable enormously different ways of looking at and managing the gaming floor. We initially planned only a three-part series on this topic but are now so passionate about how much this sea of change rewrites the rules of analytics, we have decided to expand this series into 12 parts.
As we have discussed in this article, downloadable games can change almost every aspect of gaming floor management. We show that to gain value from this approach, operators need a new set of management methods and analytics. Gaming made the math of probability,6 and we are excited to be part of the latest extensions to this art.
1 Singh, A. K. and Cardno, Andrew. “Gaming 2018: Searching for a Simpler World.” Casino Enterprise Management, October 2008.
2 These will be further explored in future articles
3 Singh, A. K. and Cardno, Andrew. “The Petabyte Era of Gaming Data.” Casino Enterprise Management, September 2008.
5 Singh, A. K. and Cardno, Andrew. “The Demise of the Slot Manufacturer.” Casino Enterprise Management, July 2008.
6 Singh, A. K. and Cardno, Andrew. “The Math That Gaming Made.” Casino Enterprise Management, November 2008.
Andrew Cardno has more than 16 years of experience in business analytics, ranging from modeling health care drive times to casino gaming floor analytics. He often presents on the future of analytics across the world and has spent the last seven years living in the United States and working with corporations around the world. He can be reached at andrewcardno[at]yahoo.com.
Dr. A. K. Singh is a professor at UNLV. He has taught statistics, mathematics and operations research courses at New Mexico Tech and advanced statistics classes including Time Series Forecasting and Data Mining at Harrah Hotel College at UNLV. He has more than 80 publications in theoretical and applied statistics and can be reached at aksingh[at]unlv.nevada.edu.
Dr. Ralph Thomas is Vice President of Database Marketing for Seminole Hard Rock Gaming. During his years in the casino industry, Thomas has focused on maximizing profitability by applying statistical analysis to the company database. Previously, Thomas spent 15 years in academia, as both a student and a lecturer of mathematics.