Importance of Slot Analytics
With slot operations becoming an important part of any gaming company’s business, Slot Analytics are getting more essential than ever.
In some parts of the world slots are generating higher revenues than Table Games and have become the most important source of gaming income. For example, Slot Win in Nevada during the period of six months (October 2019 – March 2020) amounted to 65% of total revenue, compared to just 50% in 1980’s (https://gaming.unlv.edu/reports.html).
In Macau, with Baccarat as the most popular game (88% of the total revenue), Slot result is just 5% from the total revenue but still a lot higher than Roulette 0.4% and Black Jack 1%. Undoubtedly, Baccarat is “the game” in Macau and I would like to separate Baccarat win for the sake of this calculation – therefore Slot Win is 44% from non-Baccarat revenue. Interestingly, Sic Bo Win is half of the Slot Win but 3 times BJ and 8 times Roulette Win (2017-2018 data, https://gaming.unlv.edu/abstract/macau.html ).
With Slot revenues gaining a higher share of Casino revenues, Slot Analytics has to be taken seriously. You have to know your slot numbers and be able to analyse them to figure out your players’ preferences, playing habits, and effectiveness of your promotional campaigns to be able to optimize your operations.
I know, some might find numbers dull and boring but do not underestimate their importance. There is a saying “information is power”, well, the statistical information about your slot operations and ability to use that information is your power.
Slot Analytics involves a wide range of analytical activity and should be separated into at least two categories:
- Slot Performance Analysis
- Slot Comparison Analysis.
Slot Performance Analysis
Slot Performance Analysis is used to compare actual slot machine/game performance to the settings and game’s paytable, and generally involves answers to the following questions:
- Is the actual hold of the machine the same as the theoretical hold?
- If not, then is it within an acceptable interval?
- If not, then why, etc.
Slot Performance Analysis is mostly used to make sure a game is performing to spec and to eliminate potential fraud or malfunction. To properly investigate differences between the Theoretical Hold and Actual Hold we need more specific data from the game producer – Confidence Interval, Total Games and Volatility Index.
Unfortunately, most of the game manufacturers label their games as Low/Medium/High Volatility without an exact VI (Volatility Index) number. This means that in some cases with high variance between the actual and the theoretical hold, the analyst should “flag” the game and keep monitoring it until the variance is within accepted range.
Slot Comparison Analysis
Slot Comparison Analysis as part of Slot Analytics is used to find out which slot machines or slot machine groups are more profitable for the casino by analysing a set of KPIs (Key Performance Indicators).
The most simple approach would probably be using Win Per Machine but in most cases it would not give us a fair picture because the sample size is not big enough (not enough games played leading to larger differences between Theoretical and Actual Win).
The KPIs used in basic Slot Comparison Analysis are:
- Coin-In per Machine per Day (actually, it is the Total Bet per Day. The term “Coin-In” comes from old slot machines when coins were used and Coin-In was equal to Bet).
- Actual Win per Machine per Day.
- Theoretical Win per Machine per Day.
- Variance % of Actual to Theoretical Win.
- Games Played.
- Average Bet.
- Occupancy %.
Other KPIs used in Slot Comparison Analysis are:
- Game’s Max Bet Capacity.
- Break Even Theoretical Win Per Machine.
- % of Machine Days.
- % of Machine Win.
- GPM (Game Speed – Games Per Minute).
- Actual RTP%
Most Slot Performance KPIs are easy to understand. Theoretical Win, however, sometimes causes confusion. If you are a teacher or lecturer, you probably know the feeling very well – you are in the classroom explaining something and you see students looking at you with confusion in the eyes and a silent question: “What are you talking about?”
Theoretical win of a slot machine is the expected win, based on the total amount wagered on that machine. All you need is theoretical RTP% and the total amount wagered (Total Bet) to calculate it:
Theo Win = (1 – RTP%) x Total Bet
Example: (1 – 96%) x 100,000 = 4% x 100,000 = 4,000
Slot Mix Analysis
One of the most important analyses for Slot Analytics is Slot Mix analysis. You can look at your slot floor mix through various dimensions such as Game Set (see the sample below), Manufacturer, Cabinet Model, Denomination, Minimum Bet, Maximum Bet, Fixed Lines, and Number Of Lines to see if there are any imbalances indicating a need to change the supply to better serve the demand.
In the example below you can see that VIP Orange Collection and Egypt Quest JP game sets are in equilibrium – their share of the slot floor is very close to their share of the Theoretical Win. VIP Orange Collection game set is %6,6 of the slot floor and the game set’s share of Theo Win is %5.7. Demand for CF2 Novostar game sets is lower than supply – %12.7 from the floor and %6.2 from Total Theo Win.
Obviously, there are many factors influencing performance of slot machines. The same machine with the same games might perform completely differently depending on the location on the floor, Min-Max bet settings, denomination settings, etc.
Sometimes it is not easy to understand the reasons why some machines are performing better than others.
Evaluating the effect of changes we make to the game mix can be challenging. It is mostly because besides game mix adjustments we also have a continuous stream of multilayered promotional activities and seasonal business fluctuations.
In such cases we have to use more advanced statistical techniques, such as multiple regression analysis. Regression allows us to better understand the effects of seasonality, promotional campaigns, and other variables that might influence turnover.