Calculating Slot Machine Volatility Index

Marketing has many avenues, and when we speak of slot machine marketing there are many focuses. One is the introduction of new games, and the second is ability to keep the interest of the player. There is at every marketing person’s fingertips information that allows them to profile the player, know the player, their likes and dislikes. This is a player’s loyalty system and slot accounting system.

Now when I mention likes and dislikes, I am not talking about that drink they like, or what days of the week they dislike coming to the casino. I am talking about their volatility profile as outlined by the machines they like to play. To know a player's volatility, marketing needs to know and understand the volatility of the machines the player likes to play. This (the players volatility profile) in itself can be an issue depending on the players playing habits.

What is a playing Habit?

Video keno Volatility. For instance, on some video keno machines that are set to return roughly 92 percent payback, a player picking only two numbers is actually playing about a 90 percent return game, while playing six returns nearly 95 percent. A player picking 10 numbers is back down to 93 percent. This, generally speaking. Help for a math student from China who is calculating a slot machine's volatility. Advanced search Game Types Bonuses Slots More. Ask the Slot Expert: Calculating a slot machine's volatility. 5 Apr 2017 By John Robison, Slot Expert™. Where CI is the index of confidence level, N is the total cycle of the game, Pi could be zero (for non.

In simple terms it is the 'standard routine of play the player does each time they come to the venue'. This can be complicated or simple. Below is a complicated example to help you to better understand these issues.

Player’s budget: $200.00, 3 hours (yes, time is a budget item)

They start out on a low volatile game, wagering $0.50 per spin. From this game they realize they will get “time on device” and have the opportunity to win a few medium sized awards (20%-30%) of their cash budget.
Now once they have won, they move to the next higher volatile game, wagering $0.75 per spin that may or may not give them a win in their cash budget of (30%-50%). This is a warm-up game, but also has shown them from time to time, to give them good wins and long play, but is one they do not normally stay on.
Assuming they have won a little more from the “warm-up game”, they move to the higher volatile games wagering $1.25 per spin, and as their money swings high and low between these machines, they are hoping to win big at the moment their time budget of 3 hours runs out.
What is really the most important to the player, is that their $200.00 cash budget lasted the 3-hour time budget.
Note: The player does not know the volatility indexes of these machines, but they know from play what they can expected on a consistent bases, they can feel and understand the volatility of these games.

What is volatility index for slot machines in simple terms?

In many cases volatility is expressed by, “low, medium or high”. This is NOT a volatility index! This simplistic defining scale is as detailed to say, 'I wear a large shirt'. This may be true... I could wear a large shirt, but depending on the designer, it may fit tight or loose. Further, the sleeves may be too longer or too shorter for my arms, and the neck may be loose or tight; or it may fit just right. Getting it “just right” now becomes a low probability (less chance getting it right), as there are many attributes and variations to my fit.

Volatility of slot machines are the same way. So, what is volatility index? It is the calculated sum of each combination (winning or losing), known as its value (zero to maximum award), and how often this combination occurs in the total sum (all possible variations). Each combination has two attributes to it that are used for a volatility calculation:

Its value
How often it occurs

Calculating Slot Machine Volatility Index Calculator

Each combination is set by the manufacture to occur a number of times in the total combinations. This is calculated simply by calculating the number of the symbols (for each symbol) for that combination on each reel, then multiplying it by the number of that symbol on the next reel and so on.

Example:

There are 5 cherries on reel one, on reel two is 4 cherries and reel three is 8 cherries.
5 (reel one) x 4 (reel two) = 20, then multiply 20 (sum of reel one x reel two) x 8 (reel three) = 160 times there will be three cherries in the total combinations.
For the example, three cherries on the pay line will pay 50 credits.
So, a winning combination of three cherries will occur 160 times and will pay 50 credits each time.
Note: This is done with each combination and set of symbols, this becomes the total combination.

By calculating this for each combination, winning and losing combination volatility can be calculated (will be fully explained on a different post detailing slot machine volatility calculations) and then a volatility index is calculated. The manufactures provide this information on their PAR (Program Accounting Report) sheet.

Issues with volatility exists with a few factors, these are:

The Z score used to calculate the volatility. Depending on the manufacture and the jurisdiction, this could range from a score factor percent of 90, 95 or 99. (Do not get this mixed up with PO% “payout”, this is only a factor percent rating for a standard of accounting).
Volatility is calculated assuming within the acceptable variance, that all combinations, or using a standard of 10,000,000 games are achieved. Understandably, when a player plays a game, they will not get the exact volatility as they have only played a small sample of the total combination.
Changes in a games PO%, will change its volatility index. So, two identical machines, but with different PO%, will have different volatility.
If you use a volatility scale and not an index, the results are opinionated to the scale design. What does this mean?

A scale of low, medium, and high, is useless without knowing the minimum and maximum indexes for each of the scale levels.
In the case that the attributes of each level are known, it is the opinion of the creator of the scale as to what is low, medium, or high.

Now that the volatility index is known as an overall volatility for that machine, this information needs to be recorded into a data base (or slot accounting system). In addition to volatility, features such as bonus types, free spins features, and game type/style with any other play attributes should to be recorded.

Combining player’s habits with volatility index and scaling.

Marketing now needs to look at the games the player plays, this is can be done easily by the use of a player's loyalty card. Now if the player does not use this card, the process is much harder to track. Assuming they do use the card, most systems will provide data on each game the player plays, such as:

Handle (money wagered),
Number of games played and,
Average wagered per spin,
Each machines asset number (accounting) and/or location number (known as its identification number).

What is needed, is a way to tie the player's play to the data base; that has the games attributes such as volatility index and other attributes. This becomes an issue as many systems do not record this data. So, a separate data base can be created and merged on the back end, as simple as that sounds, it may be complicated.

For this example, we will assume that data is known and in the system. Now it’s a choice to use either total games played or total handle (money wagered) to decide on what percentage of data you want to use for play reference. In this example, I am using total games played. In the data collection, you want to be able to reference percentage minimums such as nothing lower than 15% of games played, this will help to weed out games they tried and did not like.

There is an issue with clones (games that have exact math models) that may not be liked, by the player depending on color or symbol designs, or its physical location on the gaming floor. This would be added to the percentage of play to a particular volatility index, but would be cut as a singular game from the total. To correct this, the data could be pulled by volatility index, total play then weeded out.

Example: Players plays:

29.36% of their games on a 8.73 volatility index game,
17.07% of their games on a 11.45 volatility index game,
30.39% of their games on a 14.56 volatility index, and
23.18% of their games on a 16.45 volatility index game.

The player’s volatility profile is a 12.76, this is calculated by the number of games played times the volatility of that game, then summing the totals (basically a weighted average).

Below is a spread sheeting showing the calculation Figure 1, Player's Volatility Profile:

Figure 1, Player's Volatility Profile

If we look at where 12.76 falls from the four samples above, it falls in line with the lowest number of games played (3,820). This becomes an issue as the lowest volatility game is 4.03 and the highest is 3.69 points away respectively.

From this, the players volatility scale needs to be created, where 12.76 is the center. From this number I recommend taking 33.33% and subtracting it from the 12.76, this becomes the lowest and add 33.33% to this number and this becomes the highest number.

Calculating Volatility Index On Slot Machine

Example:

Medium volatility index is 12.76
12.76 x 0.3333 (33.33%) = 4.25
12.76 + 4.25 = 17.01, this is the highest volatility
12.76 – 4.25 = 8.51, this is the lowest volatility
This player's volatility scale is a, 8.51 lowest, 12.76 middle, and 17.01 highest.

This example 'player’s volatility evaluation' is one of the hardest to target since they play a large spread of volatility index games on a consistent basis.

There is a second evaluating that can be done using volatility index numbers and this is more simplistic and targeted. When a new game is introduced to the gaming floor, the players data base can be evaluated to find anyone that plays that volatility index within a particular range decided at the time of the evaluation (example: 8.50 +/- 1.25 points). Then find players that play at least 30% of their time on these machines. Now you need to evaluate this on a collective basis of all machines in this range, not just a single machine.

Slot Machine Volatility

Either way it’s evaluated, you are now targeting your player, next is getting the information out to the player. Here is a simple form letter you can use. Note, the words in BOLD are information pulled from the data base to be entered into the form.

“Dear Mary,

Our Slot Director, Bob Smith just told me about a new game he put on the casino floor last week, it’s called, Pink Lemonade. Bob mentioned that you like to play Hot Balloons, Green Tomatoes and River Walk. Bob tells me that it plays very similar to these and has free spin bonus, random wilds, and is a cascade style game.

Slot Machine Volatility Index

This new game Pink Lemonade, is located in Section 4, row 23, location 5, yes…I know this is not a big help, so if you head towards the buffet, then to the right of the buffet its between there and the bingo room. Or just ask any of our team members forsection 4, row 23, location 5 and they can take you to it.

Bob was very persistent I tell you about this game, so we are loading $20 on your account infree play just for this machine, and we are loading into your account $10 to use on any machine including this one if you like.

As always we welcome your feedback and you can reach me and my staff at 'marketing email for slot related comments', we would love to hear back from you if you liked or disliked Bob’s recommendation.”

High Volatility Slot Machines

List of issues to consider.

Having the data of the games attributes to include volatility index, bonus features and game type/style in a data base.
Multi-Game machines are an issue. If your system can not record the play on the individual games on a multi-game machine, knowing what volatility index the player plays is not possible.
Players gaming habits, identifying them and properly targeting.
Local or tourist market. In a local market, this helps to keep the players interested, in a tourist market, this evaluation would not be as beneficial unless the player is a consistent player over a certain time period.
This can be used, when a player stays at the resort, and can be sent to them as information to remind them what games they liked last visit, and what is new since there last visit.
IT support should be involved to help evaluate how the data can be extracted from the system, creating a secondary data base if needed to merge the information and what reporting software to use to extract the data that is merged.
Slot department recording the data needed for the evaluation is limited, many may not understand how this information can be used, or where to find it. The information can be found in the follow information from the manufactures.

Par Sheet
Cut sheet (information from the manufacture, also known as a sale sheet)
GLI (Gaming Laboratory International) information sheet, or other regulatory data sheet.

This information has been used by me in my career and experiences. It has been written solely from my experiences having over 25 years’ experience in gaming related to slot machines management.

Darrin Pachman

Slot Machine Volatility Formula

Email: D.Pachman@EGMAdvising.com