Reading Time: 5 minutes
by Lorenzo Nardiini, Head of Technical Compliance and Maths ComplianceOne
You have likely encountered problems with Random Number Generators (RNGs) and maths models for games of chance if you’ve ever been through the certification process. Both parts of the certification have probably slowed down the time to market.
This article will help you to understand the best practices and reduce risks by sharing with you some tips and best practice that I learned from my six years of experience as a math analyst.
The Start
Maths requirements, in principle, are a very small part of all the criteria that go into a gaming platform. They do not differ greatly between land-based and iGaming platforms. They can still be difficult to verify.
The Random Number Generator is usually the first step. It is important to note that, just as car manufacturers want to ensure the engine is working to enable the vehicle to move forward, game producers need to create unpredictable outcomes for the game that are fair to players.
RNGs can seem like a frightening topic that should be discussed with only a small audience. It is not always the case if you are well-versed in the technical standards that govern this component, and have a team of experts to help you.
Generally, regulators are looking for RNGs that can produce unpredictable numbers and do not introduce bias. In some markets, the requirement for unpredictability can be strict. Only secure RNGs are allowed (also known as cryptographically robust RNGs). RNG technology is constantly evolving and what’s considered secure today may not be so in the future. Cryptographic security has been extensively researched in the field of information technology, and its development affects many other areas than gambling: from Finance to Healthcare to E-commerce. You don’t have to be a crypto expert to use a good cryptographic RNG. Many operating systems include cryptographically secured sources of entropy. Different cryptographic RNGs are readily available. If you prefer a true source of random numbers, you can use hardware-based solutions which use the laws physics in order to produce truly unpredictable events. It is not recommended to create your own RNG algorithm, as it can be very difficult and time-consuming.
Tips for managing RNG certification
The following tips will help you focus your efforts on what you need to do to ensure compliance with your RNG and reduce the time it takes to get to market.
- Even if your current market does not require this level of security, you should opt for a RNG that is cryptographically secured from the start. The reason is that as technology advances, so do the requirements. More and more countries are following this trend. Once you obtain a license in your targeted market, you may want to expand to other jurisdictions which might require a RNG. This will save you a lot of time. You can ask some experts for help if you’re unsure of what algorithm to use.
- The most common errors I’ve seen are related to a component of RNG solutions that is much simpler: the mapping of final outcomes without bias. RNGs produce numbers that are usually in binary format (most commonly 32bit or 64bit). You will then need to include some functions to your code in order to be able to generate results in a particular format. It is common to include a range of outcomes (for example, 0-36 for a roulette with a single zero) and the option to shuffle the cards to produce unique results within a play. Experts can help you avoid these mistakes.
- Set up your system so that RNGs are encapsulated. It is important to only modify your RNG when it is necessary. Once it has been approved, it will be an integral part of Gaming Platforms. Once you have a reliable RNG algorithm and unbiased mapping and scaling methods, as well as, if necessary, dynamic monitoring, it is best to only update your RNG if technological advances require you to. Experts can review your RNG architecture to ensure you have a compliant RNG.
- Verify that your RNG has been tested for general usage. In many regulated markets, the RNG must pass statistical tests for all possible configurations in which it will be asked to generate random results by the gaming applications. You can avoid having to obtain updated RNG certificates every time you release a new product by testing your RNG in advance for many different scenarios.
Once your RNG has been approved, you can now begin the certification process for the games that are associated with it.
Maths in Games
The maths of a game can make the difference between a great game and a poor game. Maths is the place where you can implement new ideas to make your game more attractive to players.
To ensure compliance, all games must pass independent testing. Standard maths requirements are mainly concerned with the following: the minimum Return to Player allowed, the odds of certain outcomes, such as the top prizes, and the limitations enforced by regulators, such as maximum wins in a single game or the maximum loss an hour.
For you to pass these requirements, you will need a mathematical model that is accurate and reproducible. My experience has shown that insufficient/unclear data is the main cause of a math evaluation taking too long. Young suppliers often submit incomplete and inaccurate math documentation (PAR sheets) due to inexperience or to protect their intellectual properties. It is inevitable that this will lead to a long testing process with many email threads, JIRA tickets and other issues. My team is more than happy to revise these documents thoroughly before certification begins.
The same math model underlies all markets, so it is possible to find efficiencies by testing simultaneously on different markets. To do this, you must first have done a thorough compliance check for all requirements.
Conclusion
Preparation is the usual advice. A quick certification requires a high level of expertise, good maths documentation that is submitted to the lab. It is also important to remember that complexity is not always a good thing.
