top of page
  • Writer's pictureKevin Jones

Henry Newman: C&N Sporting Risk - Understanding Data with in-play Sports Betting.

Gaming Eminence grabs a moment with Henry Newman Co-Founder of C&N Sporting Risk as we touch upon data infrastructure, modelling, algorithms, and the future of data science within the industry.

GE) In order for an Operator not to be at a competitive disadvantage, accurate and timely information is vital as it relates to in-play sports betting. When considering connecting with a real-time third party data provider, what technology infrastructure checks must both parties be aligned with to avoid a poor experience for the bettor?

HN) Technology infrastructure investigations are very dependent on the data being provided on the operator site. In the case of data where speed is paramount, the most important check is ensuring the live feeds that both parties are working off are synchronised. When Sporting Risk undertakes any connection projects with operators this is the very minimum requirement and the ultimate goal is ensuring the data is completely aligned in terms of the delivery time.

Another strategy to provide the best user experiences is never allowing the data to be delivered before any event or pricing update by the operator. Learnings from the In-Play prompts service we provide has led to us introducing a seven second delay on delivery of the prompts after an in-play event has triggered their generation.

This allows the event or odds to update on the operators platform and the user to digest the latest development. Then the user can read the bet-prompt that has been displayed and act accordingly on the operator’s platform.

GE) How much of a blend do you feel there is between predictive modelling and descriptive modelling, as it relates to sports betting currently? Are both being used in the industry and to what extent is it beneficial to the Operator?

HN) Much of the predictive modelling being deployed in the industry is related to user activity and trying to understand historical journeys. Once armed with this information, then it is easier to predict user behaviour and their subsequent betting patterns to the obvious benefits to the operator.

This empowers operators by either being able to adjust prices to reflect this expected behaviour, or adjust marketing and event prioritisation in line with this.

Descriptive modelling is more prominent in the content space. It helps describe statistical pieces of information that have occurred in the past, that have some predictive utility and are easily digestible by the user.

GE) Do you feel that data algorithms have variations based on the sport type in informing betting decisions? Could you give an example of why that must be considered for an Operator looking into different sporting events?

HN) In any predictive model, algorithms used will be very specific to that sport. For instance, a sport with a fewer number of players on the field at any given time would have much greater weight placed on individual player actions (rather than team outcomes), given the marginal significance of an individual player action on the overall outcome in competitive sports with few players on each team.

If you are an operator, you will want to be confident that this is in place and functions accurately to ensure that when there are significant pieces of team news released (e.g. a big player in a basketball game not being available), the algorithms adjust the predictive outputs of the model to reflect this.

GE) In the majority of sports and countries you will find numerous leagues/games to collect data from. How do you think third party providers determine what sporting events and markets to collect data from? Is that decision led by the engagement levels, a regulatory perspective, or something else?

HN) The sporting events and markets that data is collected from is led by engagement levels. Once the targets are established it’s up to companies to follow the correct legal framework set out by regulatory requirements and what data can be collected. A final layer comes from the contractual obligations of leagues to specific data collectors whom they agreed partnerships with.

GE) How do you see the development of data science and analytics being leveraged across the gambling industry over the next five years? What influence do you think Artificial Intelligence will play?

HN) There are two key developments that the gambling industry will be driven by during the next five years.

Firstly, the shift will come from behavioural observation of the consumer. Those who have a deep understanding of betting patterns, player preferences and their responsiveness to certain messaging will be best positioned to succeed. If this information is to hand then it becomes far easier to create betting experiences that truly resonate thanks to them being hyper-personalised to the individual as much as is physically possible. This is where AI capability will excel as it helps accurately track and record user behaviour.

A second element is the content provision from the operator themselves. Presenting users with more in depth data and analytics to inform betting decisions builds a much deeper engagement relationship between players and operators. This establishes the perfect foundation for a mutually beneficial relationship whereby insightful data nurtures brand loyalty for the operator and stronger user engagement for the consumer.

About our contributor

Henry has spent 12 years associated with professional football and utilising sports data to maximise performance. This has included co-founding a betting syndicate for 7 of those years, which was the brainchild behind C&N Sporting Risk; the In-Play Sport Betting Experts

The BetitRight team provide market leading stats and advice. We combine the most in depth statistical analysis with professional betting experience to provide market leading content to improve every type of punter. We are the brains behind the big bets. Visit


bottom of page