Pulse

10star: pricing and risk management key functions

By William - 13 June 2024

The last year has seen a growing focus on pricing and risk management as a subject of conversation within the sports betting industry. Given the strength of headwinds that operators now face, their importance is only likely to increase.

Simon Trim, strategic consultant to market-makers and risk management specialists 10star, talks to G3 about how data-driven strategies should be shaping the industry.

What is the biggest trading challenge that you think operators face day-to-day, that they might not be giving enough attention to?

Put simply, risk management. Many sportsbooks do not possess the tools or information to know and react to what their book exposure is, which is a situation that has evolved through the “internet age” of sports betting.

A shift from pre-match to in-play precipitated a raft of additional market types made possible through pricing automation software, generating exposure for operators that is impossible to manage appropriately in the traditional, manual fashion. For many operators however, in-play was a new product vertical generating incremental revenue from customers who were generally less price sensitive.

Rather than try and manage this additional exposure – which requires complex automated risk management – many sportsbooks chose to offer in-play pricing as generic content, pushed to a soft customer base through promos and free bets, concentrating their risk management efforts on restricting all but the most “recreational” customers. The issues with this are now coming home to roost.

In a world of vastly increased cost of operations and against a backdrop of tighter regulation, sportsbooks are reliant on pushing poorly priced but high-margin products such as SGPs and don’t possess the capability to differentiate from all the other lookalike sportsbooks. The problems with this are two-fold.

Firstly, whilst SGPs are boosting profits in the short term, they are also leaving many sportsbooks as sitting ducks of a catastrophic payout due to the large, unknown liabilities they are running up on an outcome that will eventually land – but they have no means of mitigating because they don’t have visibility of them.

Secondly, the poor value that these bets create for customers means that recreational players burn through their disposable betting income more quickly and churn more readily. For any regulated market there are a finite number of potential users and as these churn it leaves operators fighting over fewer customers but with no means to differentiate their product.

Automated risk management can solve all of these problems by managing liabilities correctly, producing better prices for sportsbooks – that are also in-tune with more efficient marketing efforts – and allowing them to differentiate from the competition by having the confidence in laying a price that allows them to be different.

This focus on pricing, trading and risk management as specialist functions is core to the ethos of 10star and is how we are able to generate higher profits for our industry partners than other existing B2B solutions.

Data-driven strategies are shaping the future of the industry. What is 10star’s approach?

The saying ‘data is the new oil’ was first coined around 20 years ago, reflecting the fact that data would be the new wealth creator for industry in the 21st century in much the same way that oil had been since the mid 19th century.

However, there’s also a parallel in how value is created, because in their raw forms, neither oil nor data are worth as much in their unrefined states. For operators, how they extract and monetise data that they possess is becoming increasingly important.

It’s estimated that the volume of data in the world will double every two years, meaning that operators are increasingly going to require powerful tooling to understand what the data they possess is telling them and how they should react.

At 10star, we have a deep history of using data for both real-time pricing and trading decisions and historical analytics. We’ve combined these skill sets together to release a first-of-a-kind suite of SaaS analytic tools for sportsbooks, called neural, which filters all aspects of an operator’s betflow to surface insights such as customer sharpness, strength of supplier models, trader performance, margin sustainability and customer welfare.

Since 10star entered the market in 2023 there’s definitely been a growing focus on pricing and risk management as a topic of conversation. Why do you think the dial has moved like this?

In many respects, this can be summed up in the saying “you don’t know what you don’t know”. The way that the industry has evolved over the last 20 years has meant that odds compilation – i.e. the ability to make a price without an industry price to copy from – has become a dying skill.

Very few firms in the world can now be regarded as price setters, with Pinnacle being the global line that virtually all sportsbooks refer to in some form or another for pricing information.

10star shares a joint owner with Pinnacle and, through Jasis, has been a supplier of tennis pricing to Pinnacle for nearly a decade, so we share a lot of trading practices, methodology and ethos when it comes to best- in-class pricing, trading and risk management practices.

Since 10star launched there’s been a lot of noise from existing suppliers in respect of their risk management expertise, even though these suppliers do not come from a trading background and have never run risk on their own book.

Given the strength of headwinds that operators now face, and the importance that price differentiation driven by risk management will play in overcoming them, it is understandable that incumbent suppliers want to position themselves as experts in the field in order to stay relevant in a changing market, even though they’re lacking in more than 25 years’ experience of trading skill and have no means to back the claims up.

How challenging is it to separate the signals from the noise when it comes to pricing?

A common approach for operators offering sports betting is to buy in pricing from content suppliers who copy and track where the market is trading. Operators who haven’t got knowledge of how these prices are being created generally limit deviation away from the market and manage exposure by restricting more knowledgeable customers from placing bets.

This is a business model that requires very little in the way of processing real-time information, let alone deciphering what signals you need to listen to. However, as a price setter and a market maker, being able to react to relevant information is crucial for our business, and by consequence is a huge competitive advantage for the partners of our services.

Fitness, form, mentality, motivation, rule changes and such like are all elements contributing to a team or player’s likelihood to prevail. Aside from a leading team of compilers and traders, 10star is unique in the way we extract relevant information from bet flow and automate this back into price generation.

It is a unique business model based on decades of market making experience, incorporating techniques of portfolio management that are closely aligned to financial fund management. Being able to decipher how you should react to information contained in an anonymised data set is challenging enough, and in itself is a huge barrier to entry.

However, beyond that, you need to build the models and infrastructure that can incorporate and process this information with zero latency and that requires a very different technical stack to the basic algorithmic models producing generic derivative markets that are prevalent elsewhere.

Incumbent suppliers are poorly equipped to pivot to models that can react to this proprietary information – even if they can work out a way to distill it – which is why price followers cannot become price leaders quickly, if at all.

Your integrated alpha solution claims to offer higher profits and efficiency for operators. How does it accomplish this?

In financial terms, generating alpha refers to generating profits in excess of the rest of the market. Generating alpha isn’t the same as a supplier offering a service that improves returns by 10 per cent if the existing solution is 10 per cent worse than the market to start with – this is simply achieving parity.

10star generates alpha through a variety of methods, which either have high barriers to entry – such as the model and data inference processes already described – or through the asymmetric information captured through our proprietary trading processes that can’t be replicated elsewhere.

This combination of advanced modelling, unique data analytics and greater knowledge of market liquidity creates an enduring best-in- class proposition for our partners.

10star’s founding team has its roots in Jasis, a tennis betting syndicate. Why is this relevant to your operation now and how does it set 10star apart?

Alongside supplying pricing on a B2B basis to operators such as Pinnacle, Jasis is also a dominant market maker on the exchanges, so we are experts in dealing in zero latency, low margin trading and handling anonymised liquidity. To do this successfully, you need to build models and understand the market in a completely different way to anyone else.

B2B supply means we must produce a raft of markets that always have to be available, and both the exchanges and Pinnacle have a “winners welcome” policy. The fact that they don’t kick out customers means our prices, trading and risk management expertise has to be on a level that is way above that which other B2B suppliers are capable of.

All of our models are highly automated, perfectly correlated and risk driven, meaning there is no weakness in any derivative and are also fully extensible across in-play. Put together, we feel we have an operation that solves problems for sportsbooks in ways that B2B solutions cannot.

Why do you think operators fail to make the most of their existing data?

If you go onto a modern trading floor it isn’t unusual to see some form of BI reporting tool surfacing objective reports from a sportsbook’s underlying data, and in respect of personalisation and customer management, some sportsbooks are using data in really interesting ways.

However, because of the way the industry has evolved in regard to attracting recreational customers and removing more skilled ones, there is no experience of modelling customers in the same way as there is of modelling sports.

Not only that, but modelling customer behaviour and being able to decipher meaningful predictive outcomes from it is a much harder task. This is why we’re seeing so much market interest in our neural product: we are able to surface outputs to operators based on betting data that is proprietary to them and which they already possess, but previously have had no means of accessing and monetising.

In turn, because these data outputs are incorporated into our integrated alpha solution, we’re using real-time data processing in a way which is moving the profit possibility frontier for our partners, and ultimately for the industry as a whole.