Bet on Yourself products can be the driving force behind introducing casual gamers to the world of esports betting, according to Max Polaczuk (pictured) – CEO of Sportsflare – but in order to engage and retain new players, sportsbooks must ensure that their odds are ‘solid’.
Speaking to SBC News, Polaczuk shared his views on the integrity challenges associated with account sharing (or smurfing), and why Sportsflare has taken a player versus environment (PvE) approach to its Bet on Yourself product.
SBC: Firstly, can you tell us a little bit about Bet on Yourself – what does this entail?
It does what it says on the tin – it quite literally allows you to bet on yourself to play your favourite video games and achieve certain skill challenges.
What we then do with our product, which is specifically designed for B2B, is that we add pricing, risk management and data components. So rather than be a B2C brand, we supply this product to sportsbooks – and we’re seeing an increasing number of media companies interested in our product too.
SBC: And what was the thought process behind creating Bet on Yourself?
We didn’t actually create the idea. I wish we did, but unfortunately I can’t take credit for that! There have been a few companies on the B2C side who have offered Bet on Yourself products, but they have been specifically for the head-to-head markets. For example, if I was to play against a friend for money, one vs one in a game, there are quite a few products where we could actually do that today. The winner gets the pot, and the house takes a rake effectively.
The challenges that they have there are really around queueing times and finding an appropriate person which matches your skill level. As you can imagine, this is quite a difficult model to run for many companies.
For us, what we do is the PvE style. So I can play when I want on demand. What that then means is that we have to be able to generate a fair price for the player who wants to play the game and achieve a particular outcome. That then removes any queueing time and stops you from needing to matchmake the players. But it does also mean that the prices have to be solid.
The reason we took this kind of approach and invested in the PvE style is because skill-based wagering is going to be the future. There are 250 million esports fans in the world, and more than 2.8 billion gamers. That’s a 10x larger segment to consider other than esports. And esports betting is solely focused to these esports fans, it doesn’t consider the gamers. That’s the main reason why we see a huge opportunity with this product.
SBC: How is Bet on Yourself helping sportsbooks to reach new demographics of players?
Right now, the stage we’re at is that we’re live on a site which has rolled out our BoY solution. Historically, they’ve offered peer-to-peer style betting which is where I can play against another player. Our solution has been an absolute game changer for them in terms of bottom line, as well as customer engagement and retention.
But effectively, it’s really to do with that casual gamer versus esports fan segment. You can play Fortnite or Call of Duty, but there isn’t really a developed esports scene around those titles. You might want to just play those games as opposed to being able to bet on yourself. We see this as an opportunity for sportsbooks to actually move casual gamers into betting.
One example is offering the Bet on Yourself product as a free-to-play game where you can get paid out in bonus bets etc. It can really form part of that funnel of moving casual gamers into esports betting, or even towards traditional sports betting.
We’re actually launching a sportsbook in a new country using this as the leading product to really tap into that younger demographic as well. You’ll have to keep your eyes peeled for that announcement in the near future!
SBC: From an integrity point of view, what are the challenges associated with ‘bet on yourself’ markets? And what measures does Sportsflare have in place to mitigate instances of suspicious betting activity?
The integrity is definitely a key thing! It’s a really interesting challenge though because unlike a typical sportsbook, where you have several people betting on one match, we only have one bet per match as the players are betting on themselves.
It’s different than seeing suspicious betting behaviours in the sense that at a sportsbook, everyone has a line that may be mispriced, and there’s a lot of liquidity on one side but not on the other.
The other thing that’s interesting is that you can bet on yourself to win. So firstly, you can’t bet on yourself to lose or score lower than a certain amount. This then means that issues such as match fixing are less of a problem than they are for traditional esports betting markets so to speak. That’s not to say that there are no integrity challenges though!
Cheating, account sharing, smurfing and hacking are all issues – however there are a number of game engines which do have anti-cheat systems built into them to try stop cheating specifically from happening, and we have a number of measures in place to protect our clients from those who may try to game the system.
The biggest challenge, which isn’t really against a game’s terms and conditions, is using someone else’s account to play. So for us, that is definitely our main focus of integrity. Can we therefore detect when someone else is playing on a low-level account? The way that we mitigate this is through a combination of algorithms and AI solutions which are running 24/7. They’re updating prices, updating limits, all of which are based on behaviours and a number of other factors.
Then the other side of the coin is that we also do the regular checks in terms of business intelligence. We manually look at customer profiles, we’ll dive into statistics and look through what players are doing, who are the sharps and who aren’t.
So in answer to your question, a combination of data analytics and real-time machine learning is helping to eliminate those risks of fraud.
SBC: So earlier on, you mentioned Fortnite and Call of Duty. Why did you choose to add Bet on Yourself to these games in particular? And can we expect any new games to be added to the Bet on Yourself roster in the coming months?
The reason we chose those games is firstly because they are both Battle Royale titles. To give a bit of context, in gaming and esports you have a range of titles: You have first person shooters, MOBAs, Battle Royale, there’s a whole range of game types.
With the sports simulation games, they are usually a one v one style – but with Fortnite and Call of Duty, these are more like one versus one hundred. So it makes it really exciting from a PvE perspective, and there are always games on. One vs one games don’t really make sense for a PvE style Bet on Yourself product. That’s why we picked these two games as a starting point.
Another reason for picking these two games in particular is that Fortnite has 270 million accounts, and Call of Duty Warzone has 100 million accounts. That’s a large range of games that have been played around the clock, and is a huge market to tap into.
That said, we’re adding top tier titles in the future such as Valorant, Call of Duty, Dota 2 – we’re working on those right now. So the goal is that we have as many games as possible, and that we can keep doing down the list in terms of market size and how easy it is to deploy.
SBC: How important has artificial intelligence been in ensuring that players benefit from the best odds possible?
With respect to artificial intelligence, it’s interesting because when we first started this, we thought that most of the effort would be placed on the pricing side. But actually, the real value is on the risk management side of things.
But unpacking the idea of players benefitting from the best odds, what happens if we cut the odds too low is that those players which are better are going to exploit these odds, and those which aren’t skilled esports players are effectively going to be beaten – and will eventually leave. As you can imagine, this is very unfair for those players.
We’ve seen some products which have been launched with a fixed odds model. So, for example, I get $2 for doing this certain challenge, you get $2 for completing that challenge. The issue with this model is that unless you’re exactly at that skill level, the prices don’t really make sense. If you’re worse than that level, you’ll end up leaving, and if you’re better then you’ll stay and exploit. So ultimately you get a business model which just doesn’t work.
With our pricing, it updates automatically based on the stats of the player, their history on the platform itself, the betting behaviours. So many factors are taken into account. For it to happen automatically, AI has to be used – especially given that so many games occur at once. You can’t really have a team of traders watching every single game that takes place!
The pricing component of AI is used to actually transfer player statistics, betting statistics and match history into a price that says how likely they will achieve certain challenges – whether that’s winning a game, getting a certain number of kills, whatever it might be. So that’s the first part of AI.
The more challenging, and probably more interesting part, is actually to do with the risk management side of things because what you get on these platforms is disingenuous people joining your platform and doing things like cheating, account sharing (which is also known as smurfing). So what they then do is try to get better odds than those currently on offer.
To combat this, what we’ve done is implement a range of algorithms which monitor things: could this player be cheating? could they be smurfing? AI can then come in on a number of levels – it then also looks at betting history. Is the player winning more? Are they winning when they shouldn’t be? Are they winning huge amounts and then losing little bits? It considers all of these patterns.
The other one that’s a lot more challenging, is that it comes down to whether we can look at this person’s in-game actions and detect if it’s a different person playing this match than the one before. There are a lot of of telltale signs in games, and it’s really one of the benefits of being able to get such granular data.