HotStreak’s Esports Same Game Parlays Change The Game
HotStreak’s recently-launched esports offering is unique in the industry—with more markets and unrestricted same game parlays. That’s the result of a deep stack of technology and machine learning under the hood to manage a complex and unique esports market.
There are two major differences between what Inplay Innovation offers and our competitors: In terms of markets, we offer more markets. We offer kills, assists, deaths, creep score, double kills, triple kills, quadra kills, penta kills and plan to add more soon. Our competitors offer kills and sometimes fantasy points.
Secondly, we offer unrestricted same game selections (what are sometimes known in the industry as bet builders or accumulators) in esports, while others do not offer them or have heavy restrictions on combinations of players offered in the same game. We have no restrictions on the number of players in a game and our payouts reflect this. Some competitors offer 2 players per team max. How are we able to offer esports same game parlays, when many other companies don’t or have them with heavy restrictions?
Inplay Innovation has invested heavily in understanding the specifics of esports and how they work. HotStreak has already built machine learning technology to analyze data and provide customers with a simple user experience and the best lines and odds. We’ve also built AI Trader, which is our AI-based risk management technology to automatically manage risk on individual markets and across markets.
Another way HotStreak is different: we have focused on individual players, not teams. For background: in the U.S., sports fans have grown up watching sports from a player-first perspective—the NFL and NBA are dominated by superstars, and fans have drafted their favorite players on their fantasy teams for years. This has led to explosive growth of player props in the U.S. We suspect this trend will hold true internationally as well.
Most gaming operators that offer esports have been focusing on team-based markets, so we see a huge opportunity to build the world’s best player-centric esports product by offering the most stat categories. Most DFS operators offer just kills, while HotStreak now has 8 markets with more coming soon.
One of the biggest factors in being able to offer unrestricted esports same game parlays is correlation. Correlation is well understood for traditional sports by customers and industry gaming operators. For example, it’s well known in football that the quarterback and the wide receiver are highly correlated in a relatively simple way: if the quarterback goes over on yards or completions, the wide receiver on his team would have increased chances of also going over on yards and receptions. As a result, many industry operators move odds to account for this correlation.
However, in esports, correlation is much less well understood for several reasons. For one, it’s relatively new and has received less attention than the major pro sports. Also, correlation in esports is deeper, wider and more complex than in traditional sports. In League of Legends for example, there is a relationship between virtually every player and every stat pair in a match. (See the graphic above)
Also, in League of Legends, the map is divided into 3 lanes and for a good part of the match everyone sticks to their own area as each champion tries to win their lane. You essentially have groups of players playing entirely different games from each other, with each of those game-within-the-games being highly correlated. It would be like a Lakers/Nuggets game where Anthony Davis is playing against Nikola Jokić on one net, LeBron James against Jamal Murray on another, and so on. That’s esports.
As a result, many other companies haven’t been able to figure this out yet and just handle esports correlation by using the sledgehammer of putting restrictions on entries for same game parlays, which is a bad experience for users.
Because esports is still relatively new, correlation may be harder to determine or perhaps the solutions employed by other companies for other sports are too brittle for esports because they’re entirely different. The beauty is our solutions are specific to each market in an entry. At Inplay Innovation we’ve worked really hard to precisely measure how correlated markets are with each other and have seen great results using our work to price entries.
To add a further complication, for correlation there’s even more mathematical complexity when you add multiple players on one slip—because the correlation isn’t always transitive. Just because A is correlated to B and B is correlated to C, doesn’t necessarily mean A is correlated to C. And then what happens when you add D, E, F to the equation, and they all have correlations? Pricing all this quickly with a good user experience is not easy. The math behind correlation is complex, but HotStreak’s solutions have elegantly addressed all these issues.
One other point on correlation: many other companies will only lower the odds for a customer when there is high correlation—such as quarterback and wide receiver yards—because obviously that increases the probability for a customer. But these same companies won’t increase the odds when there is inverse correlation—they’ll just keep the same odds, even though they know the odds should be higher. The reason: in the latter case inverse correlation helps the customer. At HotStreak, we always keep our customers in mind, so we strive to adjust the odds both ways, so that customers can benefit from inverse correlation.
In addition, in esports, there are a number of unique features to specific games. For example, in League of Legends the game duration can vary widely. It would be as if a baseball game could end in the third inning. But that is standard in LoL—which can affect odds and lines.
Patches, Nerfs, Buffs
The other big difference in esports is patches, the regular updates that change the game by balancing champions, items and mechanics. The goal is to improve gameplay, address any issues, and keep the game fresh by adding modifications that can have big impacts on game strategies and tactics.
For example, when a character or item is too strong, the game developers can make it weaker, which is a “nerf”; or when a character or item is too weak, they can make it stronger, which is a “buff’. These are all standard parts of an ordinary software update or patch. This can make certain characters more or less popular. And patches will change what champions are played in professional games—and also have a big impact on player performance in these pro games.
This is something that doesn’t typically happen in traditional sports. It also makes pricing esports markets an ongoing challenge.
For all these reasons, esports is a challenging market to price and offer odds and lines. But with a rabid fanbase, it’s one that we want to serve and serve more. We’re committed to building out more esports offerings and believe we have the technology and expertise to lead the industry.