LINECRAFT
Sophisticated, actionable insights for bettors of all levels. New episodes every Wednesday.
By subscribing, you opt-in to marketing communications from Novig
Matt Buchalter (known as @plusEVanalytics on Twitter) is the real deal. With a degree in Actuarial Science and Statistics, Matt's not just a guy on a heater - he's a number-crunching machine. He's dabbled in everything from poker (where he once ranked #1 worldwide in $100 single-table tournaments) to horse racing syndicates. Now, he's deep in the world of sports modeling and analytics, always on the hunt for that sweet, sweet edge.
In this episode, we explore the calculated gamble that is modern sports betting through the lens of a seasoned quant. Matt speaks to how data-driven approaches are reshaping the industry, from sophisticated modeling techniques to the challenges of market efficiency. Through Matt's experiences, a clear picture emerges of an industry in flux, grappling with the implications of ever-sharper players and evolving market structures.
Our conversation lands at the intersection of analytics, game theory, and good old-fashioned sports knowledge, all set against the backdrop of a booming industry that may be approaching a fork in the road.
Check out the full episode here:
Here's what we cover:
BetBash is an annual conference where the best of the best in sports betting gather and talk shop. This year, Matt presented on a topic that's gaining a lot of attention:
The topic I presented on is correlated parlays... There's all kinds of cool stuff you can do, both the mathematics of how correlated parlays work, as well as some material on where to look for them, what correlation even means, what it looks like, and how to most effectively find and execute on those plays.
Correlated plays are about identifying connections between different events that are more likely to occur together. For example, if a team's offense is performing exceptionally well, they might be more likely to win and cover the spread. By combining these related outcomes, skilled players can find edges that others overlook. It's a blend of sports knowledge and statistical analysis that can provide a pretty significant advantage if executed properly.
Developing live models comes with its own set of challenges, especially when it comes to testing their accuracy. Matt shares his experience with a tennis model he's created:
I have this tennis model and I have no idea whether it's terrible or excellent because I've never been able to get historical live odds to back test it against. I could be sitting on a gold mine or I could be sitting on a pile of crap, and I just don't have the time to actually test it in a live environment.
This highlights a huge hurdle in the world of sports modeling. Without access to historical live odds data, even skilled modelers like Matt struggle to verify if their work is valuable. It's like trying to practice your tennis serve without a court or a ball - you can go through the motions, but you don't know if you're improving or just developing bad habits. This lack of data potentially slows down innovation across the entire industry. After all, having a great model is only half the battle - you need to be confident it works before putting real money on the line.
Matt draws an interesting comparison between the current state of sports betting and the poker boom of the mid-2000s:
I think there are a lot of parallels to poker... We are really at the high point, at least in recent history, of sports betting as a thing that people are talking about and caring about, not just in our little niche world, but in the public as a whole. Will that last forever? Absolutely not.
From about 2003 to 2011, poker experienced a massive surge in popularity. It was featured prominently on TV, and there was a widespread belief that anyone could become the next big poker star. Sports betting is experiencing a similar moment now, with widespread media coverage and increasing mainstream acceptance. However, the poker boom eventually slowed down as the player pool became more skilled and casual players found it harder to compete. It's worth considering whether sports betting might follow a similar trajectory as the market matures.
As players become more sophisticated, it creates challenges for sports betting operators. Matt offers a balanced perspective:
The books have all kinds of lobbyists on their side. So it's only fair that players should have some as well, at least from an idealistic point of view... But nobody owes them anything. They are gaming a system. There's nothing wrong with that. I love gaming systems. That's what sharp players do. But this system doesn't exist to cater to sharp players.
This highlights the complex relationship between skilled players and operators. While sharp play can help create more efficient markets, it also poses a threat to operator profitability. As a result, many operators limit or exclude consistent winners. This practice is controversial, with sharp players arguing for their right to use their skills freely, while operators maintain they need to protect their business model. How this issue is resolved could significantly impact the future of sports betting.
The sharp player dilemma leads to a broader issue known as adverse selection. Matt explains:
This is a big challenge for centralized sports books. It's a big challenge for exchanges that have market makers, because if you're a market maker on an exchange, you're basically just a sports book. You have the same motivations, challenges, and opportunities that a centralized book would have.
Adverse selection occurs when a product or service attracts primarily high-risk or less profitable customers. In sports betting, this means operators might end up with a customer base dominated by highly skilled players who consistently find edges, while casual players (who are generally more profitable for the books) go elsewhere. This issue is closely tied to the sharp player dilemma and is one of the biggest challenges facing the industry as the overall skill level of players increases.