Sports Analytics Weekly by kubeia.io - 39/2024

To sum: It's not the greed that wrecks you in gambling, it's the desperation.According to a complaint filed in a Brooklyn court, Porter, the former Toronto Raptor banned for life from the NBA for conspiring to fix results on prop bets involving his performances, had run up massive gambling debts and agreed to feign illness or injury to allow gamblers to bet the under on his stat lines and win what federal agents estimated to be over $1 million.

You may have seen this recent paper on the financial effects of legal online sports betting floating around econ twitter a few weeks ago:Gambling Away Stability: Sports Betting's Impact on Vulnerable Households by Baker, Balthrop, Johnson, Kotter, and Pisciotta (2024).The headline result: legalizing online betting increases betting by about $25 and decreases investments in stocks by about $50 per household per quarter.

Over the weekend, millions of Americans watched football. They cheered, they ate, and—more than ever—they gambled. The American Gaming Association expects $35 billion in bets to be placed on NFL games in 2024, about one-third more than last year’s total.If you follow sports, gambling is everywhere. Ads for it are all over broadcasts; more than one in three Americans now bets on sports, according to a Seton Hall poll. Before 2018, sports gambling was prohibited almost everywhere. Now it’s legal in 38 states and the District of Columbia, yielding $10 billion a year in revenue. [...] A growing body of social-science literature suggests that, yes, this is in fact quite different. The rise of sports gambling has caused a wave of financial and familial misery, one that falls disproportionately on the most economically precarious households. Six years into the experiment, the evidence is convincing: Legalizing sports gambling was a huge mistake.

We estimate the causal effect of online sports betting on households' investment, spending, and debt management decisions. Employing household-level transaction data and a staggered difference-in-differences framework, we find sharp increases in sports betting following legalization. This increase does not displace other gambling activity or consumption but significantly reduces households' savings allocations, as negative expected value risky bets crowd out positive expected value investments. These effects concentrate among financially constrained households, who become further constrained as credit card debt increases, available credit decreases, and overdraft frequency rises. Our findings highlight the potential adverse effects of online sports betting on vulnerable households.

Tactical fouls are a polemic subject within football. Many view them as a valid defensive resource but there are plenty more who see them as a cynical exploitation of the rules.They are also something that I find interesting to investigate from a data standpoint. Earlier this year, on our Spanish website, I analysed tactical fouls in La Liga; now, my attention turns to the Brasileirão, the top flight of Brazilian football.

Why study footedness? Footedness is relatively unexplored as a concept in football performance, yet there are key stakeholders

whose impact could be accelerated with a better understanding of this important area.

A player wins possession deep in their own half, and immediately, shouts echo from the sideline: “BOOT IT!ˮ… Up until the early 2000s, this was a common refrain at soccer matches across all levels and age groups. Back then, the only responsibility of a defender was to ensure goals were not conceded—passing skills were hardly a concern. Likewise, when a goalkeeper gained control of the ball or took a goal kick, the most frequent expectation was to "send it long"

In their first public pronouncements after formally acquiring Chelsea in May 2022, Clearlake Capital co-founders Behdad Eghbali and Jose E Feliciano pledged to “expand the club’s investment across infrastructure, technology, and sports science to support the incredible Chelsea football and commercial teams — all with the goal of leveraging this growth to fuel even more on-pitch success”.

A key element of Clearlake’s plan from the outset was to dramatically improve the quality and scale of Chelsea’s data analytics to inform every aspect of the sporting operation, from performance and injury prevention to player recruitment. That push has now led them to hire Sachin Gupta, executive vice president of basketball operations for the Minnesota Timberwolves and one of the most highly regarded analytical minds in the NBA.

Slowly but surely, this medium account is turning into a more meta-analysis place where I discuss methodology, coding and analysis concerning data specifically used in football. And, honestly, I love that. I always try to be innovative, but that’s not always the right thing to do. Sometimes you need to look back at your process and see if there’s something you can optimise or improve.That’s something I’m going to do today. I’m going to look at plain outliers in the data for specific metrics and what the case of outlier analysis tells us about the quality of the data analysis. Of course, there are a few problems that arise and I think it’s really good to take a moment and express worries about that.

Team play style is one of the most important factors in the success of a football team. At an elite level, the difference between obtaining European Football and missing out, or relegation and safety, can come down to a single point, so how a team defends, passes, and creates goal-scoring opportunities is crucial in gaining as many points as possible – meaning the role of team Head Coach is arguably the most important in football. A cliché among football pundits is “the best teams don’t have a plan B, they just make plan A better”, and so my analysis of team play style uses tracking and event data to determine how each team plays their own style of football introspectively, to gain an understanding of the number of the ‘plans’ each team possesses

In this paper, I introduce RisingBALLER, the first public transformer-based model trained on football match data to learn match-specific player representations. Inspired by advancements in language modeling, RisingBALLER treats each football match as a unique sequence where players act as tokens, with their embeddings conditioned by the specific context of the match. By leveraging masked player prediction (MPP) as a pre-training task, RisingBALLER learns foundational representations for football players, akin to how language models learn representations for text and vision models for images. I propose next match statistics prediction (NMSP) as a downstream task to demonstrate the utility of the learned player embeddings. The NMSP model outperforms a strong baseline commonly used in the community for performance forecasting. Additionally, I conduct a comprehensive analysis to explore how the embeddings learned by RisingBALLER can be applied to various football analytics tasks, including conditional optimal squad generation, player similarity queries, player and team style clustering, player-team fit estimation, and more. RisingBALLER is not just a machine learning model; it’s a framework designed to revolutionize football analytics by learning high-level

foundational player representations while accounting for the context of each game—a paradigm that has already transformed fields such as natural language processing and computer vision.

Matthew Trenhaile (or ‘Trench’) has worked in various roles within the industry for operators such as Mustard Bet and most recently, Pinnacle.

Alongside this, he has compiled odds for several sports, worked at spread firms (in both betting and finance) and developed an insight into the betting ecosystem that few possess.

Trench’s newest venture is a consultancy, Anubis Trading, and with the launch of his own business, he has been on somewhat of a ‘whistlestop tour’ of leading podcasts in Europe, North America and South America.

With that in mind, we did something a bit different for his appearance on the SBC Podcast, with an ‘Ask Me Anything’ special where you, the audience, posed questions.

Those questions and Trench’s answers can be found below with everything from sharp bookmaking, tipsters, moral questions and even ice cream covered. Enjoy!

Learn why calculating an exponentially weighted variance doesn’t yield a correct estimation of variance.

One of the most frequent questions I get from readers is “How do I win at Russian Roulette?”