Sports Analytics Weekly by kubeia.io - 6/2024

Here at Smart Betting Club we showcase tipsters and services that can help you bet profitably.Whilst this is great, and propels our members towards success, it is not all plain sailing! Account restrictions bite if you make too much profit and bookmakers have come up with ever more ingenious ways to detect (and stop) winners.

The United States, Canada and Mexico will host the first 48-team edition of the FIFA World Cup in 2026 — and now we know where all 104 matches in the biggest knockout tournament in soccer history will be taking place.

When Jim Ratcliffe bought Manchester United, everyone understood it as a fixer-upper. United haven’t won a league title in over a decade and that doesn’t look like it’s about to change any time soon. Bad decision after bad decision has defined the club and ensured a downward trajectory. A club of almost limitless potential has been mismanaged in just about every way on and off the pitch. Old Trafford is in a joke of a state, not getting so much as a lick of paint since the Glazer family took over the club in 2005. Even the revenue isn’t as impressive as it “should” be, with United down to fifth place in the Deloitte Money League after finishing first every year from 1997-2004. Everything is bad. This hasn’t bothered the Glazers too much. I think it’s entirely gone to plan.Manchester United have been enshittified.

In the span of just four years, 1.FC Nürnberg W has grown from a regional side playing on municipal pitches into a top flight club going head-to-head with the likes of Bayern Munich and Wolfsburg. 

Reprising the last post on tactics and dice rolls and then a deep dive into what Randomness .. is(?)

Expected Goals (xG) has emerged as a popular tool for evaluating finishing skill in soccer analytics. Intuitively, the idea is dead simple. An xG model provides an estimate of the number of goals that would have been scored by an average player from a collection of situations, while controlling for shot location, shot type, assist type, and some other things depending on the model. Therefore, the difference between a player’s actual goal tally and the expected outcomes of their shots can be interpreted as an indicator of the player’s skill with respect to the “average player”. We refer to the resulting measure as goals above expectation (GAX).But, there is a problem with this approach. The value that a player adds via shooting does not appear to be stable on a season-to-season basis. A positive GAX in one season does not necessarily elevate the likelihood of a positive residual in the subsequent season. This leads to the troublesome implication that finishing ability is a random effect.

Expected Goals (xG) has emerged as a popular tool for evaluating finishing skill in soccer analytics. It involves comparing a player's cumulative xG with their actual goal output, where consistent overperformance indicates strong finishing ability. However, the assessment of finishing skill in soccer using xG remains contentious due to players' difficulty in consistently outperforming their cumulative xG. In this paper, we aim to address the limitations and nuances surrounding the evaluation of finishing skill using xG statistics. Specifically, we explore three hypotheses: (1) the deviation between actual and expected goals is an inadequate metric due to the high variance of shot outcomes and limited sample sizes, (2) the inclusion of all shots in cumulative xG calculation may be inappropriate, and (3) xG models contain biases arising from interdependencies in the data that affect skill measurement. We found that sustained overperformance of cumulative xG requires both high shot volumes and exceptional finishing, including all shot types can obscure the finishing ability of proficient strikers, and that there is a persistent bias that makes the actual and expected goals closer for excellent finishers than it really is. Overall, our analysis indicates that we need more nuanced quantitative approaches for investigating a player's finishing ability, which we achieved using a technique from AI fairness to learn an xG model that is calibrated for multiple subgroups of players. As a concrete use case, we show that (1) the standard biased xG model underestimates Messi's GAX by 17% and (2) Messi's GAX is 27% higher than the typical elite high-shot-volume attacker, indicating that Messi is even a more exceptional finisher than people commonly believed.

In the preceding article, YOLO Loss Functions Part 1, we focused exclusively on SIoU and Focal Loss as the primary loss functions used in the YOLO series of models. In this article, we will dive deeper into the YOLO loss function and explore two other interesting loss functions: Generalized Focal Loss (GFL) and Varifocal Loss(VFL).

If you’re wondering where ergodicity economics comes from, where it sits in the world of science, then this blog post if for you. I won’t go into detail about any of its results. Instead, I will focus on its origins and history. This history is superficial and selective. It is intended to hint at the culture of which ergodicity economics forms one small part.

The field of finance is deeply afflicted by experimenter bias, because of backtest overfitting, namely the usage of historical market data to develop an investment model, strategy or fund, especially where many strategy variations are tried on the same fixed dataset. Note that this is clearly an instance of the post-hoc probability fallacy, which in turn is a form of experimenter bias (more commonly termed “selection bias” in finance). Backtest overfitting has long plagued the field of finance and is now thought to be the leading reason why investments that look great when designed often disappoint when actually fielded to investors. Models, strategies and funds suffering from this type of statistical overfitting typically target the random patterns present in the limited in-sample test-set on which they are based, and thus often perform erratically when presented with new, truly out-of-sample data.

Scientific journals publish extraordinary results, so studies whose results are statistically insignificance aren’t published. Rather, they are abandoned or silently stowed away in academic filing cabinets.For this reason, this practice is called the “file-drawer effect,” and it’s a particularly insidious form of survivor bias because it is invisible. Peter Norvig sums it up nicely:When a published paper proclaims “statistically, this could only happen by chance one in twenty times,” it is quite possible that similar experiments have been performed twenty times, but have not been published.

Before we can make a prediction interval, we need to convince ourselves that prediction is possible. What data can we collect about the past to predict the future? We clearly need a model before we can do anything.

Have you ever wondered how to objectively and scientifically evaluate probabilistic predictions produced by statistical, machine and deep learning models?