Fair Rankings in Shortened Sports Seasons

Many professional sports leagues face the challenge of suspending games due to various reasons, such as the COVID-19 pandemic. When a league is ready to resume, a key issue is deciding which games to play to fairly conclude the season within a shorter time frame. This article discusses a method to select games that can help determine Rankings that resemble those if the full season had been played.

#Problem Definition

Suspensions of professional sports leagues can take place for various reasons. One recent example is the COVID-19 pandemic, which caused many leagues, including the NBA, to stop their seasons. When reopening, leagues must consider how to finish the season without playing all originally scheduled games. The main goal is to find a small number of games that can fairly determine the end-of-season rankings.

#Relevance

Most literature focuses on planning sports from scratch, but concluding a season that was already started presents a different set of challenges. This method aims to recreate rankings similar to those that would have resulted had the entire season been played.

#Methodology

This proposal uses data analysis to create a schedule that includes a selection of games from those that were originally planned. The approach involves the following steps:

1. Prediction And Planning: Using data analysis to forecast Game Outcomes based on previous matches.
2. Optimization: Selecting the best subset of games to play that minimizes the differences in rankings between the shortened season and the complete season.

We introduce a model that assesses team rankings based on the games selected, aiming to produce a final ranking as close as possible to one that would have emerged from a full season.

#Managerial Implications

The proposed framework allows leagues to manage shorter seasons with reduced numbers of games while still targeting competitive outcomes. For instance, it may help finish a season with 25-50% fewer games played, yielding similar rankings.

#Background on Sports Suspending

The recent pandemic forced many sports leagues, including the NBA, to suspend games. As a result, teams faced uncertainty about how to continue their seasons. Several options were considered:

1. Cancel the Season: All remaining games and playoffs are canceled, with a champion decided by vote.
2. Skip to Playoffs: Cancel remaining games and move directly to playoffs based on pre-suspension rankings.
3. Full Season: Resume play and complete all games before the playoffs.
4. Shortened Season: Select a few games to play before the playoffs begin.

#Shortened Season Consideration

Aiming for fairness is challenging when games are cut short. Some teams may have played fewer games or faced easier opponents by the time of the suspension, leading to unfair rankings. The focus here is on selecting a number of games that would allow for coherent rankings while maintaining the competitive integrity of the league.

#Key Concepts in Game Scheduling

When deciding which games to include in the shortened season, the model considers various factors:

• Team Performance: Analyze each team's historical performance before the suspension.
• Game Outcomes: Predict outcomes based on data from past games.
• Rankings: Measure how closely the end-of-season rankings align with what would have happened in a full season.

#Proposed Method and Models

This approach leverages a two-phase methodology that combines prediction and decision-making:

1. Predict Game Outcomes: Use historical data to create a model that predicts outcomes of the remaining games.
2. Optimize Game Selection: Employ optimization techniques to determine which games to include in the shortened season.

#Predictive Models

A variety of predictive models can be used to analyze game outcomes, including:

• Machine Learning Algorithms: Classify and predict based on historical performance data.
• Statistical Analysis: Assess the likelihood of outcomes based not only on win percentages but also on other features like home/away game performance.

#Prescriptive Models

Once outcomes are predicted, a prescriptive model helps select the games that will lead to the most desirable end rankings. This model can help ensure that the selections are fair and reflective of true team strength.

#Results from Previous Seasons

The proposed model has been tested against several past NBA seasons, evaluating its effectiveness in producing rankings that are comparable to those resulting from full seasons.

#Effectiveness of Game Selection

Using simulations, the model demonstrated significant success in maintaining accurate rankings. It was found that the selections made under this model often lead to rankings closely resembling those from seasons where all games were played.

#Practical Implications of the Model

The model not only offers a theoretical solution but can also be practically applied to conclude a season more fairly. The insights gained can help league managers make informed decisions about scheduling, ensuring that the end-of-season rankings benefit from the most competitive games.

#Application to the 2019-20 Season

Applying this model to the NBA’s 2019-20 season can illustrate its practicality. The league faced unprecedented challenges, and this approach provided a way to manage the shortened season effectively.

#Conclusion and Future Directions

The ability to select games that fairly determine rankings in shortened seasons holds significant value for professional sports leagues. This method not only addresses immediate needs but also sets the foundation for future research. Improvements could be made by exploring more sophisticated predictions and further optimizing scheduling processes.

#Summary

In summary, this article discusses a model for selecting games in professional sports leagues when a season is shortened. It highlights the importance of predicting outcomes and optimizing game selection to ensure fair rankings, using past seasons as a benchmark for effectiveness. This method can significantly assist sports managers in navigating the complexities of scheduling under challenging circumstances, such as those caused by the COVID-19 pandemic.