This exercise stages a data-analysis task within the world of US men's college basketball. The National Collegiate Athletics Association (NCAA) Division 1 Men's Basketball Championship Tournament, known as ""March Madness,"" begins every March with 68 college teams and concludes in April with 1 champion. It is one of the biggest sporting events of the year in the United States-a multi-billion-dollar endeavor that is wildly popular with fans and that brings tremendous energy to the specific schools involved. The tournament, which is played across three weeks, begins with a ""First Four"" play-in round where 8 teams play four games to bring the field down to 64 teams. Two of the First Four games are played between the 4 lowest-ranked teams, and two are played between the four lowest-ranked ""at-large"" teams. Then a six-round single-elimination tournament is conducted with the 64 remaining teams, with two rounds played each weekend. The last weekend is known as the ""Final Four."" Each year, a selection committee comprising university athletic directors and conference commissioners chooses which teams will participate in the tournament and then seeds the teams-that is, ranks them from best to worst to assign matchups. But how the committee chooses to define ""best"" is imprecise. Students are tasked with examining the provided dataset, which contains information on prior selections and subsequent team performance, and then determining what past committees have prioritized in selecting at-large teams. Students must then take on the role of advising the committee on how to proceed with its next selection round. Does the committee get it right, or should it do things differently going forward?
In many cases, data analysts are given the question to answer or the data set to be analyzed. It's important, however, to understand that there are numerous steps, before and after the empirical analysis is done, that define an overall rigorous scientific process. This note outlines the basic steps that are important in asking an empirical research question, answering it, and presenting findings. The process described in the note generalizes to any empirical research domain. Given the rising prominence of data analytics in sports, we illustrate the process here using a question that is often discussed by basketball observers: Does calling a timeout in basketball end a run being made by the opposing team? At Darden, this note is used in the MBA and Executive MBA class "Data Analytics and Leadership Judgment in Sports"; it would also be suitable in many data analytics courses or a module in any course introducing the basics of an empirical research process.
In this case, students take on the role of the president of Kansas State University (K-State), a public university located in Manhattan, Kansas. In response to an offensive tweet from a university student, many Black athletes at K-State are demanding action be taken against the student, saying they will not participate in any sporting events until something is done. Other stakeholders at the school are opposed to the athletes' actions, arguing in defense of free speech. A public response cannot be delayed any longer. Acting as the president of the university, students must decide what to do. Students are therefore asked to read the case in class and immediately write down their answers to questions about how they view the students' demands, about what the university president should decide and do, and about how best to communicate with the various key stakeholders. Ultimately, whatever is done next will be highly scrutinized and will likely have long-lasting consequences for the athletics department, the university as a whole, and the president's and potentially others' individual reputations. This makes the case useful for helping students simulate real-time, concrete decision-making about difficult situations with major implications.
This case set follows Alex Stewart, who has built and runs a green energy development firm in the United States. After finding success finding and using underutilized rooftops to erect solar panels and wind turbines in major US cities where the existing power grid couldn't meet growing demand, Stewart is now looking for similar opportunities in Europe having spent the last two years doing research, forming the necessary networks and partnerships, and beginning the long process of due diligence and negotiations with countless public and private stakeholders. However, a person who had become Stewart's major competitor (someone he had shared the idea with because he assumed this person had no intention to do anything) has begun to show up either before or right after his team's meetings with local officials as a competing bidder. After this happened in three separate countries, Stewart knew he had a big problem on his hands. He hired an internationally renowned security firm, who advised him that while there were likely several sources to the leaks, including phone tapping and office bugging, the most certain one was the nightly trash. Students are asked to read Part A of the case-which describes a (disguised) situation that actually occurred-in class and immediately write down their thoughts about what Stewart should do, starting with whom he should (or should not) approach for advice. Part B presents a difficult conversation between Stewart and his wife about what to do next.
This case set follows Alex Stewart, who has built and runs a green energy development firm in the United States. After finding success finding and using underutilized rooftops to erect solar panels and wind turbines in major US cities where the existing power grid couldn't meet growing demand, Stewart is now looking for similar opportunities in Europe having spent the last two years doing research, forming the necessary networks and partnerships, and beginning the long process of due diligence and negotiations with countless public and private stakeholders. However, a person who had become Stewart's major competitor (someone he had shared the idea with because he assumed this person had no intention to do anything) has begun to show up either before or right after his team's meetings with local officials as a competing bidder. After this happened in three separate countries, Stewart knew he had a big problem on his hands. He hired an internationally renowned security firm, who advised him that while there were likely several sources to the leaks, including phone tapping and office bugging, the most certain one was the nightly trash. Students are asked to read Part A of the case-which describes a (disguised) situation that actually occurred-in class and immediately write down their thoughts about what Stewart should do, starting with whom he should (or should not) approach for advice. Part B presents a difficult conversation between Stewart and his wife about what to do next.
This exercise puts students into the role of an associate at a marketing and sales agency. They are asked to develop specific recommendations to present to some senior leaders about how the firm can and should become more truly inclusive and just. They also must be specific about the objectives and metrics they believe must be agreed to, and who should be accountable for them, in order to insure progress.
To prepare for each year's NFL draft, each team creates a "model" for player selection. This involves developing a holistic view of every player (especially those the team is most interested in), such that it can compare two different players. In this exercise, students are tasked with building and operationalizing (but not actually collecting the data or testing) a model for player selection in the NFL draft, including deciding on the dependent and independent variables.
In this exercise, students are put into the role of newly appointed general manager of an NFL team and must use the data provided to conduct draft research and make a tentative decision for their team's first four picks in the NFL draft. The data show the team's prospect analysis and include each prospect's grade on the 1-100 scale (and thus tier) and ranking among other prospects that play the same position. Their task is to consider all information summarized, review the information about the dozen or so best prospects likely to be available when they select in each of the first four rounds, and make their decisions about what to do in each round. The companion exercise (UVA-OB-1367) is from the head coach's perspective, and the two can be used in a negotiation scenario.
In this exercise, students are put into the role of head coach of an NFL team and must use the data provided to conduct draft research and make a tentative decision for their team's first four picks in the NFL draft. The data show the team's prospect analysis and include each prospect's grade on the 1-100 scale (and thus tier) and ranking among other prospects that play the same position. Their task is to consider all information summarized, review the information about the dozen or so best prospects likely to be available when they select in each of the first four rounds, and make their decisions about what to do in each round. The companion exercise (UVA-OB-1366) is from the newly appointed general manager's perspective, and the two can be used in a negotiation scenario.
The game of baseball offers a team's manager relatively few tactical decisions to affect their team's chances of winning on any given day. The manager chooses the batting lineup and where the fielders are positioned; the pitchers who start the game and appear in relief, and to some extent what pitches are thrown; and if or when to attempt a stolen base or sacrifice bunt. Those decisions comprise all the levers in baseball a manager can pull to gain an advantage. Given this, the question facing baseball managers is how to optimize this limited set of decisions such that they maximize their team's odds of winning as many games as possible. In this exercise, student will measure how one of the strategic decisions a manager can make-whether to sacrifice bunt or not-affects runs generated or lost.
This case set challenges students to consider the upcoming free agency decision-making of an NBA team by conducting some basic statistical analyses of three possible targets, all of whom are upcoming unrestricted free agents-Danny Glover, David Bowie, and Mike Meyers. Each of these players could potentially fit the team's need to find a shooter and versatile defender at the forward position. Students are asked specifically to evaluate what each of these three players might be worth paying to acquire based on their past performances.
This case set challenges students to consider the upcoming free agency decision-making of an NBA team by conducting some basic statistical analyses of three possible targets, all of whom are upcoming unrestricted free agents-Danny Glover, David Bowie, and Mike Meyers. Each of these players could potentially fit the team's need to find a shooter and versatile defender at the forward position. Students are asked specifically to evaluate what each of these three players might be worth paying to acquire based on their past performances. The B case puts students into the role of a player's agent responsible for finding the best contract for their client.
This case set challenges students to consider the upcoming free agency decision-making of an NBA team by conducting some basic statistical analyses of three possible targets, all of whom are upcoming unrestricted free agents-Danny Glover, David Bowie, and Mike Meyers. Each of these players could potentially fit the team's need to find a shooter and versatile defender at the forward position. Students are asked specifically to evaluate what each of these three players might be worth paying to acquire based on their past performances. The C case puts students into the role of a team representative responsible for leading negotiations with players and their agents to get the best contract for the team.
Despite the important role that assumptions about intentions play in our judgments of others' actions and how we respond to these actions, most of us haven't thought much about the accuracy of the inferences we make, the ways our decisions about "what he meant to do" or "why she did that" might be unintentionally biased, or the potential negative effects that result when inferences about intentions affect how we judge a behavior. Nor have we thought about steps we might take to question or correct our initial inferences, or whether we should aim to set aside inferences altogether in judging and determining the consequences of a behavior. The goal of this note is to help you think about exactly these things.