Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
A research paper investigating the impact of analytic versus non-analytic decision makers on win percentage after in-season player acquisitions in MLB and NBA teams. The authors, Tyler Armijo, Xin Gao, Brandon Lovette, and Kenneth Siemers from the University of San Francisco, collected data on winning percentages before and after acquisitions, and the presence of analytic decision makers, to determine if there is a significant difference in win percentage between the two types of decision makers. The study found no statistical significance in the difference in win percentage after an acquisition, trade, or signing between analytic and non-analytic decision makers.
Typology: Study notes
1 / 15
This page cannot be seen from the preview
Don't miss anything!
Running Head: ANALYTIC VERSUS NON-ANALYTIC DECISION MAKERS Analytic Versus Non-Analytic Decision Makers and Their Effect on Win Percentage after In-season Player Acquisitions Tyler Armijo, Xin Gao, Brandon Lovette, Kenneth Siemers University of San Francisco
Abstract This paper examines whether teams with winning records that employ analytic decision makers see a higher increase in win percentage after in-season player acquisitions versus teams that employ non-analytic decision makers. This was accomplished through analyzing the in-season player acquisition data from the MLB and NBA over the last four completed seasons through descriptive and inferential statistics. There was not a significant difference between the two groups in average win percentage after total acquisitions, trades, or signings. Using inferential tests we found no statistical significance between the type of decision maker, and the difference in win percentage after an acquisition.
baseball using detailed performance rather than qualitative methods” based on simple statistics such as batting average (Beneventano, Berger, & Weinberg, 2012, p. 67). Sabermetrics do not have much value for traditional baseball statistics, but focuses on on-base percentage and slugging percentage (Beneventano et al., 2012). The most famous use of sabermetrics took place in 2002 when Billy Beane, general manager of the Oakland A’s, developed advanced models to predict player performance in order to determine which players were undervalued (Armstrong, 2012). Once added to the roster, these players allowed the team to go on a 20 game winning streak, and win a division championship with one of the MLB’s lowest payrolls (“Oakland Athletics Team History,” 2015). Since this historic accomplishment by Beane, teams within the MLB and NBA have adopted similar tactics using advanced statistics to inform personnel decisions during the season. The focus of this paper will be to determine if analytic decision makers in the MLB and NBA see an increase in winning percentage in all games before and after acquiring players during the season. Only teams that had winning records over the past four seasons will be considered in order to control for teams that were not actively seeking to win games. It is hypothesized that those teams who employ analytic decision makers will be found to have a higher winning percentage after player acquisitions than teams that do not. An “analytic decision maker” will be defined as someone who makes the final decision on player acquisitions, and who has a connection to using advanced statistics through their educational background or through the employ of staff with such a background. This decision maker will have no more than five years of professional playing and coaching experience in the MLB or NBA. Player acquisition refers to player signings, and trades that take place during the regular season.
In research conducted at the University of Minnesota, the merits of clinical and mechanical judgment were analyzed and compared in order to quantify which way of decision making outperformed or underperformed relative to the other (Grove, Zald, Lebow, Snitz, & Nelson, 2000). Mechanical judgment was defined as any decision made based upon statistics or computer programs, and clinical judgment was defined as formal decision making based upon subjective methods (Grove et al., 2000). The findings of this study suggested that mechanical judgment outperformed clinical judgment on average, and was therefore equal to or superior to clinical judgment in most situations (Grove et al., 2000). One limitation of this study is that it was conducted over a decade ago, however the current trend in the NBA is to hire staffs that have more familiarity with advanced statistics (“NBA Teams That Have Analytics Departments,” 2014). This trend seems to support the long-standing notion of the validity of using quantitative data or advanced statistics to make informed decisions. In the MLB most teams are in favor of hiring general managers that have a more statistical approach regardless of that person’s previous playing or coaching experience (Wong & Deubert, 2011). Each team may vary the extent to which it makes use of statistical analysis, but the new wave of decision makers within the NBA are more open to this approach than their predecessors were (Wong & Deubert, 2011). Our paper uses the term “decision maker” to signify the person with the final say over player acquisitions because the president of basketball operations may supersede the general manager in these decisions as the hierarchy of front office positions vary within each NBA franchise (Wong, & Deubert, 2011). One thing that Wong and Deubert (2011) make clear in their analysis of NBA general managers is that many have had some form of playing experience at either the collegiate or professional level. This fact, in
a lack of evidence to support the notion that using analytics to acquire players is conducive to increasing winning percentage. Our research will be a step in this direction, but further analysis of specific frameworks used by analytic decision makers in specific player acquisitions would be the next step in the process of determining the benefits of using analytics versus more traditional methods of decision making. Methodology Research Design We gathered cross-sectional data to determine whether analytic decision makers have a higher increase in winning percentage after player acquisitions during the season compared to non-analytic decision makers. The presence of the type of decision maker for each team was the independent variable, and the change in winning percentage after any player acquisitions was considered the dependent variable. Data to Collect The data we collected were the winning percentages before and after an in season player acquisition of teams with a winning percentage above .500 at the end of each of the previous four MLB and NBA seasons. We took the winning percentage the day before any acquisition and then the winning percentage for the rest of the season. Acquisitions were further categorized into trades and signings. If a team made multiple trades we considered that as one acquisition, and used the first one to mark the point from which we measured the “after” winning percentage. We needed to see how many analytic decision makers were in each league for the past 4 seasons. We then determined which league had more analytic decision makers. Analytic decision makers are defined as general managers (or presidents of basketball operations) by
educational background, and whether or not they employ any advanced statisticians. They were not considered analytic decision makers if they had five or more years of professional coaching or playing experience. The goal of this data was to see if there was an increase or decrease in winning percentage when a team with an analytic decision maker made a move during the regular season. There may be other factors that affected winning percentage such as schedule, or the quality of competition, but we felt that those factors were roughly equivalent for teams at the end of the season. Method of Data Collection The method of data collection for our research was a secondary data content analysis. We collected data regarding player acquisitions, winning percentages, and decision makers from professional sports data websites such as www.baseball-reference.com and www.basketball- reference.com. We also made use of the official websites of the MLB and NBA. We conducted web searches on each decision maker of teams that fit our criteria in order to determine their relevant educational background, playing or coaching experience, and whether they hired analytic staff. This was done for each decision maker over the past four seasons. Population and Sample Size The population for this research project was every MLB and NBA front office over the last four completed seasons (2010-2014). Each franchise counted as a separate entity on a year- to-year basis because of possible shifts in philosophy within its front office (for example the 2013 and 2014 Padres have no relation to each other and are different members of the population). This study sampled the population using stratified random sampling. Only teams
is no significant difference between the type of decision maker and the difference in win percentage after an acquisition, trade, or signing. We tested the effectiveness of the decision maker in terms of win percentage by league using the same inferential tests, and found that there was no statistical difference between the two variables for either the MLB nor the NBA. In order to try to explain these results we tested whether or not more acquisitions were being made by one type of decision maker over another during any particular season. The average number of acquisitions for each group can be seen in Figure 1. The averages are similar except in the 2014 season, when analytic decision makers had a higher average number of acquisitions. There was no statistical significance (p=0.695) between the type of decision maker, and the number of acquisitions made per season. Figure 1. Average number of acquisitions per season Discussion
In this research project we sought to explore the possible relationship between two variables: type of decision maker, and the difference in winning percentage following an in- season acquisition. The focus of the study was to try to provide some insight into whether or not one form of decision making was more effective over another during the course of a regular season. As opposed to a stakeholder analysis (Martinez & Martinez, 2011) or a study of general managers (Wong & Deubert, 2011) we opted to gather secondary data to see if there was hard evidence to support either form of decision making. It was originally hypothesized that teams with analytic decision makers would see a higher increase in win percentage after an in-season acquisition. This stance was based on the study conducted by Grove et al. (2000), which found that a quantitative approach to decision making was equal to or superior than a qualitative approach to decision making. In addition to Grove’s study, the success of MLB’s Billy Beane and his methods of using analytics to build a roster led us to believe that other decision makers across the MLB and NBA would be able to capitalize on using analytics during the season (Armstrong, 2012). However, our findings did not support this notion. At this point we cannot convincingly argue for one form of decision making over the other. It may be possible that the newly acquired player may need time to adjust to new team and get use to play style of said team. Maybe the moves made by decision makers during the off season are making an impact rather than the one during the season. Analytics may have more of an impact in the MLB than in the NBA and therefore canceling each other out. Limitations and Recommendations This research project was limited by the amount of access that was available to information regarding the inner workings of every NBA and MLB front office. There was no
Conclusion The purpose of this research was to find out whether there was significant evidence showing that teams with analytic decision makers experience a greater increase in win percentage after in-season player acquisitions compared to teams without analytic decision makers. All player acquisitions over the past four MLB and NBA seasons were examined in this study. Effect on win percentage was deciphered by calculating the win percentage before the team made the acquisition and subtracting it from the win percentage after the team made the acquisition. Before the study, we believed there would be significant data supporting analytic decision makers having a significantly more positive influence on win percentage via in-season acquisitions (trades and signings) and number of acquisitions made than non-analytic decision makers. However the data did not support our hypothesis. Of the four areas studied (player acquisitions, trades, signings, and number of acquisitions), there was no significant data favoring either analytic or non-analytic decision makers. The results were surprising because they do not provide evidence for the current trend of hiring analytic decision makers. This study’s findings are intriguing because, in relation to in- season acquisitions, we would expect teams to see a similar level of improvement after acquisitions regardless of the type of decision maker they employ. This is very important information for owners to know when deciding on a decision maker because if they are hiring based off of the criteria of someone who can make key in-season acquisitions that will help the team down the stretch run towards the playoffs then they don’t necessarily have to focus on hiring someone analytical. This broadens their options significantly and can allow the team to feel more confident hiring a non-analytic candidate who is an all-around better fit for the organization.
So why is it that team’s continue to hire analytic decision makers? For the four years analyzed in our study, teams with analytic decision makers on average had a higher win percentage at the end of the season (.600) than teams with non-analytic decision makers (.584). Seeing as no tests were done in this study to measure the significance of the end of the season win percentage data, this would be a very interesting focus for future research. It may turn out that the effectiveness of analytic decision makers is most greatly noticed when examining acquisitions made in the off-season, which could be the reason this type of decision maker continues to get hired more frequently. There is still plenty of research to be done in this field and it will be interesting to examine how future studies compare to this one.
