ADS Capstone Chronicles Revised

5

performed exploratory data analysis, data quality checks, and feature engineering. The intended goal of these is to gather data that will aid in decision making and provide meaningful insights as it pertains to passing versus running in the NFL. 4.1 Data Acquisition and Aggregation The dataset was acquired through Kaggle (n.d.), an online and open-source data repository. The data collected consists of 22 seasons of in-game statistics across the regular and postseason for a total of 5929 records and 53 columns. Contained within the statistics are breakdowns of offensive categories by play type, such as passing or rushing attempt. Additionally, the data are organized by home and away teams to achieve separation of the two, as play styles differ between teams. 4.1.1 Exploratory Data Analysis In the analysis of wins by team type, it is evident most victories occur when teams play on their home turf, constituting 56.4% of all recorded wins. This statistic highlights the advantage home teams often enjoy, whether through familiarity with their surroundings or the support of their fans. Conversely, away teams still comprise 43.36% of the total count, which underscores the competitive nature of sports, where teams must perform well irrespective of the location of the game. Ties are relatively rare, making up only 0.24% of the outcomes, but they add another layer of unpredictability to the sporting

landscape. Figure 1 displays the win breakdowns by team type. Figure 1 Counts of Number of Wins by Team Type

Additionally, there are four scatter plots, displayed in Figure 2 and Figure 3, depicting the relationship between pass attempts and score for both home and away teams, as well as pass yards and score for both home and away teams. These scatter plots consistently reveal a similar trend: an increase in passing yards correlates with a higher score, regardless of whether the team is playing at home or away. There does not appear to be much correlation between the passing attempts with scoring, but this is likely due to the team with a higher score attempting to run the clock out towards the end of the game, while the trailing team becomes pass heavy. This trend underscores the importance of passing efficiency in contributing to overall scoring outcomes in football matches

32

Made with FlippingBook - Online Brochure Maker