I was in London this week for my spring break and, unfortunately, I came back with an unwanted souvenir, a minor cold. This week's analysis is going to be much shorter than usual as I am not feeling well enough to do an in-depth post. As I said in my first post, I wish to find a relationship between statistics and success, specifically, and most obviously, in the playoffs. For this week's post, I want to test the famous quote from football coach Paul "Bear Bryant, "Defense wins championships." Therefore, I want to test GF% versus number of wins. This statistic is to test whether a larger ratio between the number of goals scored and given up has any determination in a team's success during the playoffs. In other words, whether a team needs to have an overwhelming offense to bring home the glory or just needs to keep themselves marginally in front of their opposition. My analysis is below:
Let's break this down. As you can see by the graph and as there are pretty similar statistics as last week, this isn't a great relationship. However, if you look at the graph, you'll notice two teams were swept, yikes! These two teams, and one team that went out in 5, are obviously outliers and are ruining any opportunity to find a relationship, so I'm going to take them out. Now, let's look again!
Still not a great relationship as evidenced by the p=0.086, which is relatively high and the low F= 3.55. Moreover, the R-sq = 24.40% indicates that very little of the variation is explained by the model. Ok, I understand that you likely have no idea what I'm talking about. As a result, I'm gonna translate everything to English and use the graph to explain. As you can see in the graph, there is a lot of variation up and down. This means that it isn't easy to directly say that a higher GF% means more wins. If you notice the top of the graph, teams with 16 wins, the Pittsburgh Penguins, and 14 wins, the Nashville Predators, both did not have the highest GF%. In fact, there are a number of teams with mid-range GF% that won a large number of games during the playoffs. As a result, it is true, defense does win championships. Obviously, using last year's playoffs as a sample, it's not a matter of whether you're putting a large number of goals on the scoreboard, but instead, whether you're scoring more goals than the other team, even if the difference is relatively marginal. Thank you guys for taking the time to read my blog. Please provide any suggestions in the comment section or feel free to email me, jne264@stern.nyu.edu . Thanks and until next week, Jonathan!
I'm going to attach screenshots below. In short, I have a p-value that equals 0.020, an F= 6.83, r^2 = 32.78%, and y=-4.22+.2140. I will, however, attach the graph below to help better understand these statistics.
Let's break this down. As you can see by the graph and as there are pretty similar statistics as last week, this isn't a great relationship. However, if you look at the graph, you'll notice two teams were swept, yikes! These two teams, and one team that went out in 5, are obviously outliers and are ruining any opportunity to find a relationship, so I'm going to take them out. Now, let's look again!
Still not a great relationship as evidenced by the p=0.086, which is relatively high and the low F= 3.55. Moreover, the R-sq = 24.40% indicates that very little of the variation is explained by the model. Ok, I understand that you likely have no idea what I'm talking about. As a result, I'm gonna translate everything to English and use the graph to explain. As you can see in the graph, there is a lot of variation up and down. This means that it isn't easy to directly say that a higher GF% means more wins. If you notice the top of the graph, teams with 16 wins, the Pittsburgh Penguins, and 14 wins, the Nashville Predators, both did not have the highest GF%. In fact, there are a number of teams with mid-range GF% that won a large number of games during the playoffs. As a result, it is true, defense does win championships. Obviously, using last year's playoffs as a sample, it's not a matter of whether you're putting a large number of goals on the scoreboard, but instead, whether you're scoring more goals than the other team, even if the difference is relatively marginal. Thank you guys for taking the time to read my blog. Please provide any suggestions in the comment section or feel free to email me, jne264@stern.nyu.edu . Thanks and until next week, Jonathan!
Comments
Post a Comment