The tasks that I cover are:<\/p>\n
- \n
- Exploring features of the data<\/li>\n
- Sorting the columns<\/li>\n
- Filtering the columns<\/li>\n
- Creating new features<\/li>\n
- Calculating summary statistics<\/li>\n
- Building pivot tables<\/li>\n
- Data visualization<\/li>\n<\/ol>\n
The data I use comes form the pybaseball<\/a><\/strong><\/span> library, freely available for install in python. I’ll be using the pitchers<\/strong><\/em> dataset from years 2012 to 2016.<\/p>\n
The entire jupyter notebook is accessible on my GitHub page<\/a><\/span><\/strong>.<\/p>\n
Libraries and Data<\/strong><\/em><\/p>\n
The libraries I use are:<\/p>\n
- \n
- pandas <\/strong>— for working with data frames<\/li>\n
- numpy <\/strong>— for additional computational support<\/li>\n
- matplotlib & seaborn <\/strong>— for plotting data<\/li>\n
- pybaseball <\/strong>— for data acquisition<\/li>\n<\/ul>\n
I called the data set pitchers<\/strong><\/em>. <\/em>The data consists of 408 rows and 334 columns. After doing a bit of exploring the data set (seeing how large it is, checking columns for NA values, etc), we begin by sorting the columns.<\/p>\n
Sort Columns<\/strong><\/p>\n
Sorting columns is done by calling the name of the data set and using the sort_values()<\/strong> function, passing it the column you’d like to sort on (in this case, sorting alphabetically by pitcher name)<\/p>\n
\r\n## Sort the data by pitcher name\r\n\r\npitchers.sort_values(by='Name')\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-3.49.27-PM.png\")
<\/a><\/p>\nIf you have a specific direction you’d like to sort by, set the ‘ascending’ <\/strong><\/em>argument to either True or False. In this case, setting ascending = False<\/strong><\/em> allows us to sort the ERA from highest to lowest (descending order).<\/p>\n
\r\n## Sort the data by ERA from highest to lowest\r\n\r\npitchers.sort_values(by = 'ERA', ascending = False)\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-3.50.12-PM.png\")
<\/a><\/p>\n(For additional sorting examples, see the GitHub<\/span><\/strong> code)<\/em><\/p>\n
Filtering Columns<\/strong><\/p>\n
Filtering can be performed by explicitly stating the value you’d like to filter on within the square brackets. Here, I call the data frame (pitchers) and add the .loc <\/strong>function after the data frame name in order to access the rows that are specific to the condition of interest. Here, I’m only interested in looking at the 2012 season. Additionally, I only want a few columns (rather than the 334 from the full data set). As such, I specify those columns AFTER the comma within the square brackets. Everything to the left of the comma is specific to the rows I want to filter (Season == 2012) and everything to the right of the comma represents the columns of interest I’d like returned.<\/p>\n
\r\n\r\n## Filter to only see the 2012 season\r\n# Keep only columns: Name, Team, Age, G, W, L, WAR, and ERA\r\n\r\nseason2012 = pitchers.loc[pitchers['Season']== 2012, ['Name', 'Team', 'Age', 'G', 'W', 'L', 'WAR', 'ERA']]\r\nseason2012.head()\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-3.57.03-PM-1024x285.png\")
<\/a><\/p>\nIf I only want to look at Clayton Kershaw over this time period, I can filter him out like so:<\/p>\n
\r\n\r\n## Filter Clayton Kershaw's seasons\r\n# Keep only columns: Season, Name, Team, Age, G, W, L, WAR, and ERA\r\n# arrange the data set from earliest season to latest\r\n\r\nkershaw = pitchers.loc[pitchers['Name']=='Clayton Kershaw', ['Season', 'Name', 'Team', 'Age', 'G', 'W', 'L', 'WAR', 'ERA']].sort_values('Season', ascending = True)\r\nkershaw.head()\r\n\r\n<\/pre>\n![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-3.57.57-PM.png\")
<\/a><\/p>\nTo make the data set more palatable for the rest of the tutorial, I’m going to create a smaller data set, with fewer columns (pitchers_small<\/strong>).<\/p>\n
\r\n\r\n## Create a smaller data set\r\n# Keep only columns: Season, Name, Team, Age, G, W, L, WAR, ERA, Start-IP, and Relief-IP\r\n# arrange the data set from earliest season to latest for each pitcher\r\n\r\npitchers_small = pitchers[['Season', 'Name', 'Team', 'Age', 'G', 'W', 'L', 'WAR', 'ERA', 'Start-IP','Relief-IP']].sort_values(['Name', 'Season'], ascending = True)\r\npitchers_small.head(30)\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-3.58.43-PM.png\")
<\/a>
\nCreating New Features<\/strong><\/p>\nI create three new features in the data set:<\/p>\n
- \n
- A sequence counter that counts the season number of each pitcher from 1 to N seasons that they are in the data set.<\/strong><\/strong>This is done by simply grouping the data by the pitcher name and then cumulatively counting each row that the pitcher is seen in the data. Notice I add “+1” to the end of the code because python begins counter at “0”. NOTE: <\/strong>To make this work properly, ensure that the data is ordered by pitcher name and season. I did this at the end of my code, in the previous step.<\/li>\n
- An ‘age group’ feature that groups the ages of the pitchers in 5 year bins.<\/strong><\/strong>To accomplish this task, I use a 3 step process. First, I specify where I want the age bins to occur and assign it to the bins<\/strong> variable. I then create the labels I would like to correspond to each of the bins and assign that to the age_group<\/strong> variable. Then I use the np.select()<\/strong> function to combine this information, assigning it to a new column in my data set called ‘age_group<\/strong>‘.<\/li>\n
- A ‘pitcher type’ feature that considers anyone who’s starter innings pitched was greater or equal to the median number of starting innings pitched as a ‘starter’ and all others as ‘relievers’.<\/strong><\/strong>To create the pitcher_type<\/strong> column, I use the np.where() <\/strong>function, which works like ifelse()<\/strong> or case_when()<\/strong> in R or like IF()<\/strong> in excel. The first argument is the condition I’d like checked (“did this pitcher have starter innings pitched that were greater than or equal to the median number of starter innings pitched?<\/em>). If the condition is met, the function will assign the pitcher in that row as a “starter”. If the condition is not met, then the pitcher in that row is designated as a “reliever”.<\/li>\n<\/ol>\n
<\/p>\n
\r\n## Add a sequence counter for each season for each pitcher\r\npitchers_small['season_id'] = pitchers_small.groupby(['Name']).cumcount() + 1\r\n\r\n## Create a new column called 'age_group'\r\n# create conditions for the age_group bins\r\nbins = [\r\n (pitchers_small['Age'] <= 25), (pitchers_small['Age'] > 25) & (pitchers_small['Age'] <= 30), (pitchers_small['Age'] > 30) & (pitchers_small['Age'] <= 35), (pitchers_small['Age'] > 35) & (pitchers_small['Age'] <= 40), (pitchers_small['Age'] > 40)\r\n]\r\n\r\n# create the age_group names to be assigned to each bin\r\nage_group = ['<= 25', '25 to 30', '31 to 35', '36 to 40', '> 40']\r\n\r\n# add the age_group bins into the data\r\npitchers_small['age_group'] = np.select(bins, age_group)\r\n\r\n## Create a pitcher_type column which makes a distinction between starters and relievers\r\npitchers_small['pitcher_type'] = np.where(pitchers_small['Start-IP'] >= pitchers_small['Start-IP'].median(), 'starter', 'reliever') \r\n\r\npitchers_small.head()\r\n\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-4.11.33-PM.png\")
<\/a>
\nCalculating Summary Statistics
\n<\/strong><\/p>\nThe GItHub repo<\/a><\/span><\/strong> for this post offers a few ways of obtaining the mean, standard deviation, and counts of values for different columns. For simplicity, I’ll show a convenient way to get summary stats over an entire data set using the describe()<\/strong> function, which is called following the name of the data frame and a period.<\/p>\n
\r\n\r\n## Get summary stats for each column\r\n\r\npitchers_small.describe()\r\n\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-4.23.31-PM.png\")
<\/a><\/p>\nPivot Tables<\/strong><\/p>\n
There are a few ways to create a pivot table in python. One way is to use the groupby()<\/strong> function for the class you’d like to summarize over and then call the mathematical operation (e.g., mean) you are interested in. I have an example of this in the GitHub post<\/a><\/span><\/strong> in code chuck 42 as well as several examples of other pivot table options. Another way is to use the pivot_table()<\/strong> function from the pandas<\/strong> library.<\/p>\n
Below is a pivot table of the mean and standard deviation of pitcher age across the 5 seasons in the data set.<\/p>\n
\r\n## Pivot Table of Average and Standard Deviation of Wins by Season\r\n# Round the results to 1 significant digit\r\n\r\nround(pitchers_small.pivot_table(values = ['Age'], index = ['Season'], aggfunc = (np.mean, np.std)), ndigits = 1)\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-4.27.43-PM.png\")
<\/a><\/p>\nYou can also make more complicated pivot tables by setting the columns to a second grouping variable, as one would do in Excel. Below, we look at the average WAR across all 5 seasons within the 5 age groups (which we created in the previous section).<\/p>\n
\r\n## Calculate the average WAR per season across age group\r\n\r\npitchers_small.pivot_table(values = ['WAR'], index = ['Season'], columns = ['age_group'], aggfunc = np.mean)\r\n<\/pre>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-4.30.07-PM.png\")
<\/a>
\nData Visualization
\n<\/strong><\/p>\n - An ‘age group’ feature that groups the ages of the pitchers in 5 year bins.<\/strong><\/strong>To accomplish this task, I use a 3 step process. First, I specify where I want the age bins to occur and assign it to the bins<\/strong> variable. I then create the labels I would like to correspond to each of the bins and assign that to the age_group<\/strong> variable. Then I use the np.select()<\/strong> function to combine this information, assigning it to a new column in my data set called ‘age_group<\/strong>‘.<\/li>\n
- numpy <\/strong>— for additional computational support<\/li>\n
- pandas <\/strong>— for working with data frames<\/li>\n
![\"\"](\"https:\/\/optimumsportsperformance.com\/blog\/wp-content\/uploads\/2021\/06\/Screen-Shot-2021-06-23-at-4.37.51-PM-1024x715.png\")
<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"