Eberlicious
Well-known member
- Messages
- 4,599
I was inspired by Hooper to test and see if players age had any impact on their overall regression, compared to their Potential. Donkey was kind enough to supply his player database from our previous iteration, so thank you Donkey as this would not have been possible. I am going to break down my process from start to finish and then at the end you will find links to my public Tableau in which you can interact with the data and come to some of your own conclusions as well. Here are the following steps I took.
1. My first step was to clean out all the irrelevant data points like college, draft position, salary, hometown etc...
2. Next I determined the letter grade associated with each rating for Inside, Outside, Handles, Defense, and Rebounding and assigned the corresponding grade to each player
3. Repeat steps 1&2 for years 2041-2050
Now that I have some appropriate data to work through I wanted to start drawing some conclusions. I quickly came to realize that since players who are getting older also will inevitably have lower potential as well. So I wouldn't be able to draw a proper conclusion to my original question I still wanted to use the data that I spent a few hours sorting through and cleaning. So I decided to compare 3 players, a guard, a wing, and a big, and their regression from their age 31 season until age 38. Below are my findings.
The bottom two rows for each player represents the correlation between a rating category and age or a rating category and pot. The highlights figure represents which one has a higher degree of correlation. Now only comparing 3 players, all from different positions is too small of a sample size to draw any conclusions. And compiling this data for every player in the data set would frankly just take way too long. Now I felt like I hadn't provided you fine folk in FBB with enough data or conclusions so I wanted to go a step further. I wanted to let you all go through the data and make some of your own conclusions as well, but excel is messy and not everyone is too savvy with it so decided to use a program called Tableau to present the data with greater visualizations and interactivity. However if you would like me to send you the excel files I have compiled I will gladly do so.
To create a proper Tableau workspace I needed to clean up the data further.
1. First, I needed to compile all the individual years into one sheet with every player at every age between the years 2041 and 2050
2. Second I sorted them by player name and then age to group them in a neat order and see them all formatted similarly to the screenshot above.
3. I then imported the data into the tableau program and began to create dashboards separated by the same attributes we used to use for coaching. (Athletics, Offensive, Defensive, Potential)
4. The parameter I set was a player must have at least 3 years of data out of the 10 year sample to qualify.
And now we have a finished product. I will post the link where you can go to my Tableau profile and look through each of the different dashboards and filter them by player and position and age. In the "Potential" dashboard I also added the letter grades of each player to see where they start and finish
Tableau Link:
1. My first step was to clean out all the irrelevant data points like college, draft position, salary, hometown etc...
2. Next I determined the letter grade associated with each rating for Inside, Outside, Handles, Defense, and Rebounding and assigned the corresponding grade to each player
3. Repeat steps 1&2 for years 2041-2050
Now that I have some appropriate data to work through I wanted to start drawing some conclusions. I quickly came to realize that since players who are getting older also will inevitably have lower potential as well. So I wouldn't be able to draw a proper conclusion to my original question I still wanted to use the data that I spent a few hours sorting through and cleaning. So I decided to compare 3 players, a guard, a wing, and a big, and their regression from their age 31 season until age 38. Below are my findings.
The bottom two rows for each player represents the correlation between a rating category and age or a rating category and pot. The highlights figure represents which one has a higher degree of correlation. Now only comparing 3 players, all from different positions is too small of a sample size to draw any conclusions. And compiling this data for every player in the data set would frankly just take way too long. Now I felt like I hadn't provided you fine folk in FBB with enough data or conclusions so I wanted to go a step further. I wanted to let you all go through the data and make some of your own conclusions as well, but excel is messy and not everyone is too savvy with it so decided to use a program called Tableau to present the data with greater visualizations and interactivity. However if you would like me to send you the excel files I have compiled I will gladly do so.
To create a proper Tableau workspace I needed to clean up the data further.
1. First, I needed to compile all the individual years into one sheet with every player at every age between the years 2041 and 2050
2. Second I sorted them by player name and then age to group them in a neat order and see them all formatted similarly to the screenshot above.
3. I then imported the data into the tableau program and began to create dashboards separated by the same attributes we used to use for coaching. (Athletics, Offensive, Defensive, Potential)
4. The parameter I set was a player must have at least 3 years of data out of the 10 year sample to qualify.
And now we have a finished product. I will post the link where you can go to my Tableau profile and look through each of the different dashboards and filter them by player and position and age. In the "Potential" dashboard I also added the letter grades of each player to see where they start and finish
Tableau Link:
