Finding stats vs "Good" teams compared to "Bad" teams

Moderator: Doctor MJ

User avatar
Old Man Game
Head Coach
Posts: 6,281
And1: 4,317
Joined: Jul 15, 2012

Finding stats vs "Good" teams compared to "Bad" teams 

Post#1 » by Old Man Game » Sat Feb 20, 2016 1:11 pm

I apologize if this is a noob question, but does anyone know of an easy way to find a player's stats against teams of a given winning percentage?
altf4
Ballboy
Posts: 1
And1: 0
Joined: Mar 24, 2016
 

Re: Finding stats vs "Good" teams compared to "Bad" teams 

Post#2 » by altf4 » Sat Mar 26, 2016 3:11 pm

What might help your analysis is define a few questions that you would like to answer and also define what constitutes a "good" versus "bad" team.

If you're looking at a specific player then it is pretty straight forward: You can pull the player's stats from a variety of websites and then define a binary (0 or 1) or categorical variable (Good, OK, Bad) for all teams based on their win percentage. I see a lot on this board using Excel, so if that is the program you are currently using, you can simply define another column and use an 'IF statement' that takes into account a team's winning percentage or whatever metric you've used to define "Good" versus "Bad". With R or another statistical package, you're able to do the same by just defining a new column.

Why this might be difficult is because a team's status as "Good" or "Bad" can change throughout the season. What you might consider instead is comparing games where the player did well versus games where they did not. Taking this approach, you might find correlations between a certain opponent and the player's performance, or opponent characteristics that correlate with the player's performance.

Return to Statistical Analysis