By Roger Weber

 

Click here to download graphs

 

These graphs use 2002-2005 data to measure the changes in the standings over the course of a season. The first measures how close a team's record a certain number of games into the season is to its final record. For example it a team's winning percentage is 10% different from what it will be at the end of the year, its winning percentage is more or less 80% correct, because the average team will win 50% of its games. The second graph measures how many of the 30 possible places in the standings are occupied by the team that occupied it at the end of the year. After the first week only a few teams are in the same place in which they would end the year.

 

Both graphs follow clear patterns, although the patterns are opposite. It seems that teams' winning percentages quickly become steady and then level out, but the standings themselves are not shaped until the end of the year. So, making generalizations, we can say that teams following a certain trend toward a certain place will likely end up there. And gaps in the standings that are narrowing may close.

 

Click here to download graphs.