I’ve recently added a Power Rankings chart to the front page. The concept is quite simple but there is a lot that goes into the formula.
First off, I look at 3 different time periods and assign different weights for each.
1. Entire season (50% weight)
2. Last 25 games (33% weight)
3. Last 10 games (16.7% weight)
The “Last 10 games” actually have more impact on the final number than just 1/6th since the last 10 games are included in both the last 25 and entire season.
Four numbers go into each of the 3 time periods….
1. Team’s Winning Percentage
2. Team’s Pythagorean Winning Percentage
3. The average Winning Percentage of their opponents
4. The average Pythagorean Winning Percentage of their opponents
I then simply just find the average of these 4 stats for each of the 3 different time periods.
Let’s say a team goes 8-2 and in their last 10 games, while scoring 45 runs and allowing 45 (Pythagorean win % would be .500). Their opponents over the past 10 games have an average winning percentage of .463 and an average Pythagorean winning percentage of .455. Combining these 4 numbers would give them a .5545 rating over the past 10 games. Even though they won 80% of their games, they didn’t outscore their opponents and their schedule was weak, so their rating is not as high.