It started with team rankings that take the margin of victory and adjust for strength of schedule. Back in 2008, I developed an algorithm that makes these adjustments, and you can see these points based predictions on this page.
As my methods developed, I now combine predictions from a number of data sources to make the best possible predictions. For example, I use yards per play, a powerful efficiency metric, to evaluate football teams in both college football and the NFL.
I save these ensemble predictions for members of The Power Rank, as the NFL predictions went 53.1% against the closing spread during the regular season. However, you can get a sample of the NFL predictions by signing up for the free email newsletter.
To sign up for this free service, enter your best email address and click on “Sign up now!”
National Football League, Conference Championships
To check out my Super Bowl win probabilities prior to the start of the playoffs, click here.
1. Green Bay at Atlanta.
Atlanta (3) will beat Green Bay (4) by 2.7 at home. Green Bay has a 42% chance of beating Atlanta.
2. Pittsburgh at New England.
New England (1) will beat Pittsburgh (2) by 5.4 at home. Pittsburgh has a 35% chance of beating New England.
During the 2015 season, these college football predictions predicted the winner in 76.2% of games (583-182, doesn’t include cupcakte FBS vs FCS games).
Major League Baseball
National Basketball Association
To see my numbers prior to the start of the playoffs, check out the interactive visual for NBA win probabilities.
My rankings were a part of an academic study on making football predictions.
To check out my Euro 2016 win probabilities before the knock out stage started, click here.