Happy Thanksgiving everyone! Have you ever thought that your team, despite a couple of losses, is really the best team in the league because they beat someone who beat the best team in the league? Take last year: perhaps you’re a faithful Patriots fan defending the team against its loss to the Colts. The Jets beat the Colts though (even if maybe because the Colts pulled their starters), and the Pats beat the Jets (actually they went 1-1, but they won by more than they lost!), so obviously the Pats could beat the Colts. If you’ve ever made this argument, then this graphic is for you:
If you read the graphic (found here) clockwise, each team beat the one to its right by the score listed; so Atlanta beat Tampa 27-21 and Tampa beat Cleveland 17-14, etc.
What does it mean? Well, at face value it means that any team can win on any given day and parity (as the name of the graphic implies) rules. Maybe a bit more reasonably, it shows us that game outcomes are at least somewhat random and even a favored team will lose sometimes. Home field is also important; yes, the Texans beat the Colts 34-24 to make their part of the circle complete, but that was in Houston; later, the Colts beat the Texans 30-17 in Indianapolis.
And this is part of the reason why it’s nice to have a model to predict games. No team is 100% to win, but not every game is a 50/50 coin flip either. Which means that even the Lions have a chance to win today! How do today’s games look?