Here at Everydata, although we are big sports fans, we try not to spend too much time on sports applications of statistics because (1) not all of our followers love sports and (2) we don't want to be known for only being a sports stats source.
That said, when watching the Arizona-Green Bay game, I saw something interesting-a coin flip where the coin didn't flip. That was followed by an interesting article forwarded to us by our friend Richard about the evolution of coaches' decisions with regard to coin flips.
In the beginning of every game, the coin flip determines which team gets to decide if they want the football to start the game, or they would like to defer and get the ball to start after halftime. As the article chronicles, the conventional wisdom of NFL coaches used to be to always start with the football, but over the past 10 years, that wisdom has flipped (pun intended). Now, 8 out of 10 times coaches defer to the second half.
The full article from Monday Morning Quarterback explains this trend and the potential implications. But, one line from the article caught my eye: "The decrease in win percentage for deferring teams lends more support to the idea that the choice made at the coin toss doesn't lead to wins and losses." Given all that happens in a game, is it not surprising that the coin toss outcome is unlikely to be correlated with the final outcome? (Note, that is a different question than the overtime coin toss?) Also, is this saying there is no statistically discernible or significant difference? Given the small sample sizes, it wouldn't be surprising if you couldn't determine an effect that we can be sure is due to anything other than random chance.
Something to keep in mind as you enjoy watching the rest of this season's games.