Does time of possession matter in American football?
Your first thought might envisage a mighty Alabama team running on each play, gobbling up chunks of 4, 5, 10 yards each time. Moving the chains, wearing down the opponent's defense. But is this helping to win?
One way to think about time of possession is like baseball's innings and their length. In baseball, each team gets up to nine half-innings to score some runs. With football, we might call a team's half-inning a drive. The team with the ball has a drive to score 3 or 7 points (usually). If a team eats up a lot of clock on a lengthy drive, both teams have a reduced amount of time possible with the ball. If a big-play team scores a touchdown on a mere three plays, then there is only a small amount of time reduced to both teams.
Effectively, the length of a drive's main effect is on the number of drives possible in a half. The better team should want as many drives as possible since this reduces the expected impact of bad luck for themselves and lucky breaks for the opponent; the variance in expected points should decrease leading to a more likely outcome of the better team scoring more points. Conversely, the worse team should want as few drives as possible. Anything can happen on one drive, but maintaining the luck is near impossible over several drives.
This concept has long been acknowledged in basketball. Teams such as North Carolina know they are usually superior and play a high-tempo style so that a few lucky three pointers against them won't sink their hopes. In football, some of the top teams stubbornly persist with the idea that longer time of possession causes more wins; this analysis suggests they would be even better if they adapted.
The bottom line is that coaches of stronger football teams should emphasize a high-tempo style, while weaker teams should focus on tactics that lead to fewer drives. They can do this by recruiting and game-planning accordingly.