In the lost post I raised the question "Can the ratio of fumbles lost to fumbles tell us anything significant?" but did not address it. Here is an initial analysis.
For an offense, the ratio of fumbles lost to fumbles indicates how often the team is able to recover its own fumbles. The lower the ratio, the better the team is at falling on the ball.
My theory has been that protecting the ball is a taught skill and some players, and therefore teams, are better than others. Consequently, some teams are more likely to fumble than others. However, fumble recoveries are random and depend on where players happen to be when the ball comes loose, who sees it first, and how an irregularly shaped ball bounces. Put simply, fumble recovery is chaotic and therefore not indicative of a team's future ability to produce or recover fumbles.
To test that theory, we can compare the correlations of fumbles stats, including fumble recovery ratios, to winning, points scored, and points allowed. If the fumble recovery ratio correlations are less than 0.15, they are not statistically significant (p=0.05) and this would be evidence that fumble recoveries are random.
The ratio is not significant for wins, points scored, or points allowed while overall fumbles and fumble rates are significant. This result indicates recoveries are indeed random. It also suggests that fumbles, and not fumbles lost is a better stat for estimating a team's future likelihood of fumbing.
- Home
- Unlabelled
- Fumbles and Fumbles Lost
Fumbles and Fumbles Lost
By
Brian Burke
published on 4/01/2007
Subscribe to:
Post Comments (Atom)