No, not Oliver Luck. Turnover luck.
I spent a morning pulling TOM data on the last five years to see if there was a pattern or edge I could find that I could use for predicting future team win behavior. Here are the bullets I found that I will post on and feature in the 2015 digital preview magazines coming out soon.
The greater the turnover margin, positive and negative, the more likely and greater the regular season win change in the forthcoming season
Turnover margin is very random. Few teams can sustain a high or lower level of annual TOM.
Each team is ranked and patterned into predictive pools of trends. Teams with high and low TOMs the previous season have very high likelihoods of regular season win total changes.
As for his second point, check out the chart in this post from Bartoo:
… It always makes me laugh when someone calling a game ‘predicts’ anything about the winner and the turnover battle. We all know winning the turnover battle wins a lot of football games. Each coaching staff emphasizes it and coaches the hell out of the turnover battle. Offense and defense.
However, it is not so easy to predict. The results over the last 5 years are all over the board. There are a lot of teams with top coaching staffs at the top of the five year rankings, but it is difficult to have great results every year.
The teams in light green are the eight FBS teams that have had a positive TO margin each of the last five seasons. The teams in light red, those are the eight that have had a negative TO margin each of the last five years. The other 107 teams have had a mix of results.
The Oregon Ducks, the no. 1 ranked team in five year TO margin is the only team to have a double digit TO margin in four of the last five years. Northern Illinois and Georgia are the only other teams to hit positive double digits three of the last five seasons. [Emphasis added.]
You’ve only got eight teams out of 120 that have managed positive turnover margin in each of the last five seasons and eight that have done the same on the negative side. That strikes me as evidence that there’s some degree of randomness in the system.
And Georgia would have joined that first group, but for the disaster that was 2013, with its green defense and Aaron Murray having to carry the offense on his shoulders much of the season. Given what we’ve had to say about the coaching brain trust over the last five years, how much of that would you attribute to coaching and how much to statistical noise?