The 2015 NFL season is nigh unto here. I’m so ready. Part of the reason is because I’m bullish about my Cowboys, but I’ll get to them in a bit.
Since I did a post like this last year, I’ll be reviewing how I did as well as prognosticating this year.
I actually did two posts last year. The first one, A New Season, is a good background of my methods and theories about the challenges of predicting the NFL. This year, I’ll do ten posts. This one, which will explain my methodology, and a post about each of the divisions. Plus I’ll sum it all up for those who don’t want the details.
I want to emphasize two main points. First, luck is way more important than many sportscasters, coaches, and players want to acknowledge. There aren’t that many plays in an NFL game, and there are only 16 in a year. Small sample sizes are prevalent.
One of the tools we can use to gauge how lucky a team is in a given year is to look at their Pythagorean record. Basically, this is a statistic based upon a team’s point differential. Teams that score lots more than their opponents tend to have better records than teams that don’t. Given all of the years of the NFL, we can determine what record should, on average, ensue from a given point differential.
Secondly, on average, 6 out of 12 playoff teams are not in the playoffs the following year because of regression to mean, bad luck, harder schedule, injuries, or just the whimsy of the NFL universe.
To anticipate these changes, I’ll look at what a team’s Pythagorean record was, their age, and the number of injuries they suffered last year.
Teams who exceeded their Pythagorean will tend to regress to the mean. Teams who are older will tend to have more injuries. I wouldn’t be surprised if older teams fade at the end of the year, though I’ve not seen studies on this. Teams that had a bunch of injuries will tend to regress to the mean, depending upon their age.
The other post was the breakdown of each team at And They’re Off. This is the post I’ll be referring to constantly to determine where I was right and wrong.
One more thing. I’ll be making a lot of references to DVOA (Defense-Adjusted Value Over Average), which is an advanced stat developed by Football Outsiders. Here’s an explanation how it works at DVOA. The main thing to note is that not every yard is created equal. A gain of 10 yards on 3rd and 20 is useless, while a 2 yard gain on 3rd and 1 is productive. DVOA looks at every play, figures out what the average result would be, then compares the actual result to that average.
In any case, it provides a single value to compare teams on offense, defense, and special teams. It’s an esoteric system, but for the purposes of this exercise, it means I can use one statistic consistently to help make comparisons. Last year, it suggested that going into the playoffs the Seahawks and the Patriots were the best teams in the NFL.
With that said, let’s get to it. I’ll start with the AFC East and the AFC North, because Pittsburgh and New England play tomorrow night.