Predicting 2013 NFL Standings

Predicting 2013 NFL Standings

Dallas Cowboys quarterback Tony Romo (9) warms up before an NFL preseason football game against the Houston Texans, Thursday, Aug. 29, 2013, in Arlington, Texas.

Predictions are important. They display understanding of a subject matter; it's really easy to go back and explain a past event, but it's another thing completely to accurately predict a future one. That's why I try to Cyber Monday Michael Kors Deals be as transparent as possible with all of mine (here's a review Cyber Monday Michael kors Bags of my 2012 predictions). When I propose my final 2013 standings, I do so with a lot of care, knowing I'm going to return to analyze this article next year, good or bad.

So what am I examining when I project teams? I look at a few different stats, but all of them are http://www.8msoccer.com/ meant to determine how much of a team's 2012 success was due to luck, and how much was the result of skill (and thus repeatable). The Indianapolis Colts, for example, were 11 5 despite a 30 point differential. That's pretty strong evidence that they'll regress in 2013.

I also like to look at expected points for and against. I already did that here at Dallas News in my article One Image That Can Help Predict the 2013 Season. You can take a look at that post for more details, but here's that image:

In analyzing expected points, we can see how many points teams should have scored and allowed given how well they played. That helps us better identify overachieving teams (like the Colts) or underachieving teams (like the http://www.mundodasaves.com/ Panthers).

When it comes to the Cowboys, the team really has two strong predictive factors in their favor heading into 2013: injuries and takeaways. The Cowboys managed an 8 8 record despite historically poor luck with injuries and just 16 total takeaways. it's highly improbable that Dallas will again have so many injuries and so few takeaways, even if they play exactly the same.

It's kind of like rolling a die and getting "2," then betting on if the next roll will be higher or lower; even though the first roll doesn't influence the second, it's still very unlikely to be a lower number. It's the same general idea with the Cowboys.

We hear the same line every year: six teams from last year's playoffs will be miss the postseason this year. Guess what? That information is useless and, if you're using it to make predictions, it's also dangerous. I wrote about this idea a couple years ago:

While it is a virtual certainty that some (and often times, many) different teams will make the playoffs in a given season, I disagree with the notion that it is rational to displace a talented team with a mediocre one in one's playoff predictions simply to accommodate the "six new teams will make the playoffs" trend.

The reasoning is simple. For the sake of argument, let's suppose that all of Black Friday Michael Kors Bag the playoff teams from last season Black Friday Michael Kors Deals have a 50 percent chance of making the playoffs again this year, while the non playoff teams have a 30 percent chance of making it.

While we can be fairly certain the group of playoff teams will contain some newcomers, we don't knowwhichnewcomers it will be, and we don't knowwhich teams they'll replace. To predict that a team with a 30 percent chance of making the playoffs will do so at the expense of a team with a 50 percent chance is simply bad math.

The idea is that even though we pretty much know the playoff teams will be different from last season, it's foolish to predict teams to make the playoffs just because they didn't make it last year. It's the same idea as everyone picking a 12 seed over a 5 seed in the NCAA tournament; although it's likely that at least one 5 seed will go down, we're still justified in choosing each of them to win, assuming they're all the favorites, because we don't know which 12 seed will win.