Photo by John S. Peterson
Nebraska Football

Hot Reads: The Anonymity of Being Underrated

August 22, 2018
4,168

Nebraska is USA Today's sleeper team in the Big Ten for 2018. Wisconsin, like it was in the Associated Press poll, was tabbed as the best team in the conference, Illinois the worst.

Here's the quick writeup:

Wisconsin and its outstanding offensive line, underrated quarterback and Heisman Trophy-contending running back gets the nod over Ohio State. Nebraska has hit the ground running under first-year head coach Scott Frost and will surprise teams in conference play. Illinois is the weak link in a league brimming with national names.

"Sleeper team" is about as effusive as the praise is going to get for Nebraska in the ever-dwindling preseason. Why? Many reasons.

There's the transition to new schemes and coaches (which also might be the Huskers' best foot forward depending on your point of view). There's the fact that Nebraska really wasn't very good a year ago. There's also the fact that it was really a three-year run of average football, not just one bad year. There's the 2018 schedule.

Throw all of those ingredients in the pot and you're not exactly expecting an award-winning stew. Owing to the talents of the new chef, Nebraska fans are expecting a better stew, maybe even a good stew, but few are willing to go beyond that.

And there's a good reason for that, too.

"There is a random element to teams that make a sudden rise," Ed Feng writes at his site, The Power Rank. That's from a story titled "How to make accurate football predictions with linear regression." Don't be scared off by the title or the graphs you'll find when you get there. Feng's goal with this piece was to make it easier for non-data scientists (me) to understand data science. If you've ever looked at a set of preseason power rankings and wondered where the numbers are coming from, read this story.

There's a lot worth discussing from Feng's piece, but I want to focus on the fifth point, a general summary of the story as a whole. ". . . [Q]uantities in football like turnover margin show very little persistence from an earlier period to a later period. These quantities regress to the mean, as your best guess for the later period is the average," Feng writes. He also mentions a team's record in close games as another quantity with little persistence year to year.

Let's look at my personal favorite example for this upcoming season. Michigan State rebounded from a 3-9 season in 2016 to go 10-3 last year. If you look at the turnover margin from 2017 there's nothing alarmingly large there that might set off alarm bells. The Spartans were +3 on the year, 46th nationally. But you have to look a little closer.

After the first three games of the season Michigan State was -6 in turnover margin. That included wins over Bowling Green and Western Michigan and a 20-point loss to Notre Dame. Things flipped in Big Ten play. 

Over those nine games the Spartans were +8 in turnovers, a rate that ranked 11th nationally for teams in conference games. That total included Michigan State going +2 in a 17-10 win over Iowa, +5 in a 14-10 win over Michigan and +2 in a 27-24 win over Penn State.

That's the other key to formulating expectations for Michigan State in 2018. The Spartans went 5-1 in one-score games. Like turnover margin, that's something you would expect to regress to the mean. Yet Michigan State is ranked 11th in the preseason AP poll coming off its 10-win season because it has most of its major contributors back and is coming off a 10-win season. Preseason polls aren't much more complicated than that.

I'm putting out my Big Ten predictions on Friday and I won't have the Spartans that high. But that doesn't help us much when it comes to Nebraska. The Huskers' 2017 turnover margin (-7) was in line with their 4-8 record, but you might expect a slight improvement. The record in one-score games (2-3) was pretty normal. Nebraska wasn't really a victim of randomness while going 4-8 a year ago.

And that makes for modest projections in 2018. ". . . [S]ome quantities like team strength in college football tend to persist from year to year," Feng writes near the end of his story. Nebraska wasn't strong a year ago, and thus you don't find many people predicting the Huskers to be strong this year.

But if the Huskers do end up being stronger than expected in 2018, it'll be easy to understand after the fact.  It's just hard to project right now.

That's not the worst place to be. If you look at it the way I do, it definitely beats being Michigan State.

The Grab Bag

Today's Song of Today

×
Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.