BYU at Michigan and other football predictions for Sept 26-27

It’s the last weekend in September, and it’s time to separate the contenders from pretenders in college football. Let’s look at some games.

Michigan by 5.5 over BYU

There are 3 reasons I think this prediction might be low.

First, Michigan has a really good defense. As expected, they have played well in all 3 games this season. However, you might brush off the UNLV game as too poor an opponent to draw any conclusions.

However, UNLV gained 493 yards on 5.67 yards per play in their opener against Northern Illinois. Last week, Ohio State had 298 yards on 4.52 yards per play against the same defense.

This in no way says that UNLV has a better offense than Ohio State. Variance plays a huge role in these results. In addition, UNLV QB Blake Decker was a little banged up against Michigan.

However, this suggests we can take seriously the 235 yards on 3.79 yards per play Michigan’s defense allowed against UNLV.

Second, Michigan’s defense faces BYU’s back up QB Tanner Mangum, who took over after Heisman caliber starter Taysom Hill got hurt. Mangum, a former four star recruit who recently returned from his Mormon mission, has looked solid in 3 games.

However, luck has played a big role in his results. He threw a 42 yard hail mary to win the game against Nebraska. In addition, Boise State’s secondary let his receivers get behind them twice for gains of 80 and 74 yards. This most likely won’t happen against Michigan.

Third, there’s the Jake Rudock factor. Many Michigan fans have seen the QB’s play as shaky through 3 games. However, he’s completing 65% of his passes for 6.14 yards per play, just below the college football average.

I don’t think it’s time to worry about Rudock. He’s made some bad throws, some of which have resulted in interceptions. I expect his play to improve.

Get the rest of my best predictions

I save my best college and pro football predictions for members.

However, each week, I offer a sample of these predictions to the curious football fans who sign up for my email list. This content (except for samples like the BYU at Michigan preview above) doesn’t appear on this site.

To get all of these predictions, sign up for this free email newsletter. It goes out at 2pm Eastern on Friday afternoon.

To sign up for the free email newsletter, just enter your email address and click on “Sign up now!”

Primer on college football analytics

detnews_persistOver at the Detroit News, I wrote about three of the most important concepts in college football analytics. It’s part one of a two part primer on analytics 101.

I looked at the following three concepts.

  • Last season matters. Teams tend to persist from year to year.
  • Predicting turnovers. Randomness plays a large role in turnovers, but there is some evidence for predicting interceptions.
  • Efficiency, efficiency, efficiency. Yards per play is a good place to start.

To read the article, click here.

Part two will come later in the season.

Defensive line tips Oregon game in Michigan State’s favor

Screen Shot 2015-09-23 at 3.46.40 PMOver at the Detroit News, I previewed the Oregon at Michigan State game on September 12th, 2015.

Both defenses have issues in the secondary. However, I thought Michigan State’s elite defensive line would win the game for the Spartans. It was enough to overrule my numbers which gave an ever so slight edge to Oregon on the road.

The game couldn’t have gone better for me, as Michigan State won 31-28. If you give the defensive line credit for stopping Oregon on 4th and goal from the one yard line in the first half, I even got the details right.

It’s not always going to work out this well. But combining numbers with watching games is powerful in making predictions.

To read the Oregon at Michigan State preview, click here.

Win probability for the Big Ten East in 2015

Screen Shot 2015-09-23 at 3.35.27 PMOver at the Detroit News, I used analytics to assign a preseason win probability for each team in the Big Ten East.

My favorite part of the article was questioning the dominance of Ohio State.

Before the season started, people acted like the Buckeyes were the greatest team in college football history. However, looking at their track record through analytics over the past 3 season reveals a different story.

Ohio State looked amazing against Virginia Tech in a revenge game from last season (Virginia Tech came to Columbus and won). Then they have looked putrid on offense against Hawaii and Northern Illinois.

I still think they’re one of the two best teams in the nation. But let’s not put them in the playoff just yet.

Michigan State and Michigan also have a chance at the Big Ten East title, with Penn State as an intriguing fourth option.

To read the article, click here.

Predicting Michigan State’s win total for 2015

Screen Shot 2015-08-27 at 7.03.37 PMOver on the Detroit News, I wrote a preview of Michigan State’s season and predicted a win total.

I knew my numbers had Michigan State a bit low. I tried to explain that in the article.

But I got the expected backlash. This email landed in my inbox a few hours after the article’s publication.

Hey Ed… much do you want to bet on this game? I have deep pockets so name your price. You should stick to engineering because you do not know jack shit about college football.

Can’t make everyone happy.

However, the article is good, as I look back at their 30 year history and isolate the one reason that the Spartans might not live up to potential. To read the article, click here.