The Roar
The Roar

Advertisement

College Football forecasts for Week 4

Badgers at Lambeau Field. The college football season is underway. Image: By Steve Swanson - Wikicommons.
Roar Guru
20th September, 2017
0

With so much to talk about in our regular column, there wasn’t room to slip our forecasts for Week 4 into the mix. So since we spent time looking at the different types of games (from arranged cross-sectional blowouts to the close inter-conference contests to the upsets that throw every number out the window), we’re going to share some of the different types of games from the punters’ perspective.

Possible routs
Looking at games that have the potential to be over in the first quarter doesn’t require that the teams be at completely different conference levels. Here are some of the games scheduled for this week, both intersectional and within the same conference:

– Kent State at Louisville (we have UL by 32; the odds makers say 43.5!).
– Nevada-Las Vegas at Ohio State (we have OSU -38; the odds makers 40.5).
– Massachusetts at Tennessee (the numbers range from 28 to 32 points for UT).
– Robert Morris at North Dakota State (NDSU by 40 on my notes; 45 on Sagarin’s).

But there are some battles inside conferences that look just as lopsided.

– Oklahoma at Baylor (On the road, OU is a 28-point favorite over the troubled Bears).
– Boston College at Clemson (Clemson is a 34-36 point favorite).

In the Colonial Athletic Association, James Madison and Richmond are both 18 point favorites within their conference this week against Maine and Elon, respectively.

It happens. You have a really weak team in your league, or a really strong team, and their games often look like three-touchdown wins in advance. But sometimes… sometimes the favorite has an off day. Sometimes the underdog plays over their head. Sometimes lucky bounces go the ‘dogs’ way.

Closer games but we think this team will win
Many, perhaps most games have something around 7-14 point spreads, like these:

Advertisement

– North Carolina State at Florida State (FSU is a 12-14 point choice).
– Pitt at Georgia Tech (GT by 7-11 points.)
– Virginia at Boise State (BSU favored by 12).
– Michigan at Purdue (UM by 10-12).
– TCU at Oklahoma State (OSU by 9-14).
– Washington at Colorado (UW up by 8-10).

History and mathematics say that a 7-point favorite wins about 75 per cent of the time. A 10-point favorite wins about 80 per cent of the time, and a 14-point favorite wins 85 per cent of the time.

But there is no “100 per cent” category! Of the list above, it’s an easy prediction to say that at least one of those favored teams will lose this week – the trick, tipsters, is figuring out which one!

Games with a “favorite”, but…
The 60 per cent probability range looks like about 3-6 point differences between teams.

There is a favorite, and more often than not they’re going to win, but it’s not going to be earth-shattering news if they’re upset.

Usually all of us pundits agree on the likely winner, if not the point spread. Here’s a handful of those games this week:

– Wake Forest at Appalachian State (here’s an interesting intersectional: middle of the ACC versus top of the Sun Belt. Wake by anywhere from 2 to 7 points…).
– Notre Dame at Michigan State (ND by four, consistently).
– Texas Tech at Houston (another intersectional: top of the AAC hosting the lower range of the Big 12, and favored by 5-6 points).
– Utah at Arizona (It’s anywhere from a one to six point spread in Utah’s favor).
– Florida at Kentucky (UK’s played well so far and is only a 2-4 point underdog).
– San Diego State at Air Force (After beating two Pac-12 teams, you’d think SDSU would be more than a 3-6 point favorite).
– Louisiana-Monroe at Louisiana-Lafayette (we have ULL as 4-6 point favorites, but any rivalry game makes predicting the outcome so much harder!).

Advertisement

Games that really are “too close to call”
There are always going to be games that the point spread is really within the “margin of error”. Any game closer than three points in American football (or Australian, for that matter) is really just an educated coin flip.

Some of the games on this list even have different favorites depending on which of us you follow for your predictions:

– Duke at North Carolina (I have UNC by four; Vegas and Sagarin both think Duke by about 2).
– Army at Tulane (I have it a tossup, Sagarin says Army by 3, Vegas says Tulane by 2).
– Mississippi State at Georgia (With MSU’s great play in 2017, their odds are anywhere from being a five-point underdog to a one-point favorite! I have them as a one-point dog).
– Georgia State at Charlotte (two teams that are really bad this year means nobody knows what will happen when they play another really bad team! I have Charlotte by 2; others are GSU by 4).
– Utah State at San Jose State (I have a tie; Vegas sees USU by three).
– Finally, the two MAC close games: Miami-Ohio at Central Michigan and Ohio U at Eastern Michigan. Both games are even to two points either way, take your pick.

These games really are 50-50: when I’ve examined these games over the course of previous years, the games under 3.5 points come out within the margin of error of a coin flip. The short version? We’re just guessing on these!

close