2018, Week 4 Recap: The Land of Confusion

OK, I'll admit it. This weekend's result just have me a bit confused.  Coming into this week, I thought that we would learn a lot about MSU's chances going forward for the rest of the year.  Either MSU would look better, win, and momentum would start to build for the bigger match-ups coming in a few weeks.  Or, MSU would struggle, lose or almost lose, and I would start to adjust expectations for the rest of the year.

Somehow, neither of these two things seem to have happened, or maybe both happened?  I am not sure... hence my confusion.

I mean, MSU went on the road, beat a decent to approaching good Big Ten team at night by 2 TDs.  If you would have told me that on Saturday afternoon, yeah I would have been quite happy.  But, if you also would have told me MSU was going to turn the ball over 4 times, give up a TD on a coverage bust, still struggle to run the ball, still have problems with penalties, the entire OL would be injured, and Lewerke was going to be erratic, I probably would have skipped the game and spent some quality time with a 6-pack of Two-Hearted Ale instead.  Watching the game it seemed like MSU was simultaneously in control and on the ropes at the same time.  Both disaster and the TKO shot seemed only one play away.  At the end, it was the Nailor jet-sweep that delivered the TKO.  That is cartainly better than the alternative.  But, I still don't have this team figured out, and it's almost October.  This is the world we live in, and these are the hands we're given.

But, I don't think that there are too many problem.  The good news is that MSU has been tested, and a lot of the warts have been exposed.  And, there has been progress in a few areas (for example, the pass rush! yea!!!)  Also, let us not forget that MSU has been on a strictly cup-cake free diet all month. I fully expect all three of our opponents so far will be playing in a bowl game.  I can't say that with confidence about our next two opponents.  So, there is a little time to try to shore up the remaining issues before the trip to Penn State and the annual grudge match with the Wolverines.  Also, note that since the first quarter of Game 1, MSU has only trailed for about 3 minutes of game time all season.  So, despite the fact that we may be living in the Land of Confusion, and there hasn't been much love to go round so far from some segments of MSU fans, I say that this is the time, this is the place, and it's time we look for the future.  It's time to stand up and starting showing just where this season is going to.

National Overview

MSU was not the only program that left me confused this weekend.  Honestly, it was a pretty confusing week in many corners of college football.  For starters, here is my weekly chart showing the results of Week 4:



Most notable are some of the teams that covered by a lot, namely Michigan and North Texas, two teams in the same peer group that also beat teams in similar peer groups: Nebraska and Liberty.  Also notable are the teams that failed to cover by more than 10 points: Miami, USF, Washington, and Oklahoma.

As for my algorithm, it was also pretty confused this weekend.   It went only 26-29 (47%) ATS, and my covers were only 5-7 (42%).  The FPI out-did me this week, going 29-26 ATS (53%).  For the year, I am 26-20 (56.5%) for covers and 99-99 ATS (50% even) overall.  As for upsets, the table below gives my weekly summary.

Even the upsets were confusing.  It seemed like a wild weekend, right?  Well, I counted only 10 upsets, which is actually less than the 12.6 than my pre-week simulation predicted (but within one standard deviation).  I supposed the reason is that many of the upsets that occurred were very high-profile.  The ODU upset over VA Tech at a spread of 28 is the level of upset that only happens once every 2-3 yrs, and is clearly the new biggest upset of the year (which makes Wisconsin fans pretty happy).  Texas Tech's upset of OK State, Kentucky's upset of Mississippi State, Texas' win over TCU, and Stanford's come-from-behind win over Oregon were all upsets with big potential impacts on conference races.  I guess that isn't so confusing after all.  But, my poor algorithm once again struggled at 3-7 (30%) for its upset picks, while FPI was 4-2 (67%).  It's hanging its digits in shame.

Big Ten 

On one hand, the Big Ten results this week weren't confusing at all. Oddly, every single team that won covered, and there was only one true upset: Purdue's win over Boston College (which I sort of predicted).  But, there were some interesting and confusing subplots.  For example, Penn State actually trailed in the second half at Illinois, yet won by almost 40. In other words, things escalated quickly.  We all knew Nebraska was pretty bad, but are they really that bad, or is Michigan actually improving (nah... why start now?)  Then, there is Maryland.  That whole situation is down-right confusing.  Are they really the team that beat Texas and Minnesota or are they the team that got housed by Temple (a team that lost to Villanova... on the grass and not on the hardwood)? Do they actually need a head coach? Confusing. The big game of the week was obviously Wisconsin's big win over Iowa.  I really had convinced myself after the Badger's debacle with BYU that the Hawkeye were going to win and then take the West.  Until the last minute, it looked like they would.  Then, not so fast my friend...

So, where does all this leave the Big Ten?  Well, I think that it is a good time to reintroduce some "new math" that I have been working on this fall: my conference champions and playoff projections.  I introduced this idea in another post this summer.  The mechanism is roughly that since I can project point spreads and therefore victory probabilities for each game left on the schedule, I can also calculate the probabilities of various overall end-of-season outcomes.  The data for the first few weeks is so sparse that I was hesitant to share it, but after Week 4, let's give this a try.  So, below I show my current projections for the Big Ten, including my updated power rankings.

As you can see, a few things have changed since this summer.  But, in general the picture in the East is not so different than we expected.  It is a 4-team race with PSU (42%) and OSU (29%) in the front and UofM (20%)  and MSU (9%) a little farther back.  The numbers of still a bit volatile, but as of this week, PSU and UofM are on an uptick, while OSU and MSU are a bit down from the summer projections.  Right now, Maryland (48) and Indiana (49) have almost identical power rankings, which might tell us what to expect when MSU visits Maryland in a month or so.  In the West, Wisconsin's win this weekend all but locked up the West (88%), but Iowa still shows a 9% chance to sneak back into the race.  As for the 3rd best team in the West, my power rankings currently say that is Purdue, but since they already lost a conference game, their odds to win the West are close to zero.

Interestingly, my metrics have all four East contenders except PSU with current power rankings below their summer predictions with five Big Ten teams total ranked between #8 and #16.  As for the eventual Big Ten champ, Wisconsin leads the pack at 41%, but this is largely because they have such better odds to reach Indy than PSU or OSU do.  As for the Playoffs, Penn State (31%) and Ohio State (28%) are neck and neck, and those odds will obviously shift a lot this weekend when the two teams meet in State College.

SEC

From here on out, I will just lead with the updated conference and playoff odds.  Here is the table for the SEC:

With the exception of one game, it was a very pedestrian week down south.  Both Alabama and Georgia got big wins over solid opponents (A&M and Mizzou) and now I project them both to have 80%+ odds to win their Divisions.  Alabama is still #1 in my power rankings with a 61% to win the SEC and similar odds to make the playoffs (which is also the best of any team in the FBS).  But, there certainly is some intrigue farther down the standings.  In the East, UGA already owns wins over South Carolina and Missouri, Kentucky already gave Florida a loss, and as everyone knows, there aren't any good football teams in the state of Tennessee.  So, that means that the only real threat to Georgia in the East is... 4-0 Kentucky? Confused? Me, too. I must admit, I did not see that coming.  

Out West, it was really looking for a while like Mississippi State was going to be a threat to Bama (at least a small threat). But, the aforementioned Kentucky Wildcats bopped the the Bulldogs on the nose and sent them home with their tails between their legs this weekend.  As a result, LSU (11%) looks to be the biggest threat with Auburn (2%) still technically alive.  In other words, it was a good day in the SEC for cats (unless they were cats from Missouri).  CAT POWER!

ACC

If you happen to be a fan of the ACC and just want to see the conference do well, the only good news this weekend was that Clemson looked good and covered against GA Tech.  Other than that, it was a bit of a mess.  On the Atlantic side, I guess you can be happy that Florida State actually beat an FBS team, finally (sorry, Northern Illinois) and Syracuse continues to not suck (Sorry, UCONN).  But, BC went on the road and got upset by Purdue and Louisville only managed to score 3 points against Virginia.  So, as you can imagine, Clemson's odds to win the Division are currently at 91%, and NC State (8%) is the only other team seems to have a shot. That said, I am not that confident of this projection considering that NC State is down one key contest due to Florence and we don't have as much data on them as I would like. 

In the Coastal division, the plot has thickened.  Everyone heard about Virginia Tech's stunning upset at the hands of Old Dominion, a team that prior to Saturday was 0-3 and hadn't yet played a team ranked in the Top 100.  But, what might have been missed by some is Miami's lackluster performance against FIU.  As a result, my calculations now have Duke favored to win the Division with the shockingly high odds of 58%.  The Blue Devils are also ranked #11 in my power rankings. Although this is a bit confusing (and perhaps a little hard to believe) Duke is certainly a team to watch.  Overall, however, my 4th ranked Clemson Tigers are projected to have a 65% chance to win the ACC Title and a 62% chance to make the playoffs, second only to Bama.

Big 12

It may have been the first week of full Big 12 play, but a few teams already made some key moves.  Texas' rather convincing win over TCU opened my eyes, and I was more than a bit confused when I saw that Army took Oklahoma to OT.  But, perhaps just as big was Western Virginia's throttling of K-State, and Texas Tech's upset win at Oklahoma State.  At week's end, West Virginia (although with only 3 games on record) is sitting at #7 in my power poll with a projected 60% chance to win the conference.  Oklahoma is a distance 2nd at 17%, which is on par with Texas (14%) and the Red Raiders (5%).  I am not sure if I am quite ready to put the coon-skin crown on the Mountaineer's head just yet, but similar to Duke, the result is worth noting.  Also of note is WVU's 31% odds to make the play-offs, which is good enough for 4th best overall, just ahead of Penn State and Georgia.

Pac 12

This week was yet another example of why I am hesitant to actually give gambling advice.  Based on my metrics, I thought that both Stanford and Washington State were absolute locks this week.  The Cougars looked good at half time, but the Trojans pulled it out after I went to bed.  Stanford, although they did eventually win in OT, looked like they were going to get blown out for most of there game at Oregon.  Regarding MSU's strength of schedule, it looked for a few minutes like the Sun Devils might steal a victory in Seattle, but the Huskies did prevail, although they did not cover.  

After the action this week, Washington (52%) still looks to be the favorite in the North, with Stanford (43%) close behind.  Ironically, my power ranking have Stanford ranked ahead of Washington (3 vs. 6) with much better odds to make the playoffs (42% vs. 17%).  It certainly helps to still be undefeated, and Stanford is one of those teams that could, in principle, go 11-1 yet not win their division and then make the playoffs anyway.  As for the South... it's just a confusing mess.  My metrics, believe it or not, have Arizona State ranked the highest (31) and with the best odds to win the division (37%), but honesty pretty much everyone not named UCLA has a reasonable shot.

Independents / Group of Five

Notre Dame switched QBs and put up a bunch more points this week in a winning effort at Wake Forest.  The Irish, by my calculations, have a 26% chance to finish at 11-1 or better, which I project would put them into the playoffs.  Judgment Day is near, however, as Stanford is on the clock.

As for the Group of Five, I have odds for the division and conference races below for 4 of the 5 leagues (Sorry, Sunbelt).  Let's start with the Mountain West.

Not much to see here as Boise was on a bye, is coming off a loss, yet they maintain an 85% chance to with their division and a 59% chance to win the conference (and likely secure the NY6 bid).  In the West Division, there is very little change in the 2-horse race we predicted in the summer between Fresno State and San Diego State, although my math says the Aztecs are slipping a bit, perhaps because they almost got nipped by Eastern Michigan this weekend.  Also of note is that Utah State continues to put close to 60 points per game on folks since the MSU game and they have arisen as the biggest divisional threat to Boise (12%)

After a fashion, it was a slow week in the AAC with only SMU's mild upset of Navy to entertain us.  In general, league contenders UCF, Temple, South Florida, Cincy, and Houston all won this week, although USF and Cincy's wins were a bit unimpressive.  My metrics still have UCF as a healthy favorite in the AAC East at 52% with Temple and Cincy both around 20%.  In the AAC West, Houston (88%) is in very good shape.  As for NY6 hopes, I will note that UCF is still undefeated and I project a 16% chance that they run the table.  Cincinnati could do the same, but I have their odds at only 5%.  Interestingly, my algorithm has not been impressed with USF so far and pegs their odds at only 4% to win the East and less than 1% to finish the regular season undefeated.

The MAC has continued to have a solid non-conference season and they added to their resume by scoring wins over Rutgers (by Buffalo) and Nevada (by Toledo) this week.  Also, Ohio and Eastern Michigan almost got upsets over Cincinnati and San Diego State, and Akron and Kent played some Power 5 teams pretty tough.  As for the conference overall, my algorithm sees the MAC East as a 2-team race between Buffalo (53%) and Ohio (45%), while in the MAC West, it currently really likes EMU (64%), despite them having two closes losses in as many weeks.  Toledo (17%) and Western (14%) are also within shouting distance. As for a NY6 slot, Buffalo is almost certainly the only team with even a remote shot.  I have their odds at running the table at 24%.

As for C-USA, my algorithm has been very impressed with North Texas.  They haven't beaten anyone good so far (Sorry, Arkansas) but they are winning by over 30 points per game. I give them a 92% chance to win the CUSA West,  an 84% chance to win the conference, and actually a 75% chance to run the table!  If this were to happen, I think they would get a look at the NY6, even if their strength of schedule would not be great.  In the CUSA East, my maths says FIU (42%) is the new favorite with the old favorite FAU close behind at 31% and the Marshall Thundering Herd still grazing nearby at 19%.

Finally, I will close with two (hopefully) not confusing questions:

How bad is... Louisville?

The Cardinal were supposed to be a borderline Top 25 teams that I predicted might have a shot to win the ACC Atlantic Division if they could somehow find a way to upset Clemson in Death Valley.  Right now, this seems like a far off dream.  Granted, they are still 2-2 now and one of those loses was the completely understandable pasting at the hands of Bama in Week 1.  But, after beating Indiana State in Week 2, they struggled to beat Western Kentucky in Week 3 (preseason rank = 104) and got destroyed by Virginia (preseason rank = 76) this past week, 27-3.  Now, I have them ranked 81st overall and I only project them to win 2-3 more games all season, quite a bit short of the 7-8 that I projected this summer.

How good is...Kentucky?

Just down the road in Lexington, the Wildcats are off to a much better start.  In many ways, they are the mirror image of the Cardinals.  Back in the summer, UK was ranked at only 67, but they have started 4-0 including wins over Florida and Mississippi State.  My updated rankings have them at #18 with a respectable shot (12%) to actually win the SEC East.  Of course, this would very likely require an upset win over Georgia in early November, but the Bulldogs do have to travel to Lexington this year.  Do you believe in miracles?

That is all for now.  I will have my Week 5 Preview up in a couple of days.  Until then and as always, Go State, Beat the Chips!