Week Four Recap: Relentless

Let's cut right to the chase. This was not a game that the Michigan State Spartans should have won. The Nebraska Cornhuskers had the Spartans dead to rights, and the Spartan somehow managed to steal victory in the final few minutes and in overtime, thanks to just enough great plays on special teams, defense, and offense in the late hours of Saturday night.

First, it was the special teams. With the Spartans trailing by seven and with just under four minutes to play, Jalen Reed fielded a punt and raced to the end zone to tie the score at 20 all. 

Then, it was the defense's turn. After giving up three drives of 10 play or more in the second half alone, and trailing miserably in time of possession, the MSU defense stymied the Cornhusker offense in the final minutes of regulation to force overtime.  Then, on third and goal in the overtime period, Spartan cornerback Chester Kimbrough picked off a Nebraska pass and almost ended the game with a return inside the Nebraska 30-yard line. 

Finally, despite failing to pick up a first down in the entire second half, it was the Spartan offense who sealed Nebraska's fate. Kenneth Walker, who had been contained for most of the evening, took the direct snap and finally found some daylight. Walker, too, almost ended the game, but was tackled just short of the goal line. Perhaps it was fitting, though, that this stop set up a winning chip shot field goal by Matt Coughlin on an evening where MSU's special teams were the star of the show.

In reflecting back on the game, it had the feel of a contest that past MSU coaching regimes would have surely lost. The George Perles Spartans would have simply refused to pass the ball and not scored any points at all. The Nick Saban Spartans would have crumbled in the closing minutes when the potential third down catch by Jalen Reed was overturned by the official. They would have wound up losing for 14 points for no reason.

Under Bobby Williams, the Spartans may have built a bigger lead in the first half, but a string of bad plays due to a combination of poor discipline or poor execution would have resulted in a 21 unanswered points from Nebraska in the fourth quarter. The John L Smith Spartans would have done the same things, but John would have slapped himself during the post-game press conference.

Under Mark Dantonio, the game might have progressed similarly, but in the overtime session, a player of than Kimbrough would have been on the field, the Huskers would have scored in overtime, and the Spartans would have failed to answer.

So how did the Spartans get here? How did they pull this off? What makes this new Mel Tucker coached era different? I think it boils down to simply one word:

Relentless.

This word has been Coach Mel Tucker's rallying cry since the day that he arrived, and the Spartans demonstrated that attitude on Saturday night. Over the past 18 months, the team and roster has been molded into his image. When the chips were down late in the game, the Spartans kept fighting, in all phases of the game. When it mattered most, it was the Green and White who made the plays that mattered. They simply kept chopping. These Spartans look different.

Now there was certainly a lot to be concerned with regarding Saturday night's performance. The running game failed to get much traction until literally the final few plays of the game. Peyton Thorne was erratic in the second half, and the defense gave up far too many yards to Martinez on the ground. 

I will admit that Nebraska (with the exception of special teams) is perhaps better than I gave them credit for. If the wheels don't completely come off from the bus in Lincoln, that looks like a team that will steal a high profile win here or there this season. Saturday night's win might look a lot better in late November. 

Beyond that, in the long run, winning this game ugly might be just what the Spartans needed. They know that they need to play better and to continue to work if they want to keep winning. They have had a chance to learn that lesson without absorbing an expensive loss. Saturday provided plenty of film for the coaching staff to use to keep their collective feet to the fire. There can be no let up. It will be relentless.

Week Four Results

As is my weekly tradition, Figure 1 below shows the results of all 58 games in Week Four that featured two FBS level teams.

Figure 1: Results of Week Four showing the actual point differentials relative to the opening spread.

The overachievers, relative to the spread this week were Eastern Michigan, Texas, Western Michigan, Air Force, Northwestern. The teams who badly underachieve, yet still won include Wyoming, Fresno State, BYU, and Oklahoma. Also note that Michigan, Iowa, and Auburn were also very close to crossing the dotted line into the underachiever zone.

Week Four saw a total of 13 upsets relative to the opening Vegas line. Table 1 below summarizes those 13 upsets in comparison to the upset picks made by the computers last week.

Table 1: Upsets based on the opening Vegas line compared to the upset projections from last week

Overall, the total number of upsets was very much in line with what the math predicted, and there were several big upsets, both based on the spread as well as on the impact to Power Five conference races. Georgia Tech's upset of UNC and NC State's upset of Clemson (which I predicted), have thrown the ACC race into chaos. Oregon State's upset of USC has done the same for the Pac-12 South.

In addition, there were two upsets this week in games with spreads over 20 points. UL Monroe's (+21) upset of Troy was of a magnitude where it is typically only observed about four times per year. But, there was also Bowling Green's (+31) upset of Minnesota. That is the biggest upset I have on file since South Alabama (+33.5) over Mississippi St in 2016. The Gophers had a 98.6 percent chance to win. Upsets of this magnitude should happen only once per 5 years (which checks out).

As for picks against the spread (ATS) this week, Table 2 provides the result of my highlighted picks.

Table 2: Results of the highlighted picks versus the spread in Week Four

Similar to last week, my analysis of the spread data and the computer data resulted in a very large number of recommended picks. Unlike last week, though, the computers did not have a good week. Basically, both my computer and my curated picks from the FPI only got a third of the picks correct. This dropped the year-to-date performance of both systems down to around 52 percent.

Overall, my computer was only 22-36 (38 percent) ATS, bringing the year-to-date total to 91-114 (44 percent). The FPI had a better week at 30-28 (52 percent) and is doing better for the year at 107-98 (52 percent).

Updated Big Ten Odds and Expected Wins

Following the results of Week Four, I have re-run the full season Monte Carlo simulation using the updated power rankings, including the current uncertainty in those rankings, in order to update the season odds for each team. Table 3 below gives the update for the Big Ten conference.

Table 3: Updated Big Ten rankings, expected wins, strengths of schedule and season odds.

Other than Minnesota's huge upset loss to Bowling Green and Wisconsin's meltdown against Notre Dame, the rest of the results went more-or-less as expected.

A few teams (like Iowa, Indiana, Purdue, and Michigan) did not cover their spreads, while other teams (like Maryland, Illinois, and Rutgers) did. Other than that, the changes in the season odds seem largely due to some strength of schedule changes as the data accumulate. For example, Penn State did not get any help from Wisconsin or Auburn (who barely beat Georgia State) this week.

Overall my computer continues to love Michigan in the East and Iowa in the West. The Wolverines moved up to No. 2 in my power rankings (behind only Georgia) and now have 56 percent odds to win the East (according to my computer). Iowa is now at No. 3 overall with 73 percent odds in the less competitive West.

As for the rest of the East, Penn State, Ohio State and Michigan State all have between an 11 and 14 percent chance to make it to Indianapolis. Meanwhile, Purdue is now projected to be the second best team in the West with a 14 percent chance to pass the Hawkeyes. No other team in the West has current odds over five percent. 

Table 4 below gives the updated Big Ten win distribution table following Week Four.

Table 4: Updated Big Ten win distribution matrix after Week Four

Michigan State's expected win total is now at 8.33, which is consistent with projected record of 8-4 above in Table 3. If we dig a bit deeper into the numbers, there is now a 45 percent chance to get to nine wins or more. The odds of a double digit win total is up to 26 percent. 

For those out there who are only relentless in their pessimism, the odds of the Spartans falling off from a cliff and not reaching the .500 mark are now below five percent.

A closer look at MSU's remaining schedule

Figure 2 below gives my updated projections for the Vegas lines and win probabilities for all of the games remaining on MSU's schedule.

Figure 2: Updates projections for the Vegas line and win probabilities for MSU's remaining schedule

My computer is quite unwavering in its confidence in the Spartans next week against Western Kentucky. My projected spread of 16 points is higher than the opening eight-point line (which immediately moved to around 10 or 10.5).

The computer is also projects MSU to be a five to seven point at Indiana and versus Maryland. Then there is a group of three games at Rutgers, at Purdue, and versus Penn State where the math suggests that the Spartans will be a narrow underdog. The game at Ohio State only looks to be slightly tougher than the contest at Rutgers right now, and the Halloween weekend brawl with the Wolverines now projects as about an eight-point spread.

That said, we are entering a phase of the year when I am almost ready to completely drop the preseason ranking data from my power rankings analysis. I will make this change for next week and as I preview that data, it is a bit eye-opening. Just as a preview, there is data to suggest that Rutgers might be a lot better than we think right now, and Penn State and Ohio State might be a lot worse. Stay tuned.

National Overview

Finally, let's take a quick look around the country at the rest of the action. Once again, the updated conference odds can be viewed by clicking on the hyperlink attached to each conference name.

It was pretty wild in ACC action this week as as Clemson, North Carolina, and Virginia all suffered upset losses at the hands of NC State, Georgia Tech, and Wake Forest respectively. This now throws the race in both the Atlantic and the Coastal Divisions into turmoil. Currently, I have Wake Forest as the favorite in the Atlantic with about a 43 percent chance and with NC State close behind with 25 percent odds.

On the Coastal side of the conference, Virginia Tech has emerged as the potential leader with almost a 40 percent chance to win the division. With Pittsburgh (3-1) nipping at their heels at 31 percent. Both teams got wins over FCS teams this week.

In the wide open Big 12, Iowa State and TCU suffered upset losses at the hands of Baylor and SMU this week. In addition, Kansas State, who I felt had a legitimate shot at potentially challenging for the Big 12 title, suffered a double-digit loss to Oklahoma State, which dropped them significantly down in my standings. That said, right now my math gives a total of six teams a 25 percent chance or better of making the Big 12 title game, including Oklahoma State, Texas, Oklahoma, Iowa State, Baylor, and Kansas State in that order.

Moving on to the SEC. The biggest result of the week was Arkansas doubling up Texas A&M, which puts the Aggies’ potential claim on the SEC West very much in doubt. Alabama continues to lead the pack with a 43 percent chance to win the division. Interestingly, my math suggests that Ole Miss may be Bama’s biggest competition with a projected 30 percent chance in the West.

In the SEC East things remain fairly straightforward. The No. 1 ranked Georgia Bulldogs project to have an 89 percent chance to win the East, followed by Kentucky and Florida with just a little over a 5 percent chance each.

In Pac-12 action, Oregon took care of business against Arizona, and Washington managed to be beat Cal. I currently have those two teams at the top of my leaderboard in the Pac-12 North with 39 and 37 percent odds to win the Division. Meanwhile, the standings in the Pac-12 South were shaken up by USC’s shocking upset lose at the hands of the Oregon State Beavers, dropping them to two-and-two and dropping their odds to win the Pac-12 South significantly. I currently have UCLA at the top of the leaderboard in the south with just over a 50 percent chance to win the division with Arizona state currently in second place at 21 percent.

Overall, the four teams at the top of my current playoff leaderboard include Georgia with over 80 percent odds to make the playoffs, followed by Michigan, Notre Dame, and Iowa all with odds between 35 and 41 percent. The rest of the top 10 is rounded out by Wake Forest, Alabama, Oregon, Ole Miss, Oklahoma State, and Oklahoma.

Things were quiet in Group of Five action. Despite having a bye this week, Cincinnati remains the steadfast favorite in the American Athletic Conference (AAC). The Bearcats are also at the top of my Group of Five New Year's Six leaderboard with just over a one-third chance to claim that slot.

In Conference USA (C-USA) action, Texas San Antonio got a boost by upsetting Memphis of the AAC this weekend. However, the Road Runners still trail UAB as potentially the best team both the C-USA West and the conference overall, based on my calculations.

It was a good weekend for the MAC, as Bowling Green scored a major upset over Minnesota and Western Michigan easily handled Mountain West opponent San Jose State. I currently have Miami (OH) as the leading candidate in the MAC East, and Western Michigan is now potentially the best team in the West. No MAC team has a realistic shot at the New Year’s Six bid at this point.

It was a quiet weekend in the Mountain West, although a few prominent teams had surprisingly close shaves including Fresno State just edging out UNLV and Wyoming just escaping UConn. Boise State and Fresno State are both leading their respective divisions.

Finally, in Sunbelt action the most notable result was Appalachian State scoring a nice Group of Five intersectional win over Marshall. The Mountaineers remain neck-and-neck with Coastal Carolina as the leaders in the Sunbelt East. Meanwhile Louisiana maintains the best odds in the Sunbelt West.

That'll do it for this week. Stay tuned for my weekly preview and batch of bad betting advice coming soon.

Additional data tables shown below.