2024 Week Five Preview: Litmus Test

Merriam-Webster defines a litmus test as "a test in which a single factor (such as an attitude, event, or fact) is decisive."

We are just four game into the Jonathan Smith era of Michigan State football, so the use of the word "decisive" is a bit much. However, this Saturday night's match up with the Ohio State Buckeyes feels like it could be a strong indicator of things to come.

As I outline below in detail, the Spartans have struggled mightily against the Scarlet and Gray for much of the past decade. But the majority of those games came against a Mel Tucker coached team or in the waning days of the Mark Dantonio era. Even with a sample size of just four games, I have reason to believe that Jonathan Smith is sideline upgrade.

Last weekend's frustrating loss to Boston College made some fans red in the face and other just blue, but that partially obscures the basic fact that the Spartans on balance have played some pretty solid football over the past three weeks, including two road games against solid Power Five opponents.

If the Spartans' could have avoided one or two of the corrosive errors, Michigan could be sitting at 4-0 with two double-digit road wins and possibly a top 20 ranking. 

In other words, Michigan State could very well be better than anyone realizes right now.

On the other sideline, Ohio State comes to town with yet another highly touted roster and a 3-0 record including wins over a cupcake, a doormat, and a lost puppy. The schedule was so strenuous that they needed to take a break in Week Three. So far this year the Buckeyes are the definition of untested.

Then there is the opposing coach. Ryan Day has only lost eight games total and only three in Big Ten play over five seasons. Yet many observers, myself included, find him underwhelming. 

Maybe it is just his losing record against Michigan (which is something fans in East Lansing have not seen since the last John Smith) but something tells me that Day may not be the answer in Columbus.

In other words, Ohio State could be a little overrated.

When these two elements are mixed together, I think that the outcome could be unpredictable, especially if the experiment is run after dark. What happens if Aidan Chiles has a turnover-free game and if the Spartans can minimize penalties while maintaining the strong chemistry and execution on defense? The result could be explosive.

There is still a good chance that Buckeyes comes into Spartan Stadium and put up seven touchdowns the way that they did on their last two visits. But somehow I don't think that is the way this game is going to play out this time. The ingredients on the Michigan State sideline are different this time around.

If Coach Smith can keep up with Ohio State and make this a competitive game into the fourth quarter, that would certainly indicate to me that good things are on the horizon.

Michigan State Prediction

The Spartans have squared off with the Buckeyes a total of 52 times going all the way back to 1912, a game Michigan State won in Columbus. When the Spartans joined the Big Ten the series stared in earnest with the Green and White winning four of six games between 1951 and 1966.

But the rivalry has been decisively one-sided ever since. Michigan State has only won 10 times in 45 tries since 1967, leaving the overall series with a record of 15-37. That winning percentage of .288 is the lowest for the Spartans against any Big Ten team that has been in the conference longer than six months.

The recent history of the series has been caustic for the Spartans. Michigan State was able to get three wins over the five year span from 2011 to 2015, including a Rose Bowl-clinching win in the 2013 Big Ten Championship Game and in the rain in Columbus in 2015. 

But those are the only wins for the Spartans over the Buckeyes this century. Michigan State has not gotten a win over Ohio State in Spartan Stadium since Nick Saban did it in 1999.

The story against the spread is not much better. The Spartans are a sour 7-15-1 ATS against Ohio State since the beginning of reliable data in 1997. Ohio State has covered the spread against Michigan State in the last seven meetings by virtue of a 34.5 point average margin of victory.

So all that is not great.

The spread for this year's game opened at +23 for the Green and White. That puts Michigan State's odds at an upset win at around 5%, which is similar to the odds of a No. 15 seed beating a No. 2 seed in the NCAA Men's Basketball Tournament, not that something like that has ever happened to either school.

The line opened in similar territory for Ohio State's last two visits to East Lansing. Based on my records Michigan State has only been more than a 22-point underdog at home four times since at least 2001 (with the 2016 and 2023 games against Michigan being the other two contests.)

Based on all this data, it is tempting to think that Michigan State is going to get blown out. While that certainly may happen, I think that this game might be a lot closer than the Vegas line implies, for the reasons outlined above. I believe the line is inflated due to the recent history in the series, which has little to do with the current roster or staff. 

My computer tends to agree, at least with the line being inflated part.

Neither I, not my computer is bold enough to make the upset pick here, but I have a strong suspicion that the Spartans will cover and play a competitive game. My final score prediction is:

Ohio State 31, Michigan State 18

Big Ten Overview

Table 1 below gives a full summary of the action in the Big Ten in Week Five, including my projected scores, the opening point spread, and the computers' projected point differentials. 

Table 1: Summary of Big Ten action in Week Five, including my algorithms' projected scores. The teams shaded in green are projected to cover by my algorithm. The teams shaded red are projected to cover by the FPI.

Week Five marks the beginning of Big Ten conference play in earnest as every team except idle Iowa and Northwestern face a conference foe. Oregon (-23) is a huge favorite on the road at UCLA and Penn State (-18) is a big favorite at home against Illinois. Both teams should be safe, but my computer expects a more competitive game than the spread indicates in both cases.

As for the other primary conference contenders, USC (-13.5) looks to get back on track by hosting Wisconsin while Michigan (-10) plays their fifth consecutive home game of the season against Minnesota.  My computer like USC to cover but projects a closer game in the battle for the Little Brown Jug.

Last week I almost convinced myself that Nebraska was a dark horse contender, but last week's home overtime loss to Illinois has them in the bottom half of my Big Ten leaderboard. The Huskers (+9) are likely to get a feel-good win at Purdue.

Indiana and Maryland are both off to strong starts and both have aspirations to stay in the conversation for the Big Ten race. The Hoosiers (-7) are a touchdown favorite at home but my computer sees the game as a nail biter.

The biggest enigma in the conference for me is Washington. My computer has the Huskies ranked generously at No. 18 and with the sixth best odds to win the Big Ten. Washington makes the short five hour and 20 minute flight from Seattle to Newark this week to face Rutgers (-3). My computer likes the Huskies in an upset, as long as there is not mathematical correction for jet lag.

Other Notable National Action

Table 2 below summarizes the notable action in the rest of the college football scene in Week Five. Note that all rankings listed below are based on my power ranking system and not a national poll.

Table 2: Summary of other notable action in Week Five including my algorithms' projected scores.

With one exception, the action in the SEC is pretty quiet in Week Five. Of the top six contenders, Missouri and Tennessee are on a bye, Texas is a 40-point favorite at home versus Mississippi State, and Mississippi (-17) is over a two touchdown favorite at home versus Kentucky.

But the exception is a huge one as No. 4 Georgia (-1.5) visits No. 2 Alabama in what could be the biggest game of the entire regular season. The Bulldogs are a slight favorite on the road, but both the FPI and especially my computer are picking the Tide in an upset.

Last weekend the Big 12 race got turned upside down as both Kansas State and Oklahoma State absorbed upset losses. This week, the Wildcats (-5) host the Cowboys and the loser is likely out of the conference race. The computers are split on the outcome, but both machines have Oklahoma State covering.

Three of the four remaining undefeated Big 12 are all in action against manageable opponents. Utah and Central Florida are both 11-point favorites at home against Arizona and Colorado, respectively. The computer both like the home teams to cover. Iowa State (-12.5) is on the road at Houston but is a slightly larger favorite.

The Big 12 contender with the biggest challenge is BYU. The Cougars took out Kansas State last week and this week are underdogs on the road at Baylor (-3). Both computers predict a nail biter and the FPI foresees an upset.

Three teams have risen to the top of the ACC race, but only No. 7 Miami has a true conference game this week as the Hurricanes host Virginia Tech (+16.5). The computers both have Miami covering. 

No. 17 Louisville looks like the most likely ACC challenger, and this week the Cardinals (+6) have a tough road test at Notre Dame. The computers both have the Irish covering, but a strong showing from Louisville could bolster the resume for a potential at-large play off bid.

The only other ACC team with better than a 20% chance to reach the conference title game is No. 23 Boston College. The Eagles (-11) host Western Kentucky. The Hill Toppers currently sit atop my Conference USA leaderboard and are hoping to stay in the race for a potential playoff berth.

The remaining games shown in Table 2 also will potentially impact the race for a playoff spot for the highest ranked Group of Five Champion. The most significant game on the schedule is Fresno State at UNLV (-4.5). 

The Rebels currently sit at the top of my Group of Five Leaderboard, but the announcement on Wednesday that Rebel quarterback Matthew Sluka is planning to leave program due to an NIL dispute makes this game even more interesting. The line has already dropped by over two points and the rest of UNLV's season now very much in doubt.

If UNLV were to stumble, it could benefit teams such as James Madison, Boise State, Army, Memphis or Tulane. Memphis, James Madison, and Army are all double-digit favorites while Boise State (-7) and Tulane (-6.5) have slightly tougher road games against Washington State and South Florida, respectively.

Picks of the Week

The sections above highlighted several potential upsets and recommended picks against the spread. Figure 1 below gives an overview summary of all my computer's projections for the week in comparison to the opening Vegas spread. The same data generated by EPSN's Football Power Index is shown in Figure 2.

Figure 1: Comparison of the projected final margins of victory based on my algorithm to the opening Vegas lines for Week Five.

Figure 2: Comparison of the projected final margins of victory based on ESPN's FPI to the opening Vegas lines for Week Five.

A brief explanation of how to read these figures can be found in Week One's Bad Betting Advice column. This week's upset picks (the games appearing to the left of the vertical red lines in both figures) are summarized below in Table 3.

Table 3: Summary of upset picks for Week Four.

My computer has been aggressive in making upset picks so far this year and Week Five is no exception. My machine recommends eight total upset picks this week, including Washington over Rutgers, and Alabama over Georgia.

ESPN's FPI suggests a total of four upset picks. The FPI notably agrees with picking Alabama over Georgia and it also recommends a wager on Oklahoma State and BYU to both win straight up.

A simulation of the week's games suggests that a total of 12.2 plus-or-minus 2.9 upsets out of 50 games is most likely. 

The games shown above in Figures 1 and 2 which fall outside of either set of dashed diagonal lines make up the set of suggested bets against the spread. Those selections are summarized below in Table 4.

Table 4: Picks against the opening spread for Week Five. The picks are listed in order of confidence score.

The machines together recommend eight total picks. Both computer systems like Alabama (+1.5) to cover as well as win. My analysis of the FPI data results in recommendations of Boston College (-11) to cover versus Western Kentucky and for LSU (-20.5) to fail to cover against South Alabama. My computer notably suggests a wager on both Utah (-11) and Memphis (-24) against the spread.

Finally, Table 5 below summarizes my computer's recommended point total bets for the week.

Table 5: Recommended total points bets (over/under) for Week Five. Note the picks labeled as "locks" are at a higher level of confidence.

As real data from the season continues to accumulate, my point total bets are also starting to settle into a groove. My analysis this weeks results in just 13 recommendations, only two of which qualify as locks. 

That is all the advice I have for you this week. Enjoy the games and check back early next week in Against All Odds to see how the picks above panned out.