Against All Odds, 2025 Week Seven: Dark Clouds

I noticed something odd during my drive to East Lansing on Saturday morning. When I left my house, the skies were mostly clear, and the sun was shining. It looked like a beautiful fall day in Michigan. I was expecting a beautiful day for football in Spartan Stadium.

But as I approached East Lansing from the east on I-96, I noticed a large bank of grey, dark clouds rising up in front of me. It looked like some sort of meteorological depression was hovering over the city. I did not realize it at the time, but it was a dark portent for what was about to transpire on the gridiron.

I was confident that after two tough road games in a row, the Michigan State Spartans would play a disciplined, sharp game of football on homecoming against the 1-4 UCLA Bruins. I expected a double-digit, feel-good win that would set the stage for the backend of the season and a likely bowl game for the first time since 2021.

I was wrong.

Instead, it was UCLA that came ready to play. The Bruins mauled the Spartans in every phase of the game. UCLA brought a level of energy to the contest that Michigan State did not come close to matching after the first seven minutes.

I have always believed that Jonathan Smith was the correct man to lead the Michigan State program following the fall from grace of former Spartan coach Mel Tucker. Smith was the top coach on the board at the end of the 2023 season. I was mildly surprised that Michigan State was able to reel him in.

But the Spartans have struggled over the past season and half since Smith's arrived. Rebuilding a program is a difficult challenge, and stormy days are to be expected. But this loss feels different. This loss feels like one that a different John Smith took on a rainy homecoming day in 2006 against Illinois. 

That Fighting Illini team finished 2-10 on the season. That win at Spartan Stadium was only the program's second Big Ten win in four seasons. John L. Smith won only one more game that season before being fired after four years in East Lansing.

I am not saying that Michigan State should move on from Jonathan Smith. I continue to be an optimist, and I believe that any coach needs more than 18 games to establish a program and a culture. If nothing else, while it might seem like fun at the time, jumping on the coaching carousel again this quickly more often than not simply makes one dizzy, if not nauseous.

But I cannot deny that the vibes following the loss on Saturday are not good. Dark clouds have descended on the Michigan State program once again. Coach Smith and his staff desperately need to give Spartan fans something to feel good about.

But the forecast for the next few weeks does not look any brighter. The weather is likely to get worse before it gets better.

If you are looking for other words of encouragement or a path forward for the program, I don't have much more to offer. I am also lost in the fog right now. But I will leave you with this.

As I drove home from East Lansing Saturday afternoon, at some point in my drive I realized that I could see blue skies once again. The sun was once again visible in the sky. Bad weather is always temporary. The sun will rise again on the Michigan State program. I just can't forecast when. 

Week Seven Betting Results

Let's now take a look at the performance of last week's picks from my Bad Betting Advice article, starting with the overview summary shown below in Figure 1. More information about how to read this figure can be found in the Week One edition of Against All Odds.

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

A total of 12 teams overachieved by beating the spread by more than 14 points. This list includes Texas San Antonio, Western Michigan, Iowa, San Diego State, Utah, Wake Forest, Clemson, and USC. The only team to fail to cover the spread by 14 points, yet win, was Ole Miss.

A total of 13 teams lost in upset fashion, which was a few less than the value of 15 upsets predicted by weekly simulation. Table 1 below summarizes those upsets and compares them to the picks made last week.

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

The biggest upset of the week was Penn State's loss at home to Northwestern (+21.5). In back-to-back weeks, Penn State was the victim of the two biggest upsets of the year. They say that lightning doesn't strike twice, but I am sure they also said James Franklin's job was safe. Neither statement appears true.

Other notable upsets in Week Seven include Marshall over Old Dominion (+13), Pittsburgh over Florida State (+9.5), UCLA over Michigan State (+7.5), Colorado over Iowa State (+3.5), and South Florida over North Texas (+2.5).

There was also a dark cloud hanging over my algorithm's upset picks this week. It went just 2-10 (16.7%) which brings the year-to-date-performance down to 22-30 (42.3%). Just two weeks ago my upset pick performance was sitting at over 58%. The direction of the wind changed quickly.

ESPN's FPI went 1-3 (25%) for the week, bringing its year-to-date tally to an even darker 11-24 (31.4%). 

Table 2 below gives the results of the computers' picks against the opening spread.

Table 2: Results of the highlighted picks against the spread for Week Seven.

It was a little bit brighter on this front in Week Seven. My algorithm went 5-6 (45.5%) which brings the year-to-date performance up to 22-19 (54%). My analysis of the FPI data resulted in no recommended picks against the spread. ESPN's metric remains at 20-14 (59%) for the year.

Considering all 56 games this weekend, my computer went 27-29 (48%) against the opening spread, bringing the year-to-date total to 166-186 (47%). The FPI performed a bit worse, going 25-31 (45%) against the spread. This brings the year-to-date performance down to 169-183 (48%).

Table 3 below gives the results of the point total bets for Week Seven.

Table 3: Results of the recommended point total bets ("over/under") for Week Seven

The lone lock pick was wrong, but the other three recommended picks were correct resulting in a record of 3-1 (75%) for the week.

Year-to-date, the locks are a gloomy 12-14 (46%) while the full set of recommended bets are 64-69 (48%). 

Updated Big Ten Odds and Expected Wins

Following the results of Week Seven, I have re-run the full season Monte Carlo simulation using the updated power rankings, including the current uncertainty in those rankings, to update the season odds for each team. 

Table 4 below gives the update for the Big Ten conference and Table 5 shows the updated Big Ten win distribution matrix.

Note that all the rankings listed in the table and mentioned below refer to my computer's power rankings and not any of the national polls.