Against All Odds, Week 1: Feeling Strangely Fine
Michigan State’s season-opening win, a historic upset, and encouraging betting results leave Dr. Green and White feeling a little better than just OK
Week 1 of the 2026 college football season is in the books, and plenty has already happened. There were big upsets, narrow escapes, surprising blowouts, and clock controversies, and that was just in the Big Ten.
The Michigan State Spartans opened the season Friday night with a 10-point home win over Mid-American Conference contender Toledo Rockets. The Spartans started fast, struggled in the red zone, suffered a few defensive breakdowns, fell behind, and then rallied in the fourth quarter to secure a victory that was almost comfortable.
In the end, the Spartans beat the Rockets by essentially the same margin the oddsmakers projected. Overall, it was a fine performance, and we learned quite a bit about the 2026 Michigan State team.
Quarterback Alessio Milivojevic was sharp, completing more than 75% of his passes. New running back Cam Edwards showed speed and elusiveness while finishing just shy of 100 yards. The run blocking appeared improved, and the wide receiver committee consistently found openings in the defense.
The red-zone failures were frustrating, but they also revealed that new kicker Liam Boyd is capable and reliable.
On defense, the Spartans allowed three touchdowns, but more than 75% of Toledo's 273 total yards came on just five explosive plays.
Still, Spartan fans have plenty of reasons not to be completely pleased. The pass protection was spotty, the running game faded in the second half, pass coverage could be a concern, and the wide receiver room appears to lack a true game-breaker. Above all, a performance against a MAC opponent does not necessarily translate against teams of Notre Dame's or Oregon's caliber.
On balance, however, it was OK. I am feeling strangely fine.
After all, not everyone in Michigan enjoyed a double-digit win over a MAC contender this weekend. Then again, not every Big Ten fan got to rush the field after beating an in-state directional school as a 26.5-point favorite, either.
Week 1 Betting Results
For readers unfamiliar with the series, here is a quick primer. Each week during the college football season, I present my computer's recommended wagers in my Bad Betting Advice column. After the games, I recap the results and offer a few observations from around the country.
Figure 1 compares each game's final margin with the opening point spread for Week 1 of 2026.
Figure 1: Week 1 results showing actual point differentials relative to the opening spread.
Each data point represents one game. Its position above or below the solid diagonal line shows whether the favored team covered the opening spread. The farther a point sits from the center diagonal, the greater the deviation from the spread.
The two dotted diagonal lines represent one standard deviation from the mean, or just over 14 points from the opening spread. In other words, roughly one-third of college football games finish about two touchdowns away from the opening line. Favorites outside those dotted lines either overachieved or underachieved by a significant margin.
A total of 28 favorites overachieved in Week 1 by beating the spread by more than 14 points, although more than half blew out FCS opponents. Notable overachievers included LSU, Arizona State, South Carolina, UCLA, USC, and Miami.
On the other side, seven favorites won but underperformed the spread by at least 14 points including Michigan, Arkansas, Old Dominion, Texas Tech, Oregon, and Missouri.
In either category, the result may reveal as much about the underdog as the favorite. As the season unfolds, we will learn whether these performances reflected one team's surprising strength, the other's unexpected weakness, or a little of both.
The data points below the solid red line in Figure 1 represent favorites that lost outright. Table 1 summarizes those eight upsets and compares them with last week's computer picks.
Table 1: Week 1 upsets based on the opening line, compared with last week's upset projections.
There were eight upsets in Week 1, closely matching my computer's projection of 8.7. Three involved FCS teams beating Group of Six opponents, including the computers' only correct upset pick: Tarleton State over Bowling Green.
Colorado (+7) upset Georgia Tech, and Tulsa (+13) upset Oklahoma State, but the week's biggest shock came Thursday night when UMass went into Piscataway and beat Rutgers, which opened as a 30.5-point favorite.
Upsets involving spreads of about 30 points are rare. I count only five bigger upsets in college football since 2001, the most recent coming in 2021, when Bowling Green (+31) beat Minnesota.
The odds of such an upset are about 1.5%, roughly equivalent to a No. 16 seed beating a No. 1 seed in the NCAA basketball tournament. An upset of this magnitude occurs about once every four or five years in college football.
As the accounting at the bottom of Table 1 shows, my algorithm went 1-2 (33%) on upset picks. FPI finished 0-2, while SP+ went 1-1 (50%).
Table 2 summarizes the computers' highlighted picks against the opening spread.
Table 2: Results of the highlighted Week 1 picks against the opening spread.
I was unsure how the computers would perform in the first year that games against FCS opponents counted among the official picks, but the early returns were strong.
Combined, the three systems went 10-3 (77%) against the opening spread, including a perfect 1-0 from my computer and a 2-0 record from FPI.
Across all 91 games, my picks went 50-41 (55%) against the spread. FPI finished 46-45 (50.5%), while SP+ edged my computer at 51-40 (56%).
Table 3 shows the Week 1 results for point-total wagers.
Table 3: Results of the recommended Week 1 point-total wagers.
Like the picks against the spread, my revised point-total evaluation strategy had a strong first full weekend. So far, the new methodology appears to have all worked out.
Overall, the highlighted picks went 13-9 (59%). SP+ went 5-3 (63%), while my computer went 9-7 (56%), bringing its year-to-date mark to .500.
So far, so good. Only time will tell whether it is made to last.
Updated Big Ten Odds and Expected Wins
After Week 1, I reran the full-season Monte Carlo simulation using updated power ratings and their current uncertainty. Table 4 shows the latest Big Ten outlook, and Table 5 shows the conference's updated win-distribution matrix.
All rankings listed beside teams refer to my computer's power ratings, not national polls.
Want the full picture after Week 1? Subscribe to Spartans Illustrated for updated Big Ten title odds, Michigan State’s projected path to bowl eligibility, revised spreads for every remaining MSU game, and more observations from around college football. The early results look promising, but are they made to last?
