Eastern Michigan vs Michigan State Prediction, Picks & Odds — Sep 12, 2026
Eastern Michigan's chunk plays are Michigan State's soft spot
Michigan State's 73-14 win in 2014 says nothing about this one; the game turns on whether Eastern Michigan's explosive passing can outrun its own three-and-outs.
What follows is the preview as it was published before kickoff — unedited.
Box score, drives and the model's numbers →The last time these programs met, Michigan State won 73-14. That was 2014, and it is the wrong reference point for Saturday: both teams finished 4-8 last season, both scored in the mid-twenties per game, and Michigan State grades just 4.3 points the better side overall. The spread is 17.2. Nothing on either resume argues for a gap that size, and the widest single matchup edge in this game belongs to the visiting team.
The range of outcomes for this game. Each bar is how often the final margin lands in that range — the more it leans toward Michigan State, the more likely Michigan State wins.
The biggest mismatch on the field belongs to the underdog
Eastern Michigan's big-play offense against Michigan State's defense is the widest gap anywhere in this matchup, more than four standard deviations in the visitor's favor, and it is not a one-category quirk. The explosive passing split runs the same direction by two and a half, with Ben Prosper the receiver Michigan State has to account for on every snap that is not a run down. Chunk plays are how Eastern Michigan scores, and Michigan State's defense is where they are available.
Each unit matchup this season, sorted by how decisive it is — the top bar is what most tilts the game. Bars toward Michigan State are battles it should win; toward Eastern Michigan are ones Eastern Michigan wins.
Noah Kim projects 181 to 327 passing yards with an 81% chance of a touchdown pass, and the width of that band is the entire story of Eastern Michigan's offense. Kim at the top of it keeps the visitors inside a touchdown; Kim at the bottom is where a 17.2-point number comes from. Eastern Michigan scored on 14.3% of 14 drives against San Jose State with the same quarterback and the same receivers, which is what the low end of that range looks like on a scoreboard.
Kim's targets carry the same signature. Nick Devereaux projects 25 to 79 receiving yards but a 51% chance of a touchdown. Harold Mack projects 29 to 81 with a 27% touchdown chance, Santonyo Isaac 21 to 62. Devereaux and Mack are priced as scorers rather than volume receivers, which is how Eastern Michigan builds 24.5 points a game out of 20.4 yards per drive. Michigan State can win most of the snaps against this group and still concede the two that decide the margin.
Everything between the chunk plays goes the other way
The three categories that decide whether an offense ever reaches those snaps all belong to Michigan State: the run game by two and a half standard deviations, short-yardage power and staying on schedule by roughly a standard deviation and a half apiece. Amareon Blue projects 28 to 69 rushing yards against a run defense ranked 5th of 185; Braydon Bennett projects 14 to 34. Neither band lets Eastern Michigan shorten the game on the ground or convert third-and-two on demand.
Last Saturday against San Jose State, Eastern Michigan went three-and-out on 42.9% of its drives, scored on 14.3% of 14 possessions, and averaged 20.4 yards per drive in a 21-27 home loss. Ben Prosper and that chunk-play edge were already on the roster that afternoon. They produced 21 points, because an offense that never gets to second-and-manageable never gets to use the throws it is built around.
The matchup grades hand Eastern Michigan the most valuable edge on the field; the San Jose State result says the offense never got to use it, going three-and-out on 42.9% of its drives at home. Michigan State's run defense, 5th of 185 by what it allows, is a harder assignment than the one that already stalled Eastern Michigan, and Blue at 28 to 69 rushing yards is not the answer to it.
Michigan State's cleanest evidence came without its projected quarterback
Michigan State's 30-20 win over Toledo produced the best drive chart either side owns: points on 63.6% of 11 possessions, 36.5 yards per drive, a three-and-out on only 9.1%. Set against Eastern Michigan's same-week home game, the numbers separate cleanly: 63.6% scoring against 14.3%, 36.5 yards per drive against 20.4, 9.1% three-and-outs against 42.9%. That is the gap between these teams as of seven days ago.
Alessio Milivojevic has not taken a snap this season; nine games last year is the whole record. He projects 96 to 199 passing yards with a 69% touchdown chance against a pass defense ranked 58th of 185. Brandon Tullis projects 13 to 36 rushing yards and a 28% touchdown chance against a run defense ranked 24th of 185. Michigan State scored 30 on Toledo without needing either number to be large, which is the shape of this offense.
Michigan State's advantages are efficiency advantages: staying on schedule, winning short yardage, denying the run. Those produce controlled wins, and the projected 35-31 finish is what they are worth against an opponent that can score from anywhere on the field. Eastern Michigan also returns more of its own production than Michigan State does, which is why the gap between two 4-8 teams sat at 4.3 points after the offseason instead of widening.
Eastern Michigan's run defense can take away the thing Michigan State does best
Eastern Michigan has allowed 22.0 points a game and gave up 27 to San Jose State in a 21-27 loss its own offense lost. The front is the unit that matters Saturday: Michigan State's structural advantages are the run game and short yardage, and Eastern Michigan's run defense, 24th of 185, is the one Michigan State strength it can genuinely contest. Force Michigan State to throw, and Milivojevic, with no snaps this season, has to solve a pass defense ranked 58th of 185.
Points allowed per drive across all FBS defenses (lower is better). Eastern Michigan's defense is below average, better than only 0% of FBS defenses.
Empty possessions cost Eastern Michigan more than they cost the home team. The visitors got 14 drives against San Jose State and scored on 14.3% of them, while Michigan State scored on 63.6% of 11 against Toledo and owns the run edge to keep drives long. Eastern Michigan can survive a quiet quarter from Kim and Devereaux. It cannot survive another 42.9% three-and-out rate and still be inside a touchdown late.
The winner and the number are two different questions
The projection gives Michigan State a 62% chance to win and only a 24% chance to cover 17.2, a methodical favorite against a number priced for a teardown. It gives a one-score finish a 37% chance and the under just 13%, so the scoreboard is expected to move rather than stall in the twenties. The number, not the winner, is where Eastern Michigan lives on Saturday.
Eastern Michigan's route is narrow but plain: keep the possession count low, break even on turnovers, and turn two of Kim's downfield throws into touchdowns rather than field goals. Devereaux, at a 51% touchdown chance on 25 to 79 receiving yards, is the likeliest of those. The keep-it-close criteria are more achievable than the stack of things Michigan State needs for a three-score win, which is why 17.2 asks more of the favorite than the result does.
What has to happen
Michigan State needs the 9.1% three-and-out rate it posted against Toledo to carry over, and it needs the run edge to produce touchdowns rather than just clock. A favorite that trades field position all afternoon and wins 35-31 has beaten Eastern Michigan and lost the number. Holding Blue to 28 to 69 and Bennett to 14 to 34 is the necessary condition; turning those stops into short fields is the sufficient one.
Each ingredient of our projection and the side it favors — bars toward Michigan State are why it's the pick (home field is counted separately).
Eastern Michigan needs first downs before it needs explosive plays. Every chunk-play edge it holds is worthless from second-and-nine, and staying on schedule is where Michigan State grades better by a standard deviation and a half. Get to neutral downs, let Kim work the middle of 181 to 327, and the visitors have the most dangerous single unit on the field. Fail there, and 2014 stops looking as irrelevant as it should.
Frequently asked questions
- Who is predicted to win Eastern Michigan vs Michigan State?
- Our projection favors Michigan State with a 62% win probability. The projected final score is Michigan State 35–Eastern Michigan 31.
- What is the predicted score for Eastern Michigan vs Michigan State?
- We project Michigan State 35–Eastern Michigan 31, a combined 66 points.
- What is the spread for Eastern Michigan vs Michigan State, and which side does the model like?
- The betting line is Michigan State −17.2. Our projection has Michigan State winning by about 4, inside the number — the points are the value side.

Eastern Michigan wins
Discussion
Weigh in on the analysis — the best takes rise to the top.
0 Replies
Sign in to join the discussion.