That Nebraska beat Minnesota in Lincoln shouldn’t have been particularly noteworthy. The Huskers have now won five of eight in the series since joining the Big Ten, 19 of 22 going back to 1963. The 25-point margin wasn’t all that noticeable either — it’s been a lot worse in this series.
Until vanquishing the Gophers, though, Nebraska was off to a historically awful start in Scott Frost’s first season — the Cornhuskers were 0-6 for the first time in their 129-year history. The program’s fortunes have slid dramatically over the 20 years since legendary coach Tom Osborne retired. With a loss to Troy, blowout road losses to Michigan and Wisconsin, and a spectacular blown lead at Northwestern, this fall seemed to represent a bottoming out.
It was awful only in the win column, though. Based on actual team quality, Frost’s rebuild has gone almost exactly as projected.
Each year in my college football stat profiles, you’ll see something called “win expectancy” in each team’s “Schedule & Results” section. The definition from my last advanced stats glossary:
It is intended to say “Given your success rates, big plays, field position components, turnovers, etc., you could have expected to win this game X% of the time.” It has nothing to do with pre-game projections or opponent adjustments.
For preview purposes, I note when teams stray pretty far from their postgame win expectations, one way or the other. For example:
Five biggest 2017 overachievers, per postgame win expectancy:
- Army +2.8 (7.2 expected wins, 10 actual wins)
- Troy +2.2 (8.8 expected wins, 11 actual wins)
- Akron +2.2 (4.8 expected wins, seven actual wins)
- Northwestern +2.1 (7.9 expected wins, 10 actual wins)
- Kansas State +1.9 (6.1 expected wins, eight actual wins)
Five biggest 2017 underachievers, per postgame win expectancy:
- Miami (Ohio) -2.8 (7.8 expected wins, five actual wins)
- Baylor -2.1 (3.1 expected wins, one actual win)
- Ole Miss -1.8 (7.8 expected wins, six actual wins)
- Arkansas State -1.8 (8.8 expected wins, seven actual wins)
- Bowling Green -1.5 (3.5 expected wins, two actual wins)
Over time, this can tell us a little bit about certain coaches. As I note in an annual post at Football Study Hall, the coaches at the top and bottom of the lists tend to be the ones fans say are particularly good or bad coaches.
Mostly, it is a sign of which teams have endured random bounces. And it can be a sign with giant, blinking “TURNAROUND COMING” lights.
The five overachievers above went a combined 46-20 last season, a win percentage of 0.697. So far this season, they’re down to a combined 20-14 (0.588).
Meanwhile, the five biggest 2017 underachievers went 21-39 (0.350) last year and are thus far up to 17-21 (0.447).
Those are actually pretty tame changes compared to previous years.
In 2016, for instance, the biggest overachievers were Idaho and WVU, which went a combined 19-7 despite only 14.5 expected wins.
In 2017, they fell to 11-14.
2016’s biggest underachievers were Notre Dame and Michigan State, which went a combined 7-17 despite 12.2 expected wins.
In 2017, they improved to 20-6.
The further you get toward the poles, the more likely the turnaround becomes.
- From 2005-16, 33 teams had a win total at least two games worse than than their expected win totals. They improved by an average of 2.8 wins the next season. Only two saw their wins decrease.
- On the other side, 42 teams finished with a win total at least two games better than their expected win totals. They regressed by an average of 2.4 wins the next year. Only four teams improved on their win total.
If we break this into win differences per game, the record for poor fortune was set by SMU in 2007.
The Mustangs finished 1-11 despite an expected win total of 4.7, an average of minus-0.31 wins per game.
Three teams have exceeded that mark thus far in 2018.
- Nebraska: minus-0.40 per game (one win, 3.8 expected wins)
- Texas State: minus-0.34 (one win, 3.4 expected)
- UTEP: minus-0.32 (zero wins, 2.2 expected)
Now, there’s nothing saying they will continue on this path — a turnaround in fortune won’t have to wait until 2019, after all. But when S&P+ picks games, is basically views NU and TXST as 4-3 teams, and UTEP more like 2-5.
The Huskers’ awful fortunes didn’t begin with the crazy loss to Northwestern — four of NU’s six losses featured at least a 38 percent post-game win expectancy, meaning stats saw tossups.
- Colorado 33, Nebraska 28 (NU’s post-game probability: 95 percent)
- Troy 24, Nebraska 19 (55 percent)
- Purdue 42, Nebraska 28 (38 percent)
- Northwestern 34, Nebraska 31 (84 percent)
Based on post-game stats, NU’s chances of going 4-0 in these games were much higher (17 percent) than their chances of going 0-4 (0.2 percent). But 0-4 they went.
After rising eight spots with the Minnesota win, the Huskers are ranked 61st in S&P+ — exactly where they were projected in the preseason.
Before the year, S&P+ projected about 5.3 wins for NU, and it would have been closer to 5.6 had there been a projection that featured Bethune-Cookman on the schedule instead of Akron.
The most likely record now is 3-9 or 4-8, but it’s not hard to theorize that, given a normal number of breaks, NU could have gone something like 7-5.
To say the least, we would’ve been pretty impressed with 7-5, given the youth on Nebraska’s two-deep (freshman starting QB, freshman backup RB, sophomore leading receiver, three sophomore starters on the O-line, and 10 freshmen or sophomores on the defensive depth chart).
So when you find yourself next offseason thinking, “Man, Nebraska is getting awfully overhyped this year … as always…” know that a) you might be right, but b) the stats probably agree with the hype.
The numbers will view Nebraska as building on a team significantly better than one that went 3-9 or so.
By the way, who’s perched for a fall, either down the stretch in 2018 or in 2019?
Here is your current list of the biggest 2018 underachievers, per post-game win expectancy:
- Texas: plus-1.6 wins (six wins, 4.4 expected)
- FIU: plus-1.5 (five wins, 3.5 expected)
- Liberty: plus-1.5 (four wins, 2.5 expected)
- Louisiana Tech: plus-1.5 (five wins, 3.5 expected)
- UTSA: plus-1.4 (three wins, 1.6 expected)
The overachievers haven’t defied the math quite as much as the underachievers, but keep it in mind all the same.