Stat Junkie Stuff: Looking at Expected Wins on the 2020 Schedule
Post-Script on 2019
In 2019 we are 9-3 losing games to Wisconsin, PSU, and OSU
In a home field advantage adjusted model using Sagarin predicted those losses. It did predict the ND Win, but not the margin. Before the season the same model predicted Wisconsin and PSU as razor thin wins, effectively tossups.
2019 Calculated Losses using the Home Field Advantage (HFA) adjusted Model - Obviously margins are not correct
Wisconsin: 7 point loss
Penn St.: 4 point loss
Ohio State: 9 point loss
2020 Season Prediction (using HFA Adjusted Model)
Predicted Record: 11-1
W - Washington - ROAD: 3 points
W - Ball St - HOME: 35 points
W - Arkansas St. - HOME: 33 points
W - Wisconsin - HOME: 3 points
W - Penn St - HOME: 6 points
W - Michigan State - ROAD: 11 points
W - Minnesota - ROAD: 6 points
W - Purdue - HOME: 27 points
W - Maryland - HOME: 34 points
W - Rutgers - ROAD: 35 points
W - Indiana - HOME: 20 points
L - Ohio State - ROAD: 21 points
In absolute terms this schedule shows 11-1 but the games @Washington, @Minnesota, and Wisconsin & PSU at home will be absolute battles and all show as wins in this model.
This model also assumes that teams are roughly the same strength in 2020 (not literally true but usually true)
The middle of the season has 4 consecutive games: Wisconsin, Penn St, MSU, Minnesota that will be very competitive.
2020 WIN PROBABILITIES
When we apply "win probabilities" to all of the spreads (We can't win every close game)
This 2020 schedule produces 9.2 expected wins
The same as the 9.2 expected wins from the 2019 schedule
Effectively, this says Michigan's schedule difficulty in 2020 is the same as 2019
Thoughts?
Do you agree or disagree?
Too Optimistic?
Too Pessimistic?
Post-Script on 2019
In 2019 we are 9-3 losing games to Wisconsin, PSU, and OSU
In a home field advantage adjusted model using Sagarin predicted those losses. It did predict the ND Win, but not the margin. Before the season the same model predicted Wisconsin and PSU as razor thin wins, effectively tossups.
2019 Calculated Losses using the Home Field Advantage (HFA) adjusted Model - Obviously margins are not correct
Wisconsin: 7 point loss
Penn St.: 4 point loss
Ohio State: 9 point loss
2020 Season Prediction (using HFA Adjusted Model)
Predicted Record: 11-1
W - Washington - ROAD: 3 points
W - Ball St - HOME: 35 points
W - Arkansas St. - HOME: 33 points
W - Wisconsin - HOME: 3 points
W - Penn St - HOME: 6 points
W - Michigan State - ROAD: 11 points
W - Minnesota - ROAD: 6 points
W - Purdue - HOME: 27 points
W - Maryland - HOME: 34 points
W - Rutgers - ROAD: 35 points
W - Indiana - HOME: 20 points
L - Ohio State - ROAD: 21 points
In absolute terms this schedule shows 11-1 but the games @Washington, @Minnesota, and Wisconsin & PSU at home will be absolute battles and all show as wins in this model.
This model also assumes that teams are roughly the same strength in 2020 (not literally true but usually true)
The middle of the season has 4 consecutive games: Wisconsin, Penn St, MSU, Minnesota that will be very competitive.
2020 WIN PROBABILITIES
When we apply "win probabilities" to all of the spreads (We can't win every close game)
This 2020 schedule produces 9.2 expected wins
The same as the 9.2 expected wins from the 2019 schedule
Effectively, this says Michigan's schedule difficulty in 2020 is the same as 2019
Thoughts?
Do you agree or disagree?
Too Optimistic?
Too Pessimistic?
Last edited:
