Following up a post I made last week, I'm continuing to find an interesting pattern in IU's performances in the B1G. I'm not going to go over the math in detail again, but this is all based on actual performance vs. what's expected, based on each teams (adjusted for opponent) average margin of victory. The current rankings:
Rank Team Avg
1 Wiscy 11.05
2 OSU 6.12
3 Iowa 3.21
4 PU 2.72
5 MSU 2.59
6 Ind -0.11
7 Minny -0.30
8 Mary -0.83
9 PSU -1.25
10 Ill -1.30
11 UM -1.82
12 Neb -3.22
13 NW -5.23
14 RU -9.54
Each number represents how a team should expect to perform against a perfectly average B1G team on a neutral court. Anyway, here's what's interesting. If you take a list of games we've played and rank them according to expected result, you get this:
Opp IU MOV Pred O/U
RU 8 13.3 -5.3
UM 3 5.6 -2.6
PSU 3 5.0 -2.0
Mary 19 4.6 14.4
Neb 5 -0.8 5.8
OSU 3 -2.4 5.4
Ill 6 -2.7 8.7
MSU -20 -6.6 -13.4
PU -16 -6.7 -9.3
osu -12 -10.1 -1.9
Wiscy -14 -15.0 1.0
The first number is our actual MOV, the second is the expected, and the third is the difference, i.e., how much we over- or under-performed by. The interesting pattern that emerges is that our team consistently underperforms in games in which they are heavy favorites or dogs (minor exception for Wiscy), and consistently overperforms in those games which are expected to be closely contested (with a consistent but interesting outlier against Maryland). This isn't an artifact of the math. If it were, it would skew toward one end, rather than the middle. This is a team-specific trend. For whatever reasons, we are one of those teams that lays down for the big boys, plays down to the scrubs, and really knows how to turn it on for the tight matchups.
What's especially interesting is that the rest of our schedule is made up primarily of games which are expected to be close:
Mary -3.2
Minny 4.1
PU 1.1
RU 5.6
NW 1.2
Iowa 0.6
MSU 1.2
The math suggests that we will win 4 (actually, 4.04) more games to finish 11-7. But based on our pattern so far, it seems there is a very strong chance that we could outperform that, if current trends hold. In other words, my conclusion is the roughly same as it was last week: I'm expecting 11 wins, but the difference between 11 and 14 is very, very small.
goat
NB: If this formatting works, h/t Guy Fawkes.
This post was edited on 2/8 8:47 PM by TheOriginalHappyGoat
Rank Team Avg
1 Wiscy 11.05
2 OSU 6.12
3 Iowa 3.21
4 PU 2.72
5 MSU 2.59
6 Ind -0.11
7 Minny -0.30
8 Mary -0.83
9 PSU -1.25
10 Ill -1.30
11 UM -1.82
12 Neb -3.22
13 NW -5.23
14 RU -9.54
Each number represents how a team should expect to perform against a perfectly average B1G team on a neutral court. Anyway, here's what's interesting. If you take a list of games we've played and rank them according to expected result, you get this:
Opp IU MOV Pred O/U
RU 8 13.3 -5.3
UM 3 5.6 -2.6
PSU 3 5.0 -2.0
Mary 19 4.6 14.4
Neb 5 -0.8 5.8
OSU 3 -2.4 5.4
Ill 6 -2.7 8.7
MSU -20 -6.6 -13.4
PU -16 -6.7 -9.3
osu -12 -10.1 -1.9
Wiscy -14 -15.0 1.0
The first number is our actual MOV, the second is the expected, and the third is the difference, i.e., how much we over- or under-performed by. The interesting pattern that emerges is that our team consistently underperforms in games in which they are heavy favorites or dogs (minor exception for Wiscy), and consistently overperforms in those games which are expected to be closely contested (with a consistent but interesting outlier against Maryland). This isn't an artifact of the math. If it were, it would skew toward one end, rather than the middle. This is a team-specific trend. For whatever reasons, we are one of those teams that lays down for the big boys, plays down to the scrubs, and really knows how to turn it on for the tight matchups.
What's especially interesting is that the rest of our schedule is made up primarily of games which are expected to be close:
Mary -3.2
Minny 4.1
PU 1.1
RU 5.6
NW 1.2
Iowa 0.6
MSU 1.2
The math suggests that we will win 4 (actually, 4.04) more games to finish 11-7. But based on our pattern so far, it seems there is a very strong chance that we could outperform that, if current trends hold. In other words, my conclusion is the roughly same as it was last week: I'm expecting 11 wins, but the difference between 11 and 14 is very, very small.
goat
NB: If this formatting works, h/t Guy Fawkes.
This post was edited on 2/8 8:47 PM by TheOriginalHappyGoat
