Adv. Stats Win Probabilities Rehash - Week 12 Results

Submitted by Ecky Pting on

So, here’s the latest Total Wins Probabilities for U-M, Sparty, OSU and PSU (even though they’re out of it, they still factor in), updated with the Week 12 results. As was done last week, I’ll show both S&P+ and FPI-based Total Wins probability distributions in the embiggable charts below:

S&P+ Total Wins Prob Comp 2015 wk 12

FPI Total Wins Prob Comp 2015 wk 12

The S&P+ and FPI win probabilities for each team's remaining games that inform the charts above are shown in the table below, along with the wins expectation values:

Michigan
Michigan State
Ohio State
Penn State
Opp
S&P+
FPI
Opp
S&P+
FPI
Opp
S&P+
FPI
Opp
S&P+
FPI
OSU
61%
42.3%
PSU
65%
79.6%
@ U-M
39%
57.7%
@ MSU
35%
20.4%
Wins:
9.6
9.4
 
10.7
10.8
 
10.4
10.6
 
7.3
7.2

Mr. Obvious: “Does this mean ‘Sparty … yes?’”

Mr. Smarty-pants: “Just zip it. I’m not going there.”

So here’s one last look at the charts – not much to be gleaned other than:

  • Sparty is now the most likely to win by both estimates, followed narrowly by U-M in the S&P+ estimate, or OSU by a wider margin under FPI.
  • The converse of that is, of course, PSU is the least likely to win.

Of course, with only one game remaining, the pathways by which U-M, MSU or OSU, are quite easily understood. The probabilities of each scenario are quantified below, which basically amounts to subdividing the likelihood that MSU would lose by that of either U-M or OSU winning:

Scenario
S&P+

Probability
FPI

Probability
B1GE

Champ
UM wins

& MSU loses
21.4%
8.6%
U-M
 
MSU wins out
65.0%
79.6%
MSU
 
OSU wins

& MSU loses
13.7%
11.8%
OSU

So, yea. U-M & OSU, by virtue of OSU taking a dive last week, are left to fight for the statistical table-scraps of Sparty’s matchup with PSU. U-M's chances have eroded by about 1/3, thanks to the Buckeyes having screwed the pooch. Frankly, OSU deserves to be pummeled just for that bit of incompetence, so as to show them what sorry losers they really are. But, if you’re looking for some sort of solace in these numbers, then I’ll offer you this:

The likelihood of PSU beating MSU in East Lansing, is better than the likelihood was of Sparty (S&P: 20%, FPI: 20.9%) beating OSU last week.

Mr. Obvious: "Oh great. So now what's the chance of lightning striking twice..."

Mr. Smarty-pants: "Well, since the first event has already occurred ... the likelihood of the second event is the same as it was for the first. Or as your broker might say, 'Past performance is no guarantee of future returns.'"

Finally, here's the picture of how all of these probabilities (and those for the rest of the B1G) impact the projected end-of-season conference standings resulting from the S&P+ predictions:

B1G East
B1G
Overall
W
L
W
L
Michigan State *
7
1
11
1
Michigan
7
1
10
2
Ohio State
6
2
10
2
Penn State
4
4
7
5
Indiana
2
6
6
6
Maryland *
1
7
3
9
Rutgers
1
7
4
8

* - winner of tie-breaker

B1G West
B1G
Overall
W
L
W
L
Iowa
8
0
12
0
Northwestern
6
2
10
2
Wisconsin
5
3
8
4
Nebraska *
3
5
5
7
Minnesota
3
5
6
6
Illinois
2
6
5
7
Purdue
1
7
2
10

And following are the standings resulting from the FPI predictions:

B1G East
B1G
Overall
W
L
W
L
Michigan State *
7
1
11
1
Ohio State
7
1
11
1
Michigan
6
2
9
3
Penn State
4
4
7
5
Rutgers *
2
6
5
7
Indiana
2
6
6
6
Maryland
0
8
2
10

* - winner of tie-breaker

B1G West
B1G
Overall
W
L
W
L
Iowa
8
0
12
0
Northwestern *
6
2
10
2
Wisconsin
6
2
9
3
Nebraska
3
5
5
7
Minnesota *
2
6
5
7
Illinois
2
6
5
7
Purdue
1
7
2
10

Happy Thanksgiving to everyone, and Go Blue!

Comments

creelymonk10

November 25th, 2015 at 4:14 PM ^

Love the stats and charts as always. Talking to any MSU fan this week you'd never guess they had a 35% (or 20% using FPI) chance of not even making the Big Ten Championship Game. Most think it's a foregone conclusion that they're going to the playoffs.

Michigan's 21% outcome would be the sweetest of sweet days.