Defining “success” for Northwestern’s 2016 football season

Posted on insidenu.com:
https://www.insidenu.com/2016/8/23/12618842/defining-success-for-northwesterns-2016-football-season

This expands a bit on a comment I made on the “Wildcat Shootaround: What would you consider a successful season for Northwestern?” post.

Heading into last season, would anyone have thought 10 wins wasn’t a successful season? Yet some of the definitions of success for this season would deem last year unsuccessful. Right off the bat, Zach gives three criteria, and last year met NONE of them.

(Zach’s criteria were: “First, beat Iowa. Second, have no blowout losses. Third, win a bowl game.”)

Success is highly contextual. I graduated in 1994. A winning season back then would have been wildly successful. Three wins in some years felt like a success.

How do we create some context to define success? I built a simple model. I defined success as wins for the season (with no regard for margin of victory, number of games played, strength of schedule, etc). I used a 4-year weighted moving average (40/30/20/10) to give the over/under each season for wins needed to define success (Success Wins). Why four years? It’s the playing length of a standard college career; with a few exceptions, a roster turns over every four years.

I expanded the data set to go back to the start of the Ara Parseghian era; this correlates pretty closely to the formation of the Big Ten.

Season Coach Record Success Wins Success? +/-
1956 Ara Parseghian 4-4-1 1.4 yes 2.6
1957 Ara Parseghian 0-9 2.3 no -2.3
1958 Ara Parseghian 5-4 1.4 yes 3.6
1959 Ara Parseghian 6-3 2.8 yes 3.2
1960 Ara Parseghian 5-4 4.3 yes 0.7
1961 Ara Parseghian 4-5 4.8 no -0.8
1962 Ara Parseghian 7-2 4.8 yes 2.2
1963 Ara Parseghian 5-4 5.6 no -0.6
1964 Alex Agase 3-6 5.4 no -2.4
1965 Alex Agase 4-6 4.5 no -0.5
1966 Alex Agase 3-6-1 4.2 no -1.2
1967 Alex Agase 3-7 3.5 no -0.5
1968 Alex Agase 1-9 3.2 no -2.2
1969 Alex Agase 3-7 2.3 yes 0.7
1970 Alex Agase 6-4 2.4 yes 3.6
1971 Alex Agase 7-4 3.8 yes 3.2
1972 Alex Agase 2-9 5.3 no -3.3
1973 John Pont 4-7 4.4 no -0.4
1974 John Pont 3-8 4.2 no -1.2
1975 John Pont 3-8 3.5 no -0.5
1976 John Pont 1-10 3.1 no -2.1
1977 John Pont 1-10 2.3 no -1.3
1978 Rick Venturi 0-10-1 1.6 no -1.6
1979 Rick Venturi 1-10 0.8 yes 0.2
1980 Rick Venturi 0-11 0.7 no -0.7
1981 Dennis Green 0-11 0.4 no -0.4
1982 Dennis Green 3-8 0.2 yes 2.8
1983 Dennis Green 2-9 1.3 yes 0.7
1984 Dennis Green 2-9 1.7 yes 0.3
1985 Dennis Green 3-8 2 yes 1
1986 Francis Peay 4-7 2.5 yes 1.5
1987 Francis Peay 2-8-1 3.1 no -1.1
1988 Francis Peay 2-8-1 2.8 no -0.8
1989 Francis Peay 0-11 2.5 no -2.5
1990 Francis Peay 2-9 1.4 yes 0.6
1991 Francis Peay 3-8 1.4 yes 1.6
1992 Gary Barnett 3-8 2 yes 1
1993 Gary Barnett 2-9 2.5 no -0.5
1994 Gary Barnett 3-7-1 2.5 yes 0.5
1995 Gary Barnett 10-2 2.7 yes 7.3
1996 Gary Barnett 9-3 5.6 yes 3.4
1997 Gary Barnett 5-7 7.4 no -2.4
1998 Gary Barnett 3-9 7 no -4
1999 Randy Walker 3-8 5.5 no -2.5
2000 Randy Walker 8-4 4 yes 4
2001 Randy Walker 4-7 5.2 no -1.2
2002 Randy Walker 3-9 4.9 no -1.9
2003 Randy Walker 6-7 4.3 yes 1.7
2004 Randy Walker 6-6 4.9 yes 1.1
2005 Randy Walker 7-5 5.2 yes 1.8
2006 Pat Fitzgerald 4-8 6.1 no -2.1
2007 Pat Fitzgerald 6-6 5.5 yes 0.5
2008 Pat Fitzgerald 9-4 5.6 yes 3.4
2009 Pat Fitzgerald 8-5 6.9 yes 1.1
2010 Pat Fitzgerald 7-6 7.5 no -0.5
2011 Pat Fitzgerald 6-7 7.6 no -1.6
2012 Pat Fitzgerald 10-3 7 yes 3
2013 Pat Fitzgerald 5-7 8 no -3
2014 Pat Fitzgerald 5-7 6.9 no -1.9
2015 Pat Fitzgerald 10-3 6.1 yes 3.9
2016 Pat Fitzgerald 7.5

You get a successful season about half the time. I’m OK with that, though I’d guess some people think that’s too easy a hurdle to clear. But to me it means we’re moving in an overall positive direction, and that’s successful.

The Success Wins for 2016 is 7.5. Seems about right for where we are right now. For the 2016 season, 8 wins = success.

Interesting to note that NU’s high-water mark for Success Wins came after the 2012 season at 8.0. If we win 8 games this year, next year’s mark would be 7.7. But if we win 9 games, it moves to 8.1. We are at or near the peak of the program.

We’re not talking about the low-water mark.

What are our most successful seasons? Here are the top 10.

Season Coach Record Success Wins +/- Rank
1995 Gary Barnett 10-2 2.7 7.3 1
2000 Randy Walker 8-4 4 4 2
2015 Pat Fitzgerald 10-3 6.1 3.9 3
1958 Ara Parseghian 5-4 1.4 3.6 4
1970 Alex Agase 6-4 2.4 3.6 4
1996 Gary Barnett 9-3 5.6 3.4 6
2008 Pat Fitzgerald 9-4 5.6 3.4 6
1959 Ara Parseghian 6-3 2.8 3.2 8
1971 Alex Agase 7-4 3.8 3.2 8
2012 Pat Fitzgerald 10-3 7 3 10

Our 1995 season may well be the all-time college football most “successful” season by this metric. Feel free to leave a comment with any outlier seasons you can think of for other teams and we’ll see how they compare. Pat Fitzgerald has three of Northwestern’s top 10 seasons since 1956.

We can use this to rank coaches too. What Success Win level did they inherit, and what Success Win level did they leave?

Coach Seasons Start End +/- Rank
Ara Parseghian 1956-63 1.4 5.4 4 1
Gary Barnett 1992-98 2 5.5 3.5 2
Dennis Green 1981-85 0.7 2.5 1.8 3
Pat Fitzgerald 2006- 6.1 7.5 1.4 4
Randy Walker 1999-2005 5.5 6.1 0.6 5
Francis Peay 1986-90 2.5 2 -0.5 6
Rick Venturi 1978-80 1.6 0.7 -0.9 7
Alex Agase 1964-72 5.4 4.4 -1 8
John Pont 1973-77 4.4 1.6 -2.8 9

I don’t think anyone would put Dennis Green that high, but given the context of what he inherited, it makes some sense. It gives a lot of credit to Barnett (as it should!) for turning things around and taking us up a few levels. Randy Walker moved it up a bit more, and Pat Fitzgerald has taken the program to its highest point yet. If you ranked coaches by the absolute (not relative) level they achieved, Fitzgerald is #1.

I did notice that after the 2014 season, we were at 6.1, exactly what Fitzgerald inherited. There was a segment of the fan base (at least one person!) who felt we were stuck. Then 2015 got us moving back in a positive direction.

How does this look for B1G play? Our Success Conference Wins for 2016 is 4.0, though that’s based on an 8-game slate. That’s a .500 win percentage for 8 games, so we can convert that to 4.5 wins for the 9-game 2016 slate.

How does every team in the Big Ten look?

Team 2012 2013 2014 2015 2016 Rank
Ohio State 12 12 14 12 12.6 1
Michigan State 7 13 11 12 11.4 2
Wisconsin 8 9 11 10 9.9 3
Iowa 4 8 7 12 8.9 4
Nebraska 10 9 9 6 7.9 5
Michigan 8 7 5 10 7.7 6
Northwestern 10 5 5 10 7.5 7
Penn State 8 7 7 7 7.1 8
Minnesota 6 8 8 6 7 9
Rutgers 9 6 8 4 6.1 10
Maryland 4 7 7 3 5.1 11
Indiana 4 5 4 6 5 12
Illinois 2 4 6 5 4.8 13
Purdue 6 1 3 2 2.5 14

Wow, Ohio State. Just win 13 games to be successful! Or, anything short of the playoff is not a successful season. That’s crazy. Michigan seems low, but my model doesn’t know that they hired Jim Harbaugh to replace Brady Hoke. Wisconsin seems high. LOL Illinois. Northwestern’s above average! Seems reasonable for where we are at this point.

TL;DR Success for the 2016 Northwestern football season is 8+ total wins, 5+ conference wins. With 9 wins, the program will be at its best level ever. THE MODEL HAS SPOKEN.