Measuring our SOS for the past 15 years

[Go Gophers Rah]

In another thread, someone mentioned that 2010 will likely be one of our most difficult schedules ever. I don't disagree with that, but I wanted to try to apply some analysis to it.

There are probably alot of ways to look at this, but here is the method I chose:

The years when we played the most teams that were mentioned in (including "also receiving votes") the AP Poll at both the time we played them and in the final poll of the year (and the percent of such games on the schedule to account for 11-13 game seasons):

1995: 4 (36%)
1996: 5 (45%)
1997: 7 (64%)
1998: 4 (36%)
1999: excluding the bowl: 4 (36%) including the bowl: 5 (42%)
2000: excluding the bowl: 3 (27%) including the bowl: 4 (33%)
2001: 3 (27%)
2002: excluding the bowl: 3 (25%) including the bowl: 4 (31%)
2003: excluding the bowl: 2 (17%) including the bowl: 2 (15%) - might explain the 10-win season
2004: excluding the bowl: 4 (36%) including the bowl: 4 (33%)
2005: excluding the bowl: 5 (45%) including the bowl: 5 (42%)
2006: excluding the bowl: 5 (42%) including the bowl: 5 (38%)
2007: 4 (33%)
2008: excluding the bowl: 3 (25%) including the bowl: 4 (31%)
2009: excluding the bowl: 4 (33%) including the bowl: 4 (31%)


If you exclude the "also receiving votes" and only include the true Top-25 teams, then the numbers don't change all that much, with the 3 toughest seasons being:
1996: 5 (45%)
1998: 4 (36%)
2006: excluding the bowl: 4 (33%) including the bowl: 4 (31%)

I could see us playing in a season in which 5 teams are ranked in the Top 25 more-or-less throughout the season; those 5 being:USC, Wisconsin, Penn St., Ohio State and Iowa.
I think it would be highly unlikely (but not impossible) that MTSU or Michigan State could have a vote or 2 at the time we play them and at the end of the season, meaning that 7 teams could be in the "also receiving votes" or better category.

So, yes, I do think 2010 will be as difficult as 1996 and likely more difficult than any schedule we've played since.

What this doesn't measure is how few "weak" teams we play. I'm not going to do the analysis on that. In that respect 2007 was probably our most difficult year - one in which only 1 opponent finished the year with a losing record (the one opponent we did beat - Miami (OH) who ended up 6-7).

I would also note that 2 of the 3 other most difficult SOS years resulted in the firing of our head coach, so I'm not sure that the decision makers are that sympathetic to SOS.

Probably more info than anyone else cares about, but I didn't want to judge the difficulty of 2010 without thinking it through.

 

[wdgold]

In another thread, someone mentioned that 2010 will likely be one of our most difficult schedules ever. I don't disagree with that, but I wanted to try to apply some analysis to it.

There are probably alot of ways to look at this, but here is the method I chose:

The years when we played the most teams that were mentioned in (including "also receiving votes") the AP Poll at both the time we played them and in the final poll of the year (and the percent of such games on the schedule to account for 11-13 game seasons):

1995: 4 (36%)
1996: 5 (45%)
1997: 7 (64%)
1998: 4 (36%)
1999: excluding the bowl: 4 (36%) including the bowl: 5 (42%)
2000: excluding the bowl: 3 (27%) including the bowl: 4 (33%)
2001: 3 (27%)
2002: excluding the bowl: 3 (25%) including the bowl: 4 (31%)
2003: excluding the bowl: 2 (17%) including the bowl: 2 (15%) - might explain the 10-win season
2004: excluding the bowl: 4 (36%) including the bowl: 4 (33%)
2005: excluding the bowl: 5 (45%) including the bowl: 5 (42%)
2006: excluding the bowl: 5 (42%) including the bowl: 5 (38%)
2007: 4 (33%)
2008: excluding the bowl: 3 (25%) including the bowl: 4 (31%)
2009: excluding the bowl: 4 (33%) including the bowl: 4 (31%)


If you exclude the "also receiving votes" and only include the true Top-25 teams, then the numbers don't change all that much, with the 3 toughest seasons being:
1996: 5 (45%)
1998: 4 (36%)
2006: excluding the bowl: 4 (33%) including the bowl: 4 (31%)

I could see us playing in a season in which 5 teams are ranked in the Top 25 more-or-less throughout the season; those 5 being:USC, Wisconsin, Penn St., Ohio State and Iowa.
I think it would be highly unlikely (but not impossible) that MTSU or Michigan State could have a vote or 2 at the time we play them and at the end of the season, meaning that 7 teams could be in the "also receiving votes" or better category.

So, yes, I do think 2010 will be as difficult as 1996 and likely more difficult than any schedule we've played since.

What this doesn't measure is how few "weak" teams we play. I'm not going to do the analysis on that. In that respect 2007 was probably our most difficult year - one in which only 1 opponent finished the year with a losing record (the one opponent we did beat - Miami (OH) who ended up 6-7).

I would also note that 2 of the 3 other most difficult SOS years resulted in the firing of our head coach, so I'm not sure that the decision makers are that sympathetic to SOS.

Probably more info than anyone else cares about, but I didn't want to judge the difficulty of 2010 without thinking it through.

Win Baby Win that is all the decision makers are worried about not SOS like you said.

 

[Tater]

Great post. Love to see the raw info and how it compares to how we did each year. I'm not sure how else you could do this other than by using the opponent's actual records, which could be more misleading when considering certain non-con teams' records (like lower division teams or relatively good non-con teams from bad conferences).