Michigan's non-conference scheduling luck
Apologies for the earlier formating fail.
After reading Hannibal's excellent review of Michigan's bad scheduling luck in the Big 10, I was curious to see what the results for our non-conference games would look like through the same lens. Thanks to James Howell and Chris Stassen, all the data required is readily available. In keeping with Hannibal's treatment, I have considered all non-conference games since the Big11Ten entered its current format in 1993. Michigan's game against each team has NOT been subtracted from that team's record, because (1) our game against each team was part of that team's overall level of success that year, and (2) I'm lazy. The only team whose overall win-loss record was likely to be impacted significantly over the course of 16 years just from playing us was Notre Dame; their 6-6 record against us over this span is slightly lower than their net 0.616 success percentage over the same span (success percentage defined as (wins+0.5*ties)/(wins+ties+losses). If I get ambitious, maybe I'll run the numbers again with games against us subtracted out, but I suspect it won't change the conclusions much.So without further ado, here are the numbers:
Team Game score Record 1993-2008 Cumulative
Utah L 23-25 13-0 (1.000) 130-61 (0.681)
Miami (OH) W 16-6 2-10 (0.167) 106-83-1 (0.561)
Notre Dame L 17-35 7-6 (0.538) 119-74-1 (0.616)
Toledo L 10-13 3-9 (0.250) 118-70-2 (0.626)
2007
The Horror L 32-34 13-2 (0.867) 148-56 (0.726)
Oregon L 7-39 9-4 (0.692) 130-65 (0.667)
Notre Dame W 38-0 3-9 (0.250) 119-74-1 (0.616)
Eastern Mich W 33-22 4-8 (0.333) 56-125 (0.309)
2006
Vanderbilt W 27-7 4-8 (0.333) 50-127 (0.283)
Central Mich W 41-17 10-4 (0.714) 83-103 (0.446)
Notre Dame W 47-21 10-3 (0.769) 119-74-1 (0.616)
Ball State W 34-26 5-7 (0.417) 84-100-2 (0.457)
2005
Northern Ill. W 33-17 7-5 (0.583) 81-104 (0.438)
Notre Dame L 10-17 9-3 (0.750) 119-74-1 (0.616)
Eastern Mich W 55-0 4-7 (0.364) 56-125 (0.309)
2004
Miami (OH) W 43-10 8-5 (0.615) 106-79 (0.573)
Notre Dame L 20-28 6-6 (0.500) 119-74-1 (0.616)
San Diego St. W 24-21 4-7 (0.364) 77-110 (0.412)
2003
Central Mich W 45-7 3-9 (0.250) 83-103 (0.446)
Houston W 50-3 7-6 (0.539) 74-113-1 (0.396)
Notre Dame W 38-0 5-7 (0.417) 119-74-1 (0.616)Oregon L 27-31 8-5 (0.615) 130-65 (0.667)
2002
Washington W 31-29 7-6 (0.539) 95-94-1 (0.503)
Western Mich W 35-12 4-8 (0.333) 98-86-1 (0.532)
Notre Dame L 23-25 10-3 (0.769) 119-74-1 (0.616)
Utah W 10-7 5-6 (0.455) 130-61 (0.681)
2001
Miami (OH) W 31-13 7-5 (0.583) 106-79 (0.573)
Washington L 18-23 8-4 (0.667) 95-94-1 (0.503)
Western Mich W 38-21 5-6 (0.455) 98-86-1 (0.532)
2000
Bowling Green W 42-7 2-9 (0.182) 100-83-2 (0.546)
Rice W 38-7 3-8 (0.273) 76-95-1 (0.445)
UCLA L 20-23 6-6 (0.500) 109-83 (0.568)
1999
Notre Dame W 26-22 5-7 (0.417) 119-43-1 (0.616)
Rice W 37-3 5-6 (0.455) 76-96-1 (0.445)
Syracuse W 18-13 7-5 (0.583) 97-92-1 (0.513)
1998
Notre Dame L 20-36 9-3 (0.750) 119-74-1 (0.616)
Syracuse L 28-38 8-4 (0.667) 97-92-1 (0.513)
Eastern Mich W 59-20 3-8 (0.273) 56-125 (0.309)
Hawaii W 48-17 0-12 (0.000) 100-101-1 (0.498)
1997
Colorado W 27-3 5-6 (0.455) 114-81-1 (0.584)
Baylor W 38-3 2-9 (0.182) 58-123 (0.320)
Notre Dame W 21-14 7-6 (0.539) 119-74-1 (0.616)
1996
Colorado W 20-13 10-2 (0.833) 114-81-1 (0.584)
Boston College W 20-14 5-7 (0.417) 120-75-1 (0.615)
UCLA W 38-9 5-6 (0.455) 109-83 (0.568)
1995
Virginia W 18-17 9-4 (0.692) 117-79 (0.597)
Memphis W 24-7 3-8 (0.273) 81-105 (0.436)
Boston College W 23-13 4-8 (0.333) 120-75-1 (0.615)
1994
Boston College W 34-26 7-4-1 (0.625) 120-75-1 (0.615)
Notre Dame W 26-24 6-5-1 (0.542) 119-74-1 (0.616)
Colorado L 26-27 11-1 (0.917) 114-81-1 (0.584)
1993
Washington St. W 41-14 5-6 (0.455) 93-95 (0.495)
Notre Dame L 23-27 11-1 (0.917) 119-74-1 (0.616)
Houston W 42-21 1-9-1 (0.136) 74-113-1 (0.396)
TOTALS
Cumulative opponent records in years we play them: 329-318-3 (0.509)
Cumulative opponent records for 1993 - 2008: 5117-4522-36 (0.531)
The difference in success percentage of only 0.022 here is about a quarter of a game per opponent per year. Whether or not this difference registers as "statistically significant" (I didn't check), common sense tells us that a quarter of a game per opponent is probably meaningless.
So the conclusion is, over the course of 16 years, our non-conference opponents have done, on average, no better or worse in years we've played them than they have over the course of the last 16 years as a whole. Sure, we've played a few teams who were unexpectedly playing lights-out (Utah undefeated last year; 10-2 and 11-1 Colorado teams in the mid-'90s), but we've also hit our share of otherwise decent teams having dismal years (0-12 Hawaii, 3-9 Toledo (granted they beat us, but that's not the point), and a couple of underperforming Notre Dame teams). In the end, it seems to work out about even.
EDIT: Added two games I previously missed.
After reading Hannibal's excellent review of Michigan's bad scheduling luck in the Big 10, I was curious to see what the results for our non-conference games would look like through the same lens. Thanks to James Howell and Chris Stassen, all the data required is readily available. In keeping with Hannibal's treatment, I have considered all non-conference games since the Big11Ten entered its current format in 1993. Michigan's game against each team has NOT been subtracted from that team's record, because (1) our game against each team was part of that team's overall level of success that year, and (2) I'm lazy. The only team whose overall win-loss record was likely to be impacted significantly over the course of 16 years just from playing us was Notre Dame; their 6-6 record against us over this span is slightly lower than their net 0.616 success percentage over the same span (success percentage defined as (wins+0.5*ties)/(wins+ties+losses). If I get ambitious, maybe I'll run the numbers again with games against us subtracted out, but I suspect it won't change the conclusions much.So without further ado, here are the numbers:
Team Game score Record 1993-2008 Cumulative
Utah L 23-25 13-0 (1.000) 130-61 (0.681)
Miami (OH) W 16-6 2-10 (0.167) 106-83-1 (0.561)
Notre Dame L 17-35 7-6 (0.538) 119-74-1 (0.616)
Toledo L 10-13 3-9 (0.250) 118-70-2 (0.626)
2007
The Horror L 32-34 13-2 (0.867) 148-56 (0.726)
Oregon L 7-39 9-4 (0.692) 130-65 (0.667)
Notre Dame W 38-0 3-9 (0.250) 119-74-1 (0.616)
Eastern Mich W 33-22 4-8 (0.333) 56-125 (0.309)
2006
Vanderbilt W 27-7 4-8 (0.333) 50-127 (0.283)
Central Mich W 41-17 10-4 (0.714) 83-103 (0.446)
Notre Dame W 47-21 10-3 (0.769) 119-74-1 (0.616)
Ball State W 34-26 5-7 (0.417) 84-100-2 (0.457)
2005
Northern Ill. W 33-17 7-5 (0.583) 81-104 (0.438)
Notre Dame L 10-17 9-3 (0.750) 119-74-1 (0.616)
Eastern Mich W 55-0 4-7 (0.364) 56-125 (0.309)
2004
Miami (OH) W 43-10 8-5 (0.615) 106-79 (0.573)
Notre Dame L 20-28 6-6 (0.500) 119-74-1 (0.616)
San Diego St. W 24-21 4-7 (0.364) 77-110 (0.412)
2003
Central Mich W 45-7 3-9 (0.250) 83-103 (0.446)
Houston W 50-3 7-6 (0.539) 74-113-1 (0.396)
Notre Dame W 38-0 5-7 (0.417) 119-74-1 (0.616)Oregon L 27-31 8-5 (0.615) 130-65 (0.667)
2002
Washington W 31-29 7-6 (0.539) 95-94-1 (0.503)
Western Mich W 35-12 4-8 (0.333) 98-86-1 (0.532)
Notre Dame L 23-25 10-3 (0.769) 119-74-1 (0.616)
Utah W 10-7 5-6 (0.455) 130-61 (0.681)
2001
Miami (OH) W 31-13 7-5 (0.583) 106-79 (0.573)
Washington L 18-23 8-4 (0.667) 95-94-1 (0.503)
Western Mich W 38-21 5-6 (0.455) 98-86-1 (0.532)
2000
Bowling Green W 42-7 2-9 (0.182) 100-83-2 (0.546)
Rice W 38-7 3-8 (0.273) 76-95-1 (0.445)
UCLA L 20-23 6-6 (0.500) 109-83 (0.568)
1999
Notre Dame W 26-22 5-7 (0.417) 119-43-1 (0.616)
Rice W 37-3 5-6 (0.455) 76-96-1 (0.445)
Syracuse W 18-13 7-5 (0.583) 97-92-1 (0.513)
1998
Notre Dame L 20-36 9-3 (0.750) 119-74-1 (0.616)
Syracuse L 28-38 8-4 (0.667) 97-92-1 (0.513)
Eastern Mich W 59-20 3-8 (0.273) 56-125 (0.309)
Hawaii W 48-17 0-12 (0.000) 100-101-1 (0.498)
1997
Colorado W 27-3 5-6 (0.455) 114-81-1 (0.584)
Baylor W 38-3 2-9 (0.182) 58-123 (0.320)
Notre Dame W 21-14 7-6 (0.539) 119-74-1 (0.616)
1996
Colorado W 20-13 10-2 (0.833) 114-81-1 (0.584)
Boston College W 20-14 5-7 (0.417) 120-75-1 (0.615)
UCLA W 38-9 5-6 (0.455) 109-83 (0.568)
1995
Virginia W 18-17 9-4 (0.692) 117-79 (0.597)
Memphis W 24-7 3-8 (0.273) 81-105 (0.436)
Boston College W 23-13 4-8 (0.333) 120-75-1 (0.615)
1994
Boston College W 34-26 7-4-1 (0.625) 120-75-1 (0.615)
Notre Dame W 26-24 6-5-1 (0.542) 119-74-1 (0.616)
Colorado L 26-27 11-1 (0.917) 114-81-1 (0.584)
1993
Washington St. W 41-14 5-6 (0.455) 93-95 (0.495)
Notre Dame L 23-27 11-1 (0.917) 119-74-1 (0.616)
Houston W 42-21 1-9-1 (0.136) 74-113-1 (0.396)
TOTALS
Cumulative opponent records in years we play them: 329-318-3 (0.509)
Cumulative opponent records for 1993 - 2008: 5117-4522-36 (0.531)
The difference in success percentage of only 0.022 here is about a quarter of a game per opponent per year. Whether or not this difference registers as "statistically significant" (I didn't check), common sense tells us that a quarter of a game per opponent is probably meaningless.
So the conclusion is, over the course of 16 years, our non-conference opponents have done, on average, no better or worse in years we've played them than they have over the course of the last 16 years as a whole. Sure, we've played a few teams who were unexpectedly playing lights-out (Utah undefeated last year; 10-2 and 11-1 Colorado teams in the mid-'90s), but we've also hit our share of otherwise decent teams having dismal years (0-12 Hawaii, 3-9 Toledo (granted they beat us, but that's not the point), and a couple of underperforming Notre Dame teams). In the end, it seems to work out about even.
EDIT: Added two games I previously missed.
August 16th, 2009 at 4:02 PM ^
html fail?
August 16th, 2009 at 5:10 PM ^
Indeed. Also, apparently the "Preview" button posts the diary entry. Sorry for the premature post; I'm still trying to get the formatting to work.
EDIT: I stripped out the HTML table and just spaced the entries with the Tab key. Bootleg, but it should at least be readable now.
August 16th, 2009 at 6:44 PM ^
Colorado wasn't unexpectedly good; they were a great program at that time. Bill McCartney (former Bo assistant) built them into a powerhouse, only to have his successor, Rick Neuheisel, allow it to collapse. (How Neuheisel keeps getting plum head coaching jobs is beyond me.)
Back to the record comparison, if you take away the 13-38 record those teams posted against us, their record in years they played us goes to 303-269-3 (.529) - or almost identical to their overall winning percentage for those years. So it appears that our scheduling pretty much held to form.
August 17th, 2009 at 2:50 AM ^
"Unexpectedly" is probably the wrong word - they had an excellent program throughout the '90s. The point was merely that even by their historical averages, the teams we played were especially good.
August 16th, 2009 at 7:37 PM ^
Michigan played Utah in 2002 and Oregon in 2003 as part of the new 12-game schedule that eventually became permanent in 2006.
What's most noticeable to me is the rise of the MAC and other fodder over the years. Those are very quality non-conference schedules in the mid to late 90's. By 2003, it's basically Notre Dame and non-BCS teams (generally the MAC).
August 17th, 2009 at 2:46 AM ^
Thanks for catching that -- not sure how I missed those games. They've been added. Including them changes the percentage for teams in years we play them from 0.507 to 0.509 - not much of a difference.
And I agree with your point about the change in scheduling. Can you even imagine having Boston College, Colorado, and Notre Dame on the same nonconference slate nowadays? And the change didn't begin immediately upon introduction of the BCS system, either - though I'm not sure how far in advance some of these schedules were made.
August 17th, 2009 at 12:38 AM ^
last year is even more depressing in this light: its the only year we didn't win at least two OOC games in the past 16 years...(ever?)
August 17th, 2009 at 4:51 AM ^
I'm not so sure this even compares to the Big Ten study. Typically, you want to play a solid to very good (even great, heaven forbid) non conference game, then a kind of middling one (hence, the Oregon games, at least in theory), and a creamy puff puff or two to close it out.
You kind of have to toss Notre Dame out of this mix because we will play them every year, except for a couple, and I'm sure they'll have up and down years. I can't think of anyone who plays a game like this, unless it involves ND, of course anymore. Texas and Oklahoma used to play each other every year, but now they're in the same conference.
The reason I'm ticked that UM plays Notre Dame so much is that you probably couldn't, or wouldn't, go out and schedule another Big Timer because you just don't know if the Irish are going to rise from the ashes, and having two heavyweights on the schedule as non con games is something NO ONE does. It's almost certain death to any MNC hopes, especially if you have a quirky year where both of those might be road games.
November 28th, 2015 at 3:54 PM ^
-
Michigan needs better players. The top ten states that produce the most NFL talent are CA, FL, GA, IL, NC, OH, PA, SC, TX, and VA. If Michigan played more games with teams from these states, Michigan would have an easier time recruiting kids from these states.
-
Michigan fans want to see interesting matchups. No more MAC games. No more directional Michigan games.
-
The athletic department wants to make money.
-
The college football playoff committee wants Michigan to play tough opponents.
- Michigan can check all the boxes by scheduling home-and-home games with these schools: Berkeley, Southern Cal, Stanford, UCLA, Florida, Florida State, Miami, Georgia, Georgia Tech, Missouri, North Carolina, Duke, North Carolina, NC State, South Carolina, Baylor, Houston, Rice, Texas A & M, Texas, SMU, Virginia, Virginia Tech.
Comments