Back in the summer, fans of both the Michigan State Spartans and the Michigan Wolverines would have Marveled at the situation that we find ourselves in this weekend. Back then, neither team was ranked and most pundits projected that the Spartans would not even get enough wins to play in a bowl game.
Time travel forward to late October and both teams are undefeated and ranked in the top-eight nationally. The last time both teams met as members of the top-10 was in the 1960s, and back then the annual Civil War took place in the third game of the season. One could argue that this will be the biggest matchup between the Green and the Blue...ever.
Yes, the contest this Saturday in East Lansing has all the trappings of a clash of superheroes. On one side is the hero who recently came out of a few-decades-long deep freeze of mediocrity and is currently enjoying a 21st century resurgence that started around 2008. This hero is generally considered to be better looking, stronger, and generally tries to do the right thing. I have always felt that MSU should really be the Captain of Team America.
On the other side is the arrogant son of rich parents who generally thinks that he can just show up and be awesome. He may have recruited a few other heroes to his side, but they are mostly just the new guy, some old work buddies and an impressionable teenager. Typical.
He always thought that he was the smartest guy in the room until he met Bruce Banner... and Blank Panther...and Shuri...and Dr. Strange...huh. He also seems to have a strange fascination with trying to revise history. (This may not be the most iron-clad analogy, but I think that you know what I mean.)
In any event, the entire college football world will be watching this weekend and only one team will emerge victorious. There is certainly a lot to discuss about this Michigan/Michigan State matchup below, but one things feels certain: it’s going to be epic.
Figure 1: Comparison of the projected final margins of victory based on my algorithm to the opening Vegas Spreads for Week Nine
Figure 2: Comparison of the projected final margins of victory based on ESPN's FPI to the opening Vegas Spreads in Week Nine
Table 1 below summarizes the combined upset picks for the week, based on the data shown above in the two Figures.
Table 1: Upset picks for Week Nine
With their super-powers combined, the computers are only predicting a total of 10 upsets this week, including SMU over Houston, Iowa to beat Wisconsin, Kentucky to upset Mississippi State, Ole Miss to defeat Auburn and the Rutgers Scarlet Knights (or is it Scarlet Witches?) to upset Illinois. The outcome of several of those games will have serious implications on a few divisional races.
For reference, my weekly simulation suggests that 13.3 upsets, plus-or-minus 2.9 upsets, are most likely to occur.
As for my recommended picks against the opening Vegas spread this week, those can be found below in Table 2.
Table 2: Picks against the opening spread for Week Nine. The picks are listed in order of confidence score
Only seven picks this week makes this the least amount of bad betting advice that I have given out since Week Two. For whatever reason, the computers seem to have settled down this week and are now much better behaved. They morphed from Ultron to Vision seemingly overnight.
But, my computer certainly does not know how this game is going to shake out any more than your neighbor with the Michigan flag on the porch does. Based on the opening Vegas spread of just two points, Michigan’s chance to earn a road victory against a ranked team is just 56 percent. Even if I use the current line of Michigan -4.5, the odds only creep up just past 60 percent. This game is a virtual toss-up.
But, there are a lot of historical and statistical trends that, once assembled, line up clearly in Michigan State’s favor this week.
First, there is Michigan’s abysmal record of late on the road against ranked teams. As recently as 2018, the Wolverines had lost 17 straight such games. That streak was broken in 2018 when Michigan beat No. 24 Michigan State (a team that finished 7-6). The Wolverines picked up a win last year at No. 21 Minnesota (a team that finished 3-4).
Michigan’s last win over a top-10 team on the road was over No. 2 Notre Dame in 2006. That was 25 Marvel films and 16 Marvel TV shows ago. Wow, that’s like “Thor: The Dark World” bad.
Second, contrary to popular opinion in Ann Arbor, the Wolverines very rarely beat “good” MSU teams. The best way to visualize this is in Figure 3.
Figure 3: Comparison of the regular season win percentage of Michigan and Michigan State back to 1950, not including the rivalry game. The winner of the Michigan / Michigan State game is colored either blue or green.
This chart compares the regular season win percentage of each team, NOT including the result of the Michigan/Michigan State game back to 1950. Each data point is shaded either blue or green based on the team that won the big game. Data points on the right side of the graph represent years when MSU was “good” and data points toward the top of the graph represent years when Michigan was “good.”
This figure tells us a lot about the history of the rivalry over the past 70 years. As you can see, the vast majority of the blue data points fall only in the upper left hand corner, in years where Michigan was pretty good and Michigan State was, at best, mediocre.
In the years where Michigan State wins more than 70 percent of its regular season games (excluding the one against the Wolverines), the Spartans are 17-2 (90 percent) with the only losses coming in 1955 and 1989. Note that Michigan’s record appears to be irrelevant. If MSU is good, MSU wins. If you are a Michigan fan, that’s like Edward Norton’s “Incredible Hulk” bad.
The opposite is clearly not true. When Michigan’s non-rivalry win percentage is over 70 percent on the season, the Wolverines are only 28-13 (68 percent) with the losses coming against Michigan State teams that were anywhere from elite (2013) to downright awful (1969). Also note that a sub-.500 Michigan team has not beaten Michigan State once in the last 70 years.
Some Wolverines fans have seen this data and pointed out that it that MSU has only been “good” just under 20 times in the past 70 years. Yes. That’s true. After winning multiple national titles in the 1950s and 1960s,tThe Spartans essentially wandered in the college football desert from the 1970s to around 2010 as a .500 program. This is not a surprise.
The Wolverines have been more consistent. But, Michigan is not consistently the “Leaders and Best,” as it is more generally just consistently above average. The Wolverines beat the teams that they are supposed to beat, win the Big Ten when it is weak (which has not been the case in 15-plus years), and get embarrassed in bowl games or other high profile matchups against “good” teams. That is where that program has been for decades. It is what it is.
In contrast, while Michigan State has struggled as a program at times, Coach Mark Dantonio brought a new attitude to East Lansing. The Spartans have dominated the rivalry recently, winning nine of the past 13 games and going 11-2 versus the spread against the Wolverines since 2008. Along the way, the Spartans have won the division, won the Big Ten, won the Rose Bowl and made the College Football Playoffs.
While results trailed off at the end of the Dantonio era, Coach Mel Tucker already seems to have things rolling in East Lansing once again. It is looking a lot more like a reload than a rebuild. If that is the case, and Michigan State continues to be strong, Figure 3 suggests that Michigan is going to continue to struggle to find wins against the Spartans going forward.
Naysayers will find it easy to dismiss such data as irrelevant, but I think that it speaks to some of the broader issues of psychology and team and university culture that play into this rivalry. Dismissing those intangibles because we cannot quantify them is foolish.
Michigan fans have argued for decades that Michigan State plays well against the Wolverines because the game is simply more important in East Lansing than in Ann Arbor. Maybe that’s true, but if it is, isn’t that on them? Perhaps the Spartans secret is that they are always angry. If so, the big green guys seem likely to smash the puny Wolverines yet again this year.
All that said, what about the specific matchup this coming weekend? In order to have a perfect record this late in the season, both team have to be pretty good. Both teams seem to have solid defenses and good running attacks. The Spartans appear to have the edge in the passing game and have certainly been more explosive, but also perhaps less consistent than Michigan. Special teams play seems solid for both teams.
While there are dozens of ways to try to parse and compare the data for each team, I will leave you with one additional figure. While stats like total yards are nice, I think that yards per attempt is the gold standard for judging offensive and defensive efficiency. Furthermore, it is difficult to truly compare two teams that have played different schedules. Fortunately, Michigan and Michigan State have three opponents in common so far: Northwestern, Nebraska and Rutgers.
To this end, I pulled the box score data for the six games involving the Spartans, the Wolverines, and those three opponents. I then calculated the rushing, passing and total yards per play on both sides of the ball for both teams. The results are shown below in Figure 4.
Figure 4: Yards per attempt comparison between MSU and U of M's offenses and defense against all common opponents so far in 2021.
Naturally, it is better to have a big number on offense (the left half of the chart) and a smaller number on defense (the right side of the chart). On the offensive side of the ball, the Spartans have the clear edge against common defenses in both the passing game, but also in the running game. Michigan did fare slightly better against the Cornhuskers, while MSU was significantly better against Rutgers’ and Northwestern’s defenses compared to the Maize and Blue.
On the defensive side of the ball, the numbers are closer, but still generally favor the Spartans. MSU’s rushing defense was better overall than Michigan’s in all three games. Through the air, Michigan’s defense grades out slightly better. In total, the Spartans gave up slightly less yards combined per play against those three common opponents.
If I add everything up, there are a lot of reasons for the Spartans to feel good about this game. That doesn’t mean that Michigan State will win, but I like a lot of what I see above. The Spartans are usually able to muster at least an “A-” level of effort when facing the Wolverines. If Michigan cannot match that intensity, it will be a long day for them. I am sure that they are saying “this year will be different...” just like every other year going back to the 1950s. Sure.
At the end of the day, both teams are going to try to impose their will on the other. Both teams will try to run the ball and if one team can dominate the other on the ground, that team will likely win. If that phase of the game cancels out, the Spartans’ superior passing game and explosiveness may carry the day. If that happens, the Wolverines may find themselves beaten and bloodied once again and feeling like they are lost in a hole somewhere in Siberia.
Table 3: Summary of Big Ten and national action in Week Nine, including my algorithms' projected scores. The teams shaded in green are projected to cover by my algorithm. The teams shaded red are projected to cover by the FPI.
In other Big Ten action, Penn State’s visit to Columbus against Ohio State is the obvious other big game, but everyone except my computer expects a blowout win by the Buckeyes. There are also several big games in the Big Ten West this week such as Iowa at Wisconsin (-3), Minnesota (-4) at Northwestern and Purdue at Nebraska (-7). The results of those three games will almost certainly make a big impact on the final Big Ten West standings.
As for the rest of the Power Five, it will be interesting to see if Oklahoma (-22) can continue to dodge bullets against Texas Tech. We will also see if Baylor is a real Big 12 contender this week as it hosts Texas (pick’em). UCLA’s trip to Utah (-3.5) will help to sort out the Pac-12 South. While the SEC race feels almost over, at least three games down south could potentially make waves, including the very Thanos-like Georgia (-14) versus Florida.
Finally, in Group of Five action, Cincinnati (-23) is likely to stay undefeated this week at Tulane. Undefeated San Diego State may not be so lucky versus Fresno State (pick’em). If chaos were to descend on the Group of Five, the winner of the Houston/SMU (+4) game might be the beneficiary.
That is all the advice that I have for you this week. Until next time, Go State, beat the Skunk Bears!
Time travel forward to late October and both teams are undefeated and ranked in the top-eight nationally. The last time both teams met as members of the top-10 was in the 1960s, and back then the annual Civil War took place in the third game of the season. One could argue that this will be the biggest matchup between the Green and the Blue...ever.
Yes, the contest this Saturday in East Lansing has all the trappings of a clash of superheroes. On one side is the hero who recently came out of a few-decades-long deep freeze of mediocrity and is currently enjoying a 21st century resurgence that started around 2008. This hero is generally considered to be better looking, stronger, and generally tries to do the right thing. I have always felt that MSU should really be the Captain of Team America.
On the other side is the arrogant son of rich parents who generally thinks that he can just show up and be awesome. He may have recruited a few other heroes to his side, but they are mostly just the new guy, some old work buddies and an impressionable teenager. Typical.
He always thought that he was the smartest guy in the room until he met Bruce Banner... and Blank Panther...and Shuri...and Dr. Strange...huh. He also seems to have a strange fascination with trying to revise history. (This may not be the most iron-clad analogy, but I think that you know what I mean.)
In any event, the entire college football world will be watching this weekend and only one team will emerge victorious. There is certainly a lot to discuss about this Michigan/Michigan State matchup below, but one things feels certain: it’s going to be epic.
Picks of the Week
Before we dig into the details of the big game, let’s take a look at this week’s potentially dubious betting advice. As is my usual tradition, Figures 1 and 2 below give this week’s projected point spreads for each of the 53 games involving two FBS teams. Figure 1 shows the picks from my algorithm relative to the opening Vegas lines and Figure 2 shows the picks from ESPN’s FPI.
Figure 1: Comparison of the projected final margins of victory based on my algorithm to the opening Vegas Spreads for Week Nine
Figure 2: Comparison of the projected final margins of victory based on ESPN's FPI to the opening Vegas Spreads in Week Nine
Table 1 below summarizes the combined upset picks for the week, based on the data shown above in the two Figures.
Table 1: Upset picks for Week Nine
With their super-powers combined, the computers are only predicting a total of 10 upsets this week, including SMU over Houston, Iowa to beat Wisconsin, Kentucky to upset Mississippi State, Ole Miss to defeat Auburn and the Rutgers Scarlet Knights (or is it Scarlet Witches?) to upset Illinois. The outcome of several of those games will have serious implications on a few divisional races.
For reference, my weekly simulation suggests that 13.3 upsets, plus-or-minus 2.9 upsets, are most likely to occur.
As for my recommended picks against the opening Vegas spread this week, those can be found below in Table 2.
Table 2: Picks against the opening spread for Week Nine. The picks are listed in order of confidence score
Only seven picks this week makes this the least amount of bad betting advice that I have given out since Week Two. For whatever reason, the computers seem to have settled down this week and are now much better behaved. They morphed from Ultron to Vision seemingly overnight.
The Big Game
Those who have followed the projections of my computer all season likely know what is coming next. My machine projects that the Wolverines are going to win this game and as such my "official prediction" is a final of Michigan 30, Michigan State 23.But, my computer certainly does not know how this game is going to shake out any more than your neighbor with the Michigan flag on the porch does. Based on the opening Vegas spread of just two points, Michigan’s chance to earn a road victory against a ranked team is just 56 percent. Even if I use the current line of Michigan -4.5, the odds only creep up just past 60 percent. This game is a virtual toss-up.
But, there are a lot of historical and statistical trends that, once assembled, line up clearly in Michigan State’s favor this week.
First, there is Michigan’s abysmal record of late on the road against ranked teams. As recently as 2018, the Wolverines had lost 17 straight such games. That streak was broken in 2018 when Michigan beat No. 24 Michigan State (a team that finished 7-6). The Wolverines picked up a win last year at No. 21 Minnesota (a team that finished 3-4).
Michigan’s last win over a top-10 team on the road was over No. 2 Notre Dame in 2006. That was 25 Marvel films and 16 Marvel TV shows ago. Wow, that’s like “Thor: The Dark World” bad.
Second, contrary to popular opinion in Ann Arbor, the Wolverines very rarely beat “good” MSU teams. The best way to visualize this is in Figure 3.
Figure 3: Comparison of the regular season win percentage of Michigan and Michigan State back to 1950, not including the rivalry game. The winner of the Michigan / Michigan State game is colored either blue or green.
This chart compares the regular season win percentage of each team, NOT including the result of the Michigan/Michigan State game back to 1950. Each data point is shaded either blue or green based on the team that won the big game. Data points on the right side of the graph represent years when MSU was “good” and data points toward the top of the graph represent years when Michigan was “good.”
This figure tells us a lot about the history of the rivalry over the past 70 years. As you can see, the vast majority of the blue data points fall only in the upper left hand corner, in years where Michigan was pretty good and Michigan State was, at best, mediocre.
In the years where Michigan State wins more than 70 percent of its regular season games (excluding the one against the Wolverines), the Spartans are 17-2 (90 percent) with the only losses coming in 1955 and 1989. Note that Michigan’s record appears to be irrelevant. If MSU is good, MSU wins. If you are a Michigan fan, that’s like Edward Norton’s “Incredible Hulk” bad.
The opposite is clearly not true. When Michigan’s non-rivalry win percentage is over 70 percent on the season, the Wolverines are only 28-13 (68 percent) with the losses coming against Michigan State teams that were anywhere from elite (2013) to downright awful (1969). Also note that a sub-.500 Michigan team has not beaten Michigan State once in the last 70 years.
Some Wolverines fans have seen this data and pointed out that it that MSU has only been “good” just under 20 times in the past 70 years. Yes. That’s true. After winning multiple national titles in the 1950s and 1960s,tThe Spartans essentially wandered in the college football desert from the 1970s to around 2010 as a .500 program. This is not a surprise.
The Wolverines have been more consistent. But, Michigan is not consistently the “Leaders and Best,” as it is more generally just consistently above average. The Wolverines beat the teams that they are supposed to beat, win the Big Ten when it is weak (which has not been the case in 15-plus years), and get embarrassed in bowl games or other high profile matchups against “good” teams. That is where that program has been for decades. It is what it is.
In contrast, while Michigan State has struggled as a program at times, Coach Mark Dantonio brought a new attitude to East Lansing. The Spartans have dominated the rivalry recently, winning nine of the past 13 games and going 11-2 versus the spread against the Wolverines since 2008. Along the way, the Spartans have won the division, won the Big Ten, won the Rose Bowl and made the College Football Playoffs.
While results trailed off at the end of the Dantonio era, Coach Mel Tucker already seems to have things rolling in East Lansing once again. It is looking a lot more like a reload than a rebuild. If that is the case, and Michigan State continues to be strong, Figure 3 suggests that Michigan is going to continue to struggle to find wins against the Spartans going forward.
Naysayers will find it easy to dismiss such data as irrelevant, but I think that it speaks to some of the broader issues of psychology and team and university culture that play into this rivalry. Dismissing those intangibles because we cannot quantify them is foolish.
Michigan fans have argued for decades that Michigan State plays well against the Wolverines because the game is simply more important in East Lansing than in Ann Arbor. Maybe that’s true, but if it is, isn’t that on them? Perhaps the Spartans secret is that they are always angry. If so, the big green guys seem likely to smash the puny Wolverines yet again this year.
All that said, what about the specific matchup this coming weekend? In order to have a perfect record this late in the season, both team have to be pretty good. Both teams seem to have solid defenses and good running attacks. The Spartans appear to have the edge in the passing game and have certainly been more explosive, but also perhaps less consistent than Michigan. Special teams play seems solid for both teams.
While there are dozens of ways to try to parse and compare the data for each team, I will leave you with one additional figure. While stats like total yards are nice, I think that yards per attempt is the gold standard for judging offensive and defensive efficiency. Furthermore, it is difficult to truly compare two teams that have played different schedules. Fortunately, Michigan and Michigan State have three opponents in common so far: Northwestern, Nebraska and Rutgers.
To this end, I pulled the box score data for the six games involving the Spartans, the Wolverines, and those three opponents. I then calculated the rushing, passing and total yards per play on both sides of the ball for both teams. The results are shown below in Figure 4.
Figure 4: Yards per attempt comparison between MSU and U of M's offenses and defense against all common opponents so far in 2021.
Naturally, it is better to have a big number on offense (the left half of the chart) and a smaller number on defense (the right side of the chart). On the offensive side of the ball, the Spartans have the clear edge against common defenses in both the passing game, but also in the running game. Michigan did fare slightly better against the Cornhuskers, while MSU was significantly better against Rutgers’ and Northwestern’s defenses compared to the Maize and Blue.
On the defensive side of the ball, the numbers are closer, but still generally favor the Spartans. MSU’s rushing defense was better overall than Michigan’s in all three games. Through the air, Michigan’s defense grades out slightly better. In total, the Spartans gave up slightly less yards combined per play against those three common opponents.
If I add everything up, there are a lot of reasons for the Spartans to feel good about this game. That doesn’t mean that Michigan State will win, but I like a lot of what I see above. The Spartans are usually able to muster at least an “A-” level of effort when facing the Wolverines. If Michigan cannot match that intensity, it will be a long day for them. I am sure that they are saying “this year will be different...” just like every other year going back to the 1950s. Sure.
At the end of the day, both teams are going to try to impose their will on the other. Both teams will try to run the ball and if one team can dominate the other on the ground, that team will likely win. If that phase of the game cancels out, the Spartans’ superior passing game and explosiveness may carry the day. If that happens, the Wolverines may find themselves beaten and bloodied once again and feeling like they are lost in a hole somewhere in Siberia.
Big Ten and National Overview
Let’s be honest. If you are reading this, there is probably only one game this weekend that you care about. But, for completeness, Table 3 gives an overview of the other games that might be worth following this weekend.Table 3: Summary of Big Ten and national action in Week Nine, including my algorithms' projected scores. The teams shaded in green are projected to cover by my algorithm. The teams shaded red are projected to cover by the FPI.
In other Big Ten action, Penn State’s visit to Columbus against Ohio State is the obvious other big game, but everyone except my computer expects a blowout win by the Buckeyes. There are also several big games in the Big Ten West this week such as Iowa at Wisconsin (-3), Minnesota (-4) at Northwestern and Purdue at Nebraska (-7). The results of those three games will almost certainly make a big impact on the final Big Ten West standings.
As for the rest of the Power Five, it will be interesting to see if Oklahoma (-22) can continue to dodge bullets against Texas Tech. We will also see if Baylor is a real Big 12 contender this week as it hosts Texas (pick’em). UCLA’s trip to Utah (-3.5) will help to sort out the Pac-12 South. While the SEC race feels almost over, at least three games down south could potentially make waves, including the very Thanos-like Georgia (-14) versus Florida.
Finally, in Group of Five action, Cincinnati (-23) is likely to stay undefeated this week at Tulane. Undefeated San Diego State may not be so lucky versus Fresno State (pick’em). If chaos were to descend on the Group of Five, the winner of the Houston/SMU (+4) game might be the beneficiary.
That is all the advice that I have for you this week. Until next time, Go State, beat the Skunk Bears!
