2020 Yards Per Play: Big 12

Three conferences down. Seven to go. This week, we stay in America's heartland and look at the Big 12.   

Here are the 2020 Big 12 standings.

So we know what each team achieved, but how did they perform? To answer that, here are the Yards Per Play (YPP), Yards Per Play Allowed (YPA) and Net Yards Per Play (Net) numbers for each Big 12 team. This includes conference play only, with the championship game not included. The teams are sorted by Net YPP with conference rank in parentheses.

College football teams play either eight or nine conference games (typically fewer in 2020). Consequently, their record in such a small sample may not be indicative of their quality of play. A few fortuitous bounces here or there can be the difference between another ho-hum campaign or a special season. Randomness and other factors outside of our perception play a role in determining the standings. It would be fantastic if college football teams played 100 or even 1000 games. Then we could have a better idea about which teams were really the best. Alas, players would miss too much class time, their bodies would be battered beyond recognition, and I would never leave the couch. As it is, we have to make do with the handful of games teams do play. In those games, we can learn a lot from a team’s YPP. Since 2005, I have collected YPP data for every conference. I use conference games only because teams play such divergent non-conference schedules and the teams within a conference tend to be of similar quality. By running a regression analysis between a team’s Net YPP (the difference between their Yards Per Play and Yards Per Play Allowed) and their conference winning percentage, we can see if Net YPP is a decent predictor of a team’s record. Spoiler alert. It is. For the statistically inclined, the correlation coefficient between a team’s Net YPP in conference play and their conference record is around .66. Since Net YPP is a solid predictor of a team’s conference record, we can use it to identify which teams had a significant disparity between their conference record as predicted by Net YPP and their actual conference record. I used a difference of .200 between predicted and actual winning percentage as the threshold for ‘significant’. Why .200? It is a little arbitrary, but .200 corresponds to a difference of 1.6 games over an eight game conference schedule and 1.8 games over a nine game one. Over or under-performing by more than a game and a half in a small sample seems significant to me. In the 2020 season, which teams in the Big 12 met this threshold? Here are Big 12 teams sorted by performance over what would be expected from their Net YPP numbers.

Texas Tech was the lone Big 12 team to see their actual record differ significantly from their expected record based on YPP. The Red Raiders under-performed, but we can't blame close games. All three of their conference victories came by a touchdown or less and they were a respectable 3-2 in close conference games. Instead, the culprit was turnovers. The Red Raiders forced the fewest turnovers in Big 12 play (9) while committing the most (18), causing them to fall short of their expected record. Which, speaking of...

Matt Wells' Teams Almost Always Under-Perform

At least relative to their YPP numbers. Matt Wells has been a coach at the FBS level for eight seasons; six at Utah State and two at Texas Tech. I happen to have YPP data for all those seasons (and a few more going back to 2005) and in that span, his teams almost always fall short (sometimes a little, other times a lot) of their expected conference record based on YPP. 

During his six seasons at Utah State, the Aggies were undoubtedly a successful team. They played in five bowl games, won a division title, finished in the top 25, and went 30-18 against Mountain West opponents. However, in that six season span, Utah State finished with the largest average negative disparity between their conference record and their expected record based on YPP in the Mountain West.

As you might expect over a sample size of six seasons, most teams finished with a negligible average disparity. Teams over-perform one year and under-perform the next with little rhyme or reason save for the randomness inherent in a college football season. However, two teams really stood out. Air Force and Utah State. Air Force finished with a conference winning percentage that was on average .141 better than their expected winning percentage based on YPP. In an eight game conference season, this means they won about 1.13 more additional games than we might expect based on their YPP numbers. By contrast, Utah State finished with a conference winning percentage that was .109 lower than their expected winning percentage. This equates to 0.87 fewer wins in an eight games conference season. 

After guiding Utah State to eleven wins and a top 25 appearance in 2018, Wells was hired by Texas Tech. In his first two seasons in Lubbock, the Red Raiders have struggled. They are just 8-14 overall and 5-13 in Big 12 play. They lost to Kansas in 2019 and nearly lost to them again in 2020. Despite the poor record, Texas Tech has put up mediocre per play numbers. Last season, they ranked ninth in the Big 12 in Net YPP, and while finishing ahead of Kansas should be the default for any Big 12 team, their record was still slightly worse than one would expect based on their YPP numbers. Then in 2020, the Red Raiders actually posted a positive per play differential, yet lost twice as many conference games as they won. Through two seasons, Texas Tech has under-performed worse than any other Big 12 team.

That -.146 equates to 1.31 fewer wins than expected over a nine-game Big 12 conference season. Two seasons is not nearly a large enough sample to draw any meaningful conclusions, but paired with his six seasons at Utah State, a pattern has started to develop.

In eight seasons, Wells' teams have been in the top half of their conference once when it comes to performance against expected record based on YPP. Meanwhile, they have finished last three times, second to last once, and third to last once. Why have his teams consistently struggled to finish within range of their lofty YPP numbers? Well, we can't blame turnover margin. His teams are collectively +10 in 66 conference games. No, the primary culprit appears to be plain old close game luck. In conference games decided by a touchdown or less, his Utah State teams were 5-9. In conference games decided by more than a touchdown, those Aggies were 25-9. Meanwhile, in two seasons at Texas Tech, his teams are 3-6 in close conference games and 2-7 in all other conference games. Doing the math, overall, his teams are just 8-15 in close conference games (and 27-16 in all other conference games). Is Matt Wells a poor close game coach or is this record a function of small sample size randomness? History is not destiny, but if Matt Wells wants to keep his job, he better start winning more of these close games.

2020 Adjusted Pythagorean Record: Big 10

Last week we looked at how Big 10 teams fared in terms of yards per play. This week, we turn our attention to how the season played out in terms of the Adjusted Pythagorean Record, or APR. For an in-depth look at APR, click here. If you didn’t feel like clicking, here is the Reader’s Digest version. APR looks at how well a team scores and prevents touchdowns. Non-offensive touchdowns, field goals, extra points, and safeties are excluded. The ratio of offensive touchdowns to touchdowns allowed is converted into a winning percentage. Pretty simple actually.

Once again, here are the 2020 Big 10 standings.

And here are the APR standings with conference rank in offensive touchdowns, touchdowns allowed, and APR in parentheses. This includes conference games only with the championship game excluded. Since teams played a varied number of games (everyone played at least five and a few played a full nine game schedule), the rankings are on a per game basis, not raw totals. 

Finally, Big 10 teams are sorted by the difference between their actual number of wins and their expected number of wins according to APR.

I use a game and a half as a line of demarcation to determine whether or not a team significantly over or under-performed relative to their APR. Using that standard, Wisconsin was the lone team that saw their record differ significantly from their APR. The Badgers played six conference games and won half of them, but two of their victories (the first two) came by 38 points apiece and gave the impression the Badgers were an elite team. Huge margins of victory, particularly in a smaller than normal conference season, can really skew how a team rates in APR. 

Bert is Back in the Big 10

You may have missed it if you don't follow Illinois football (and why would you?), but the Illini will have a new football coach in 2021. Illinois replaced a one-time Super Bowl losing coach with a two-time Rose Bowl losing coach when they hired Bret Bielema. Bert as he is affectionately known, coached Wisconsin for seven seasons following Barry Alvarez's retirement. His tenure was very successful with three Big 10 titles, five ranked finishes, and an average of nearly ten wins per season. His Big 10 background and success makes him an obvious homerun hire for Illinois right? Well, if you look at the past three Wisconsin coaches, it sure seems like the culture surrounding the program was the primary reason for the success other than any greatness inherent in the head coach. 

We'll circle back to Bielema in a moment, but let's start with his replacement, Gary Andersen. Andersen was the head coach at Wisconsin for just two season, but in that time he led the Badgers to nineteen victories and one division championship. Following his abrupt departure, he headed to Corvallis, Oregon and coached the Oregon State Beavers for two and a half seasons, where his teams won just seven games. He resigned from Oregon State and after a little more than a year, returned to where his FBS head coaching career began, Utah State. Andersen took over a Utah State team that had finished in the top twenty-five the previous season, with a future first round pick at quarterback, and led them to a 7-6 record. In his second season back at Utah State, his team began the year 0-3 with each loss coming by at least three touchdowns and Andersen again resigned. At Wisconsin, his teams went 13-3 in Big 10 play. Since leaving Wisconsin, his teams won nine conference games over parts of five seasons.

This analysis is not completely fair to Andersen as he did lead Utah State to a pair of bowl games in his original tenure. Let's give him credit for those four seasons.

When we include his first run in Logan, his career numbers look a little better, but its important to remember context. When Andersen had his greatest success at Utah State, the WAC was imploding. Boise State left the WAC for the Mountain West after the 2010 season, with Fresno State, Hawaii, and Nevada following after the 2011. In 2012, Utah State was in a conference with two teams that had good seasons (Louisiana Tech and San Jose State), two of the worst teams in FBS (Idaho and New Mexico State), and two FBS newcomers (Texas-San Antonio and Texas State). 

When Andersen left for Corvallis, Wisconsin hired Pittsburgh head coach Paul Chryst. Chryst's teams in the Steel City had been middling, posting a 19-19 overall record as they closed out the Big East and joined the ACC. However, upon arriving in Madison, the Badgers continued to be one of the best teams in the conference outside of Ohio State. In six seasons under Chryst, Wisconsin has won at least ten games four times, played in the Big 10 Championship Game three times, and averaged nearly ten wins per year. In conference play, the Badgers have lost as many games in six seasons under Chryst as the Panthers did in three seasons. 

Now lets return to Bielema. He has not been out of coaching since 2012. After leaving Wisconsin, he became head coach of Arkansas. While his overall record at Arkansas was mediocre (29-34) and while he guided the Hogs to three bowl games in five seasons, he did not do what he was ostensibly paid for; beat fellow SEC opponents. 

Bielema's best squad (at least based on the advanced metrics) was probably his 2014 team which managed a 2-6 conference record. His team did finish with a winning conference record in 2015 (and famously denied Ole Miss their shot at an SEC crown), but that same team also lost to Texas Tech and Toledo in non-conference play. Outside of that 5-3 finish, his teams averaged 1.5 conference wins per season. Bielema's new division, the Big 10 West, is nowhere near as brutal as the SEC West and Bielema won't have to deal with Alabama, Auburn, LSU, and Texas A&M on an annual basis, but Illinois will begin his tenure at the bottom of the division hierarchy. Iowa and Wisconsin are the stalwart front runners, Northwestern is the nerd school that annually overachieves, and Minnesota, Nebraska, and Purdue have all hired young promising coaches in the past few years (though success at their current locations has been mixed). Illinois replaced a failed retread (who took them to a bowl a little more than a year ago I might add) with another failed retread who happened to have Big 10 roots. And his success at his previous Big 10 job (now nearly a decade in the past) appears to have had more to do with the program itself than anything Bielema did. Perhaps Bielema will be a success at Illinois, I've been wrong a time or two before. However, this hire was lazy. Instead of hiring a coach that succeeded at an easy job and failed at a hard one, maybe take a shot on a coach that just led Kent State to back to back winning seasons? Casual college football fans and less plugged in alums may not have known who he was, but he would have been a better choice.    

2020 Yards Per Play: Big 10

This week we move from the coastal elites to the midwest and examine the Big 10.  

Here are the 2020 Big 10 standings.

So we know what each team achieved, but how did they perform? To answer that, here are the Yards Per Play (YPP), Yards Per Play Allowed (YPA) and Net Yards Per Play (Net) numbers for each Big 10 team. This includes conference play only, with the championship game not included (but the other games from championship weekend are). The teams are sorted by division by Net YPP with conference rank in parentheses.

College football teams play either eight or nine conference games (typically fewer in 2020). Consequently, their record in such a small sample may not be indicative of their quality of play. A few fortuitous bounces here or there can be the difference between another ho-hum campaign or a special season. Randomness and other factors outside of our perception play a role in determining the standings. It would be fantastic if college football teams played 100 or even 1000 games. Then we could have a better idea about which teams were really the best. Alas, players would miss too much class time, their bodies would be battered beyond recognition, and I would never leave the couch. As it is, we have to make do with the handful of games teams do play. In those games, we can learn a lot from a team’s YPP. Since 2005, I have collected YPP data for every conference. I use conference games only because teams play such divergent non-conference schedules and the teams within a conference tend to be of similar quality. By running a regression analysis between a team’s Net YPP (the difference between their Yards Per Play and Yards Per Play Allowed) and their conference winning percentage, we can see if Net YPP is a decent predictor of a team’s record. Spoiler alert. It is. For the statistically inclined, the correlation coefficient between a team’s Net YPP in conference play and their conference record is around .66. Since Net YPP is a solid predictor of a team’s conference record, we can use it to identify which teams had a significant disparity between their conference record as predicted by Net YPP and their actual conference record. I used a difference of .200 between predicted and actual winning percentage as the threshold for ‘significant’. Why .200? It is a little arbitrary, but .200 corresponds to a difference of 1.6 games over an eight game conference schedule and 1.8 games over a nine game one. Over or under-performing by more than a game and a half in a small sample seems significant to me. In the 2020 season, which teams in the Big 10 met this threshold? Here are Big 10 teams sorted by performance over what would be expected from their Net YPP numbers.

Indiana and Northwestern overperformed relative to their YPP numbers. The Hoosiers and Wildcats nearly met in a Bizarro Big 10 Championship Game until the conference changed the rules at the eleventh hour to allow Ohio State to qualify despite playing just five conference games. The Hoosiers and Wildcats were a combined 5-1 in close conference games, but another major reason for their success relative to their performance was their absurd defensive play once their opponents arrived in the Red Zone. In Big 10 play, Indiana allowed nineteen drives inside their twenty yard line, but opponents converted touchdowns less than half the time (nine) and scored at all just over half the time (eleven). Meanwhile, Northwestern was even better at preventing touchdowns. Opponents scored just seven touchdowns in the twenty-five occasions they got inside the twenty. They did manage ten field goals to account for seventeen total scores, but were only able to punch it in the end zone just over a quarter of the time. The Hoosiers and Wildcats were the epitome of 'bend but don't break'. At the other end of the spectrum, Maryland, Michigan, and Purdue all underachieved relative to their YPP numbers. Maryland played only five games, so the small sample size of their conference schedule led to increased variance. The Terrapins were also -7 in turnover margin in those five games. I couldn't pinpoint a reason for Michigan's underachievement, but the Wolverines also played a small number of conference games (6) and had a negative turnover margin (-3). Purdue finished 2-3 in close conference games, but the real culprit for their underperformance were their own Red Zone struggles. The Boilermakers scored just fifteen touchdowns on twenty-eight possessions inside the twenty. 

Longest Bowl Losing Streaks

In many respects, 2020 was a dream season for Indiana. In an abbreviated campaign, the Hoosiers finished with one regular season loss and found themselves ranked seventh in the AP Poll when bowl season commenced. Alas, the Hoosiers met up with a volatile Ole Miss team in the Outback Bowl and lost 26-20. Despite the loss, the Hoosiers still finished twelfth in the AP Poll, their highest ranking since finishing fourth in 1967. The loss was familiar for Indiana fans in that it was their sixth consecutive bowl loss. The Hoosiers have not won a bowl game since the 1991 Copper Bowl. Six consecutive bowl losses may seem like a lot, but the Hoosiers have a ways to go to match the all-time record among current Power Five teams. And who might they be chasing for this ignominious record? Why, a team they almost faced in the Big 10 Championship Game and a small, private school with little football tradition, also located in Indiana. 

I kid of course about Notre Dame. For a private school, their enrollment is rather large. Honorable mentions for bowl losing streaks go to South Carolina and West Virginia, a pair of schools that followed starkly different timelines in those losing streaks. South Carolina's took more than four decades, while West Virginia squeezed theirs into a twelve year period. 

Take heart Indiana fans. While the bowl losing streak continues, at least you can always relish the journey of the 2020 season. 

2020 Adjusted Pythagorean Record: ACC

Last week we looked at how ACC teams fared in terms of yards per play. This week, we turn our attention to how the season played out in terms of the Adjusted Pythagorean Record, or APR. For an in-depth look at APR, click here. If you didn’t feel like clicking, here is the Reader’s Digest version. APR looks at how well a team scores and prevents touchdowns. Non-offensive touchdowns, field goals, extra points, and safeties are excluded. The ratio of offensive touchdowns to touchdowns allowed is converted into a winning percentage. Pretty simple actually.

Once again, here are the 2020 ACC standings.

And here are the APR standings with conference rank in offensive touchdowns, touchdowns allowed, and APR in parentheses. This includes conference games only with the championship game excluded. Since teams played a varied number of games (everyone played at least seven and a few played a full ten game schedule), the rankings are on a per game basis, not raw totals. 

Finally, ACC teams are sorted by the difference between their actual number of wins and their expected number of wins according to APR.

I use a game and a half as a line of demarcation to determine whether or not a team significantly over or under-performed relative to their APR. Using that standard, Louisville was the lone team that saw their record differ significantly from their APR. Astute readers will remember Louisville also under-performed relative to their expected record based on yards per play and we went over some reasons for that last week. No need to rehash that again. 

How to Spot a Championship Game Blowout

I had high hopes for the ACC Championship Game in 2020. It was a rematch of a regular season classic, and despite the fact the locale would be different and Clemson would have access to their starting quarterback, I felt it would be competitive. I could not have been more wrong. Clemson jumped out to a three touchdown lead by halftime and toyed with Notre Dame in the second half in a 24 point victory. While I was wrong about this game, I have developed a nearly fool proof way to determine if the title game you are about to watch is going to be a blowout. Read on to find out how.

I've written before about how shockingly accurate the preseason AP Poll is as a proxy for talent and team strength. This is not an original thought. Other internet folks discovered this valuable insight before me and if I could remember the exact person I stole this idea from, I would cite them here. But I digress. The conference championship game is a relatively recent college football innovation, with the SEC pioneering the event in 1992. Since that means I didn't have to comb through a hundred years or more worth of data, I decided to look at the preseason AP rank of every Power Five conference championship game participant and see how often teams that were ranked in the preseason AP Poll beat teams that were not. Since the SEC started this title game phenomenon we'll look at them first.

We have had seven matchups in SEC Championship Game history where one team was ranked in the preseason AP Poll and the other was not. The team ranked in the preseason poll won all seven games with each victory coming by at least ten points and the average margin of victory coming by more than three touchdowns. Looking at preseason top ten teams, they are 6-0 against teams that were not ranked in the preseason. Their average margin of victory actually dips a little to 19 points per game, but that is all thanks to Cam Newton.

The Big 12 was the second Power Five conference to institute a championship game. While they took a brief hiatus after some of their members departed following the 2010 season, they are back in the championship game business.

Teams ranked in the preseason are 6-1 against preseason unranked teams, winning by an average of nearly four touchdowns per game with five of the seven games being decided by double digits. Colorado was the only team unranked in the preseason poll to beat a preseason ranked team in the title game when they took down Texas in 2001.

The ACC was the third Power Five conference to institute a conference championship game. So we'll look at them next. 

Despite having significantly fewer title games than the SEC, the ACC has more instances of preseason ranked teams taking on preseason unranked teams in said title game. The teams ranked in the preseason are 10-1 overall with the average margin of victory coming by nearly two touchdowns. If we look at preseason top ten teams, they are 6-0 with an average margin of victory of more than 17 points. Interestingly, the lone preseason unranked team to win was Clemson back in 2011. That marked the last time the Tigers were not ranked in the preseason poll.

After the ACC, the Big 10 and Pac-12 expanded and started playing championship games in 2011. We'll go alphabetically and start with the Big 10. 

The Big 10 has the fewest instances of preseason ranked teams taking on preseason unranked teams in the title game. That small sample size means we should probably take the relatively small average margin of victory with a grain of salt. Before we move to the Pac-12, I would like to pay homage to the 2013 Michigan State Spartans. After a disappointing 7-6 finish in 2012, the Spartans were not ranked in the preseason poll in 2013. Riding one of the best modern college football defenses, the Spartans won each of their conference games by at least ten points (including the Big 10 title game against Ohio State). Were it not for a tight loss to Notre Dame in the non-conference, the Spartans may well have faced Florida State in the final BCS Championship Game. 

In the Pac-12, preseason ranked teams are unbeaten against preseason unranked teams, winning by more than twenty points per game on average. The numbers are similar when we look at those teams that were ranked in the top-ten.

So that is how all the Power Five conferences shook out individually. What about when we aggregate the data? First with all preseason ranked teams.

32-3 is a pretty strong record, especially in college football. Those teams won by an average of more than eighteen points per game and nearly two thirds of the games were decided by double digits. Now what about those teams ranked in the preseason top ten that drew conference championship games against teams unranked in the preseason?

Their record is slightly better and their scoring margin is also slightly better. In addition, more than 70% of the matchups resulted in a double digit win for the preseason top-ten team. 

Finally, before we leave, here is a list of all the Power Five conference championship game winners that were unranked in the preseason. We already touched on the three teams that pulled upsets in Power Five title games (Clemson, Colorado, and Michigan State), but three other teams managed to win their conferences despite not being ranked in the preseason AP Poll. How did they do that you ask? By facing other teams that were also unranked in the preseason poll. 

Wake Forest, Auburn, and Penn State were all surprise conference championship game participants that benefited from facing other surprise teams once they got their. The preseason unranked versus preseason unranked in a Power Five title game is truly a rare event, but if you come across a title game where one of the participants was ranked in the preseason poll and the other was not, it might be a good idea to change the channel.