Thursday, January 30, 2020

2019 Yards Per Play: ACC

The second conference we come to in our alphabetized recap of 2019 is the ACC. Here are the ACC 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 ACC team. This includes conference play only, with the championship game not included. The teams are sorted by division by Net YPP with conference rank in parentheses.
College football teams play either eight or nine conference games. 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 2019 season, which teams in the ACC met this threshold? Here are ACC teams sorted by performance over what would be expected from their Net YPP numbers.
Virginia was the lone ACC team that saw their expected record based on YPP differ significantly from their actual record. The Cavaliers exceeded their expected conference won/loss record and in the process won the Coastal Division for the first time. Their division title means every team in the Coastal has won the division at least once. In 2020, Virginia will attempt to become the first repeat division winner since Virginia Tech (2010 and 2011). Virginia was not especially fortunate in close games (3-2 in one-score conference games), but thanks to the malaise of the non-Clemson portion of the conference, they were able to string together enough wins to become the Tiger’s sacrificial lamb in the ACC Championship Game.

One Ranked Team 
The ACC was not a strong conference in 2019. I doubt even the most ardent Packer and Durham fan would argue otherwise. There was one elite team, Clemson, and a slew of decent to bad teams. Befitting a conference with one powerhouse and a lot of lightweights, Clemson was the lone ACC team that finished ranked in the final edition of the AP Poll. For an ostensible power conference, this is a rare ‘accomplishment’. To find out how rare, I looked up the last time each power conference finished with one ranked team. I included the five current power conferences (ACC, Big 10, Big 12, Pac-12, and SEC), a former power conference in the BCS era (Big East), and the two conferences that practically merged to form the Big 12 (Big 8 and Southwest Conference). Here is the last time each conference had only one team ranked in the final AP Poll with commentary to follow.
ACC: 1988
I should note the AP Poll was only twenty teams strong in 1988 (instead of 25). 1988 was a lot like 2019. Clemson was clearly the best team in the conference, although this time they did lose a close game to a team from the Tar Heel state. And just like 2019, Virginia finished second. 1988 was interesting in that no other ACC team entered the polls at any point during the season. At least 2019 saw Syracuse, Virginia, Virginia Tech, and Wake Forest with a ranking beside their name at some point.

Big East: 2012
In its final season, the Big East saw Louisville finish as the lone ranked team. The Cardinals pulled off a massive Sugar Bowl upset of Florida to finish ranked thirteenth. Louisville entered that game ranked 22nd, so had the game gone to form, it’s quite possible the Big East would have finished its final season with zero ranked teams (something that also happened in 2010).

Big Eight: 1983
The year Tom Osborne nearly won his first national title was also the only time the Oklahoma Sooners under Barry Switzer did not finish the season ranked in the final poll. The Sooners opened the season ranked second in the country, but four losses later, they were out of the polls. Once again, the AP Poll went just twenty deep in 1983.

Big Ten: 1982
On January 1st, 1983, Michigan lost the Rose Bowl to UCLA dropping them out of the final AP Poll, and leaving Ohio State as the lone ranked team from the Big 10.

Big 12: Never 
In every year of its existence, the Big 12 has had at least two teams finish ranked. In fact, the conference has had at least three teams finish ranked every year except 2006.

Pac-12: 1999
Oregon was probably the best team in the then Pac-10 in 1999, but they lost twice in conference play and did not get the opportunity to play eventual champion Stanford. While the Ducks were the lone team to finish ranked, seven of the league’s ten members entered the AP Poll at some point in 1999 (Cal, Oregon State, and Washington State were the only teams that did not).

SEC: 1943
Yes, if you go back far enough, even the mighty SEC has a down year now and then. For some reason, only four SEC schools fielded football teams in 1943 (and two of them are not even in the league any more). Do they even care about football in the south? I did some follow-up research and it appears there was a minor geopolitical conflict in 1943.

Southwest Conference: 1993
As the conference was circling the drain, Texas A&M won their third straight SWC title and for the third year in a row, finished as the lone ranked team.

Thanks for reading. We'll be back next week with an APR look at the ACC. Spoiler, Clemson does pretty well in that metric.

Thursday, January 23, 2020

2019 Adjusted Pythagorean Record: AAC

Last week we looked at how AAC 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 2019 AAC 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.
Finally, AAC 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. Temple was the only team that saw their expected record differ significantly from their APR. The Owls finished with a winning conference record despite allowing more touchdowns than they scored. The Owls were not especially fortunate in close games, finishing 2-1 in one-score AAC contests. No, the disconnect between their record and APR can be explained by a pair of blowouts. The Owls lost to SMU and UCF in back-to-back weeks by a combined 66 points. In their five conference wins, the Owls only outscored their opponents by 62 total points (thirty total points in the four non-Connecticut games). Those uncharacteristic spankings tempered their overall APR numbers.

I'm Your Stepping Stone
In the Playoff era (since 2014), the AAC has been the preeminent Group of Five conference. The league has had at least one member institution finish ranked in the final AP Poll each season, has grabbed four of the six available NY6 slots (no other conference has more than one), and sent more coaches to Power Five jobs than nearly the rest of the Group of Five combined. The following table lists the AAC coaches that have leveled up each offseason.
After not having any coaches leave for Power Five jobs in the first offseason of the Playoff era, the American has sent at least one coach to a Power Five job each of the past five offseasons. With eight coaches moving to the big time, the AAC is well ahead of the other Group of Five leagues. Here is how the other four G5 conferences have fared in terms of sending their coaches to Power Five jobs.
The other four conferences have all sent coaches to the Power Five at about the same rate. The MAC, Mountain West, and Sun Belt have all seen three coaches get Power Five jobs while Conference USA has had two such coaches.

While the AAC has sent the most coaches to Power Five jobs, it’s fair to question their success (or lack thereof) once they get those bigger jobs. Hiring coaches is often a crapshoot, but some of the more notable AAC graduations have already been fired. Chad Morris did not finish his second season at Arkansas, Willie Taggart lasted less than two seasons at Florida State (although that was his second job after leaving South Florida), and Scott Frost has yet to lead Nebraska to the postseason. For comparison, I have included the aggregate winning percentage at the Power Five level of the coaches from each Group of Five conference. Note that Taggart’s time at Oregon and Florida State are included here.
Thanks for reading. Check back next week when we look at the yards per play numbers for the ACC.

Thursday, January 16, 2020

2019 Yards Per Play: AAC

We begin our offseason recaps as we have in the past with the American Athletic Conference. Here are the AAC 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 AAC team. This includes conference play only, with the championship game not included. The teams are sorted by division by Net YPP with conference rank in parentheses.
College football teams play either eight or nine conference games. 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 2019 season, which teams in the AAC met this threshold? Here are AAC teams sorted by performance over what would be expected from their Net YPP numbers.
The AAC saw a pair of teams significantly exceed their expected record based on YPP and saw one team significantly under perform. Cincinnati finished ranked in the AP top 25 for the second consecutive season and while they managed to beat a few good teams (most notably UCF) and hang tough with Memphis in back-to-back weeks, they also narrowly edged East Carolina and South Florida and needed a blocked extra point returned for two to hold off Temple. Overall, the Bearcats finished 4-0 in one-score conference games during the regular season and finished with the best in-conference turnover margin of any AAC team (+7). SMU was also ranked in the AP top 25 for much of the season, but crapped the bed in their bowl game to finish unranked. The Mustangs were not especially fortunate in close games like the Bearcats, finishing 3-2 in one-score conference games nor did they boast an amazing turnover margin (+3 in AAC play). The slight advantages in those two areas helped them overcome their defensive struggles (allowed 6.65 yards play over their final five conference games) and mediocre statistical profile. Tulsa was on the other end of the spectrum. Despite a mediocre yards per play margin, they won just twice in conference play. They were not especially unlucky in close games (1-2 in one-score AAC games), nor was their turnover margin especially poor (-1 in AAC play). The culprit was non-offensive touchdowns. The Golden Hurricane allowed four non-offensive touchdowns while scoring just one of their own. Two of those touchdowns (an interception and a kickoff return) came against Houston in a game Tulsa lost by ten points. Another (a fumble return) came against Tulane in a game the Golden Hurricane lost by twelve. Those mostly random non-offensive scores made the Golden Hurricane look much worse than they actually were.

What a Con
2019 marks the end of an illustrious run for Connecticut in the AAC. The Huskies have grown tired of conference living and will be setting out as an independent beginning with the 2020 campaign. While the Huskies have served as a functional bye for their conference opponents over the past two seasons, they were not always quite this bad. In the early days of the AAC (2015), the Huskies actually qualified for a bowl game (and beat the eventual Group of Five NY6 representative). The years since that legendary St. Petersburg Bowl have not been kind. The Huskies moved on from Bob Diaco after a three win season in 2016 and attempted to recapture their (relative) glory days of the late aughts by bringing back Randy Edsall. Edsall won just two conference games in three seasons of work in the AAC and leaves with a nineteen game conference losing streak. However, perhaps more impressive than the losing streak is the streak of games in which Connecticut was expected to lose. The Huskies have not been favored against another AAC opponent since their 2017 conference opener against East Carolina (which they lost). How does that streak of 23 straight games as a conference underdog compare to other notable streaks since 2005? Well, it’s not even close to the longest streak, which I bet you can guess if you think about it for just a second.
Kansas has not been favored in a Big 12 game since 2009! The league has lost four teams and added two since the last time Kansas was favored. The Huskies are not even a third of the way to catching the Jayhawks and with the Huskies leaving the AAC, they won’t have a chance to cut into the lead this season. Of course, the Jayhawks streak is still active, and assuming things continue unabated as they have for the last decade, we may see their streak reach triple digits this November.

With so many of the longest conference underdog steaks currently active, I decided to include a separate table of active streaks of at least ten games.
Eight of the ten FBS conferences are represented with the ACC and MAC the lone holdouts. We’ll see you same time next week when we look at the AAC through the prism of the Adjusted Pythagorean Record.

Monday, January 13, 2020

Strangers in the Field Part V: How'd we Do?

I thought about never retiring this blog after my atrocious bowl season picks, but I know you were waiting with bated breath for another post. As I have done the past few years, I went to Vegas over the summer and made some wagers. Here is how they turned out.

Over/Under Win Totals

Auburn over 7.5 wins -115 ($30 to win $26.10)
Going off the trend regarding preseason top ten teams that finish unranked, I figured Auburn would improve upon their 7-5 2018 regular season and cash this ticket. The Tigers were quite fortunate to beat Oregon in the opener, and were 4-2 overall in close games during the regular season, but were probably the right side considering they beat this number by a game and a half.


Boise State under 10 wins +120 ($30 to win $25)
Despite this ticket being a loser, I think it was the right side. Excluding their lone regular season loss against BYU, the Broncos trailed in six of their other eleven games. They trailed in the second half of five of those games, including three times by double digits! Give the Broncos credit for their clutch play in winning those contests, but in the long run, they probably should have lost at least one more game.

Colorado over 3.5 wins -155 ($20 to win $12.90)
I was probably fortunate to cash this ticket as four of the Buffs five wins came by three points or less. Of course, three of their seven losses also came by a touchdown or less, so they weren’t exceptionally lucky in close games.

Florida State over 7.5 wins -110 ($40 to win $36.35)
After the Seminoles blew that big lead in the opener against Boise, this ticket was destined for the trash. The ACC was down in 2019, but the non-conference losses to Boise State and Florida meant the Seminoles would have no margin for error in the conference outside of an expected loss to Clemson.

Indiana 6 wins +110 ($30 to win $33)
I bet the under as the Hoosiers have four built in losses by virtue of the division they play in. Unfortunately, the Hoosiers won every game that looked like a tossup or potential loss in the preseason. They beat Maryland by six, Nebraska by seven, and Purdue by three. Oh, and Northwestern also fielded on of the worst offenses in FBS. This bet was probably bad, but was Indiana really eight win quality in 2019?

Notre Dame under 9.5 wins -140 ($40 to win $28.55)
With road games at Georgia and Clemson on the schedule, I figured the best Notre Dame could finish was 10-2. With seven other Power Five teams on the schedule, I figured the Irish would stumble at least one more time. Alas, close victories against Southern Cal and Virginia Tech (with Bud Foster calling the worst three man rush two minute defense I have ever seen), Bryce Perkins fumbling all over the field in the Virginia game, and Stanford’s decline allowed the Irish to get to double digit wins.

Oklahoma State over 7 wins -130 ($40 to win $30.75)
After a 3-3 start, the Pokes won four of their last five games to get this one home. I was probably a little fortunate this one cased, but at worst I was probably looking at a push.

Pittsburgh over 5.5 wins -165 ($30 to win $18.20)
This bet ended up hitting, but not for the reasons I envisioned. I thought Mark Whipple would keep the Pitt offense near the top of the ACC, but the defense carried the team to seven regular season wins.

Rutgers over 2.5 wins -165 ($20 to win $12.10)
With Massachusetts and Liberty on the non-conference schedule, I though the Knights would need just one conference win to hit the over. They took care of business against the Minutemen and Flames, but their closest conference loss came by 21 points to Penn State.

Southern Cal over 7 wins -130 ($50 to win $38.45)
I figured with the talent on hand and the addition of Air Raid concepts, Southern Cal would rebound from their 5-7 campaign in 2018. They made me sweat it out by losing their starting quarterback in the opener and losing to BYU after a 2-0 start, but behind Kedon Slovis, the Trojans did not lose another game they were favored to win the rest of the way. This bet wasn’t assured of cashing until they beat UCLA in the regular season finale, but I felt pretty good about a push once November rolled around.

Texas under 9 wins -110 ($20 to win $18.20)
I didn’t think Texas was back after their Sugar Bowl performance against Georgia. The Longhorns lost some close games (seven points each to LSU and Oklahoma), but also needed a field goal at the buzzer to edge Kansas. After the Longhorns lost to LSU and Oklahoma, I felt pretty good about at least getting this to push, but TCU, Iowa State, and Baylor came through for me to easily cash this ticket.

TCU over 7.5 wins +110 ($30 to win $33)
My blind faith in Gary Patterson was not rewarded. The Horned Frog offense looked lost for much of the season and the Frogs misses out on a bowl for the first time since 2013. Realistically, I should have stayed away since this number was more than seven. At seven I could justify the bet with the possibility of a push, but at seven and a half, it was just a bad bet.

Wake Forest over 5.5 wins -160 ($50 to win $31.25)
After a close victory against Utah State in the opener, I didn’t have to sweat this one at all. The Demon Deacons began the year 5-0 and cashed this ticket around mid-October.

Tampa Bay over 6.5 wins +105 ($30 to win $31.50)
Despite Jameis Winston’s propensity to keep the Bucs and their opponents in the game, we managed to get this one home. For a while, I thought the football gods were against me as Tampa opened the season 3-7 with a late missed field goal and a botched call costing them two games. They then won four in a row so I didn’t have to sweat out the last two weeks of the season.Overall, Tampa probably should have won eight or nine games, (3-6 record in close games and outscored their opponents on the year), so I think this was a good bet.

Games of the Year

September 13th
Houston +7 Washington State-110 ($30 to win $27.25)
I was right that Washington State would decline in 2019, but I didn’t anticipate Houston struggling as well. The Cougars (Houston edition) needed a late score to push this number, but they actually led at halftime and this game turned on a failed fourth down conversion by Houston in the second half.

September 21st
Oklahoma State +9.5 Texas-110 ($30 to win $27.25)
I doubled down on fading Texas and the Cowboys responded by getting a backdoor cover. Probably not the best bet I made all season, but the Cowboys had plenty of opportunities to win this game outright.

October 26th
TCU +3.5 Texas -110 ($30 to win $27.25)
I tripled down on fading Texas and TCU got the outright win by ten points at home.

November 2nd
Southern Cal +1 Oregon -110 ($30 to win $27.25)
The Trojans went up 10-0 early in this game, but turnovers and special teams gaffes made it a laugher early in the second half. Many were calling for Clay Helton’s firing after this embarrassing performance, but the Trojans rebounded and won their final three regular season games.

November 16th
Auburn +6.5 Georgia -110 ($50 to win $45.45)
This one hurt. The Tigers fell behind 21-0 before coming alive in the fourth quarter and cutting the lead to seven points. The Tigers actually got the ball back twice with a chance to tie, but a bad pass and another failed fourth down sealed the loss.

November 23rd
Baylor +6 Texas -110 ($30 to win $27.25)
I quadrupled down on fading Texas. Baylor was actually favored by about a touchdown in this game and the outcome was never in doubt.

November 30th
Stanford +4 Notre Dame -110 ($30 to win $27.25)
While I bought the Baylor game at a great price, I bought at a terrible price here. Notre Dame was about a seventeen point favorite when they traveled to Palo Alto. Stanford actually led 17-7 in the second quarter, but the Irish scored 31 straight points to put the game away.

November 30th
Florida State +13.5 Florida -110 ($30 to win $27.25)
Once again, placing my faith in Willie Taggart was a bad idea. Still waiting on the Seminoles to get their first defensive stop in this game.

December 14th
Navy +8 Army -110 ($50 to win $45.45)
Probably the best bet I ever made. Navy was a significant favorite by the time this game kicked off and I didn’t have to sweat it at all.

Conference Champion Bets

Tampa Bay Bucs to win the NFC South +1000 ($10 to win $100)
The Bucs were definitely better in 2019 than they were in 2018, but the Saints also reside in this division. They pretty much had it wrapped up by late October.

Reckless Parlay 1:
$10 to win $110

Game 1: August 24th
Miami +7 Florida
Winner.

Game 2: August 30th
Wake Forest -3.5 Utah State
Nope by a half point.

Game 3: August 31st
Michigan -31.5 Middle Tennessee State
Nope.

Game 4: August 31st
Auburn -2.5 Oregon
Winner.

Two out of four.

Reckless Parlay 2:
$10 to win $60

Game 1: September 8th
Tennessee +5.5 Cleveland
Winner.

Game 2: September 8th
Arizona +2 Detroit
Winner.

Game 3: September 9th
Denver +2.5 Oakland
Nope.

Money Wagered: $800
Money Won: $774.55
Return on Investment: -3.18%

If not for the bet on Tampa Bay to win the division and the two reckless parlays, I would have actually a small (very small) profit this trip. I was a few points from a very prosperous trip, as I lost the Auburn/Georgia game by a half a point and was a point away in the Notre Dame/Virginia Tech game from seeing the Irish go under their win total. All in all, I think I was a little unfortunate to lose money this year. Oh well, here's to better luck next season.

Tonight's title game is the final college football game until August. However, Statistically Speaking will help you get through the long offseason by starting the YPP and APR conference reviews on Thursday. For those that aren't regular readers, we'll review each FBS conference through the lens of Yards per Play (YPP) and the Adjusted Pythagorean Record (APR) with one post per week. As always, we'll go alphabetically starting with the AAC. That will get us close to Memorial Day and then I'll have some more sporadic posts over the summer until the football season begins anew. As always, thanks for reading and feel free to drop a comment should you feel the urge. See you on Thursday.