Stats & Info: NCF

Buckeyes show: Yards per game deceive

January, 14, 2015
Jan 14

Ronald Martinez/Getty ImagesOhio State’s defense Monday night performed better than traditional stats say.
In Monday night’s College Football Playoff National Championship Presented by AT&T, Oregon gained 465 yards against Ohio State. That’s about 56 yards more than the FBS average in a game in 2014, and it dropped Ohio State’s FBS rank in total defense from 17th to 19th. So by yardage numbers traditionally used to rate and rank defenses, Monday night’s performance was below average.

But defense is about more than yards allowed. Looking at it on a drive-by-drive basis, Ohio State held Oregon without points on 10 of its 14 drives -- six punts, two turnovers on downs, the end of the first half and the game-ending interception. Preventing the Ducks from scoring on fourth-and-goal early in the second quarter and holding them to field goal attempts on two other red zone drives were particularly strong defensive sequences the yardage numbers don’t capture well.

So even before adjusting for how strong Oregon’s offense is, Ohio State’s defense added more than four points to the scoring margin in terms of expected points added (EPA), taking into account all it did based on the situation before and after every play. That’s not a tremendous number, but given that the FBS average is about zero, it does show that the defense was a positive for the Buckeyes within the game.

To get a more accurate measure of Ohio State’s defense compared with other teams across all games, the performance should be adjusted for the fact it came against Oregon’s offense, which ranks best in FBS in terms of offensive efficiency (based on EPA per game, adjusted for opponent strength). With that adjustment, Ohio State’s defense comes out as plus-23.8 in the national title game -- an extremely positive performance. In fact, it was the third-best defensive performance of the 2014 bowl season and the 11th-best in any BCS/New Year’s Six bowl game going back 10 seasons.

In summary, if you use yards, Ohio State’s defense played worse than average Monday night. If you use expected points added, it looks better than average. And if you adjust that for the quality of Oregon’s offense, it looks way better than average -- which is more in line with reality.

Most champions rank better in efficiency than yards or points per game
National champion Ohio State finished the season fourth in offensive efficiency and 16th in defensive efficiency. Efficiency measures the contribution of the given unit to the scoring margin across all plays, adjusted for opponent strength.

The Buckeyes, like each of the eight national champions before them, ranked better in offensive efficiency than in yards per game and points per game at the end of the season. Specific to yards per game -- the default measure of offense -- the national champion ranked better in efficiency each time, by an average of 13 to 14 spots.

The story is similar but not as extreme with defense. Each of the past five and eight of the past 10 national champions rank at least as highly in defensive efficiency as yards or points allowed.

Looking across both offensive and defensive efficiency, each of the last 10 national champions has ranked in the top five of at least one of these categories at the end of the season, with nine of the 10 ranking in the top 16 of both categories. This cannot be said about the yardage or points-based measures of offense and defense.

A set of 10 good teams ranking better in efficiency than the traditional stats is by no means proof that the efficiency stats are better -- they are better because of all they take into account that the traditional stats miss or wrongly assign. But it’s also good to know that the efficiency numbers align more with reality than the traditional stats do for the offenses and defenses of the college football teams that finished at the top the past 10 years.

Efficiency hallmark of NFL-bound Mariota

January, 14, 2015
Jan 14

Ezra Shaw/Getty ImagesAmong FBS quarterbacks with 100 or more TD passes, Marcus Mariota has the fewest interceptions.
Marcus Mariota, one of the most efficient and accurate quarterbacks in college football history, announced Wednesday he will leave the University of Oregon with a year of eligibility remaining and make himself available for the NFL draft. The 2014 Heisman Trophy winner is the No. 2 player on the Scouts Inc. Big Board and No. 1 in Todd McShay’s first mock draft.

Mariota finished his Oregon career with the lowest interception percentage in FBS history (1.2 percent). Among quarterbacks with at least 14 starts, his 88.8 career Total QBR is the second best in the past 10 seasons. Of the top five players on that list, three were drafted No. 1 overall (Cam Newton, Andrew Luck and Sam Bradford).

It’s worth noting that the five Heisman Trophy-winning quarterbacks active in the NFL have a winning percentage of 46 percent. Early-entrant quarterbacks chosen in the first round in the past 10 seasons also have a winning percentage of 46 percent.

NFL draft
Mariota and Florida State’s Jameis Winston, the 2013 Heisman winner, are widely expected to be the first two quarterbacks chosen in the NFL draft this spring. Winston declared for the draft last week.

Early-entrant quarterbacks generally have struggled in the NFL. Since 2006, the first year of ESPN's Total QBR, 13 early-entrant quarterbacks have been chosen in the first round. Three -- Luck, Mark Sanchez and Vince Young -- have winning records as starters; four have posted career Total QBRs of 25 or worse.

If Mariota or Winston is chosen first, he would be the sixth Heisman Trophy-winning quarterback chosen first overall in the common draft era (since 1967). There are five active Heisman-winning quarterbacks in the NFL, and none of them has a career record above .500. The group had zero combined playoff wins until Newton and the Panthers defeated the Cardinals earlier this month.

Each of the teams in the top six of the NFL draft order finished in the bottom 10 in Total QBR last season. Mariota’s former coach, Chip Kelly, has the No. 20 pick with the Eagles.

College career
Mariota led the FBS in Total QBR this season (90.8) after ranking second each of the previous two seasons. He has 21 games with a Total QBR of 90 or greater (including an FBS-high nine games this season), the most by any player in the past 10 seasons.

From a total offense perspective, Mariota finished his career as the most efficient player in FBS history. There have been 180 players to gain at least 9,000 yards of offense, and six have averaged at least eight yards a play. Mariota is No. 1 at 8.67. Second, at 8.47 yards per play, is Luck of Stanford.

Among the 20 FBS players who have thrown at least 100 touchdowns, nobody has thrown fewer interceptions than Mariota, and it's not even close. Five players have 100 or more passing touchdowns and fewer than 30 interceptions. Mariota doesn't even have 15 interceptions.

Mariota improved against the blitz each season. His touchdown passes against the blitz have gone up every year, and his 2014 statistics against extra pass-rushers -- including a 71.6 completion percentage -- were his best across the board.

Mariota was consistently accurate (completion percentage of at least 58 percent in every game) and consistently productive (responsible for at least one touchdown in 28 of 29 halves this season). He didn’t play the second half against South Dakota and didn’t account for a touchdown in the first half of the Rose Bowl against Florida State.

Mariota finished his career with 14 interceptions in 1,167 passes. His rate of 1.2 interceptions per 100 passes is an NCAA record (min. 1,050 passes), breaking Geno Smith’s record (1.43 per 100).

Pac-12 tops final conference rankings

January, 14, 2015
Jan 14
After posting a Power 5-best 6-3 record in bowls, the Pac-12 rose to No. 1 in ESPN Stats & Information’s Conference Power Rankings -- a system that equally weighs the rankings from The Associated Press poll and ESPN’s Football Power Index (FPI) to determine the best and worst conferences in the country.

The Pac-12 not only had the best record, but it also outscored its opponents by more than eight points per game in bowls. Oregon, Utah and Stanford won by 24 or more points, and UCLA, Arizona State and USC won hard-fought contests. After an impressive bowl season, which included a trip to the national championship, the Pac-12 finished the year with six teams ranked in the AP poll, tied for the most of any conference with the SEC.

As the Pac-12 excelled, the SEC’s top teams struggled through bowl season. The SEC West in particular -- which was 28-0 against nonconference opponents entering bowls -- went 2-5, including 0-4 in New Year’s Six bowls.

Although many were quick to bury the SEC, the Conference Power Rankings account for the entire season and all teams in the conference, and the strength at the bottom of the SEC is unrivaled in college football.

We learned throughout bowl season, however, that there is more parity across the top of conferences than might have been expected. Each Power 5 conference had two teams in the top 10 of the final AP poll, and all had a team ranked in the top six.

After Ohio State's 42-20 victory over Oregon Monday night, the national champion is from the Big Ten for the first time since 2002. The Big Ten’s reputation had taken a hit in recent years, but Ohio State helped dispel the myth that the Big Ten could not compete with other top conferences by taking down the SEC and Pac-12 champions on its path to the title.

Michigan State and Wisconsin also beat top-20 opponents to help the Big Ten post its first winning record in bowls since 2009. Even more impressive is that the Big Ten was a Vegas underdog in all 11 of its bowl games and was expected to win about four games, according to ESPN’s Football Power Index. After exceeding expectations, the Big Ten rose 6.1 points in the Conference Power Rankings, the biggest rise of any conference during bowl season.

Elsewhere, the Big 12 had the worst bowl record of any Power 5 conference. Besides TCU’s 39-point victory over Ole Miss, the Big 12 was outscored by an average of 10.7 points in its other six games. The Big 12 ended the year 3-12 against nonconference opponents ranked in the top 40 of ESPN’s Football Power Index, the worst record of any Power 5 conference.

Next year, expect another great competition between the Pac-12 and SEC for the top conference in the nation. The SEC’s weakness this year, inexperience at quarterback, should become a strength as nine of 14 teams are returning their starting QBs next season.

The Big Ten and Big 12 also should make a splash with TCU, Ohio State and Baylor ranked as the top three teams heading into next season in Mark Schlabach’s Way-Too-Early Top 25. Keep an eye out for ESPN’s Preseason FPI, which will be released this spring, to see where your team and conference rank heading into next season.

Elliott, Buckeyes run over Ducks

January, 13, 2015
Jan 13

Scott Clarke/ESPN ImagesEzekiel Elliott rushed for 246 yards -- a national championship game record -- in the Buckeyes’ win.
Much of the attention heading into Monday’s College Football Playoff National Championship Presented by AT&T was centered on Ohio State’s backfield. That was a key area, but the attention, as it turned out, should have been on two players.

Ezekiel Elliott continued his late-season surge, and starting quarterback Cardale Jones belied his inexperience, helping carry the Ohio State Buckeyes to a 42-20 victory over the Oregon Ducks.

Elliott gained 246 yards and four touchdowns on 36 rushes, setting national championship-game records (since the BCS era began in 1998) for rushing yards and touchdowns. Texas’ Vince Young and USC’s LenDale White rushed for three touchdowns each in the 2006 Rose Bowl.

With his third consecutive 200-yard rushing game, Elliott also broke Ohio State’s bowl record for rushing yards, surpassing the mark of Raymont Harris, who had 235 yards in the 1993 Holiday Bowl in 1993.

Jones strong in bowl debut
Jones, thrust into the starting quarterback job after injuries to Braxton Miller and J.T. Barrett, didn’t play like a third-string quarterback. He entered the game with two career starts. That is the fewest starts for an official championship-winning quarterback, beating the 12 starts of Alabama’s AJ McCarron (2012 season), LSU’s Matt Flynn (2008) and Tennessee’s Tee Martin (1999) before the championship game.

Jones continued the type of play that had led Ohio State to 101 points in his first two starts. The 6-foot-5, 250-pound sophomore completed 16 of 23 passes for 242 yards with one touchdown and one interception.

The interception came on a drop when Oregon’s Danny Mattingly nabbed a bobbled ball at the end of Ohio State’s first drive of the third quarter. That interception was part of the Buckeyes’ minus-3 turnover margin, a stat that Ohio State is alone in being able to overcome in a championship game.

Four teams had been minus-3 or worse in turnover margin in the championship game era: Notre Dame (minus-3) in 2012, Texas (minus-3) in 2009, Oklahoma (minus-5) in 2004 and Miami (minus-3) in 2002. All four of those teams lost the championship game. And Oregon had been 31-1 in the past five seasons when having a turnover margin of plus-2 or better.

The Ducks didn’t exploit the turnovers or maximize their red zone opportunities. The Ducks scored 13 points in four red zone trips compared with Ohio State’s 35 points in five chances.

Led by Heisman Trophy winning Marcus Mariota, Oregon took the opening kickoff and drove for a 7-0 score. Mariota finished with 24 completions on 37 passes (two touchdowns, one interception), but he became the latest current Heisman winner to lose in the championship game.

Quick hitters
Ohio State’s win probability at halftime, leading 21-10, was 84 percent (Oregon’s was 16 percent). The Ducks ranked second in the FBS with on average an 80 percent chance to win at the half and never had lower than a 30 percent halftime win probability this season. … Ohio State improved its record under coach Urban Meyer when an underdog to 6-0. … The Buckeyes have trailed in four of their past five bowl wins.

Why RBs will decide CFP title game

January, 9, 2015
Jan 9

Getty ImagesEzekiel Elliott and Royce Freeman could be high-impact players on Monday.
Heading into the inaugural College Football Playoff National Championship, much of the focus has been on the quarterbacks, but it may be the running backs who ultimately decide the game.

Nine of the last 10 BCS National Championship games were won by the team that ran for more yards, and for Ohio State and Oregon, establishing a run game is essential to success.

Both teams run on more than 55 percent of their plays and use the run to open up the passing game. For this reason, Oregon's trio of running backs and Ohio State’s Ezekiel Elliott are key players to watch Monday night.

Oregon’s Trio of RBs
Oregon has utilized three running backs throughout the season – Royce Freeman, Thomas Tyner and Byron Marshall.

Freeman is Oregon’s leading rusher (95.9 YPG) and the first freshman in school history to run for 1,000 yards. He leads the Pac-12 in rushing touchdowns (18) and is tied for the lead in 20-yard rushes (12).

At 229 pounds, Freeman has added a toughness between the tackles that Oregon lacked in seasons past. He is the only player in the Pac-12 this season with at least 550 yards rushing both inside and outside the tackles, and he got stronger as the season progressed.

Tyner missed four games due to injury, but in the Rose Bowl CFP semifinal against Florida State he had a team-high 13 carries and a season-high 124 yards rushing. He adds a dimension of speed that pairs well with Freeman’s power.

Finally, Marshall was Oregon’s leading rusher in 2013, but he has been split out wide for most of this season. On 51 carries this season, he has averaged 4.7 yards before first contact and has gained at least 10 yards on 25 percent of those carries.

Why do they matter?
Oregon has attempted 63 percent of its passes off play-action this season, the highest percentage of any Power 5 offense. Establishing the run increases the effectiveness of play-action and opens up downfield throwing lanes.

Mariota leads all Power 5 players with 55 completions on passes thrown 15 yards or longer, and 44 of those completions came after a run fake.

An effective run game also forces teams to focus on the running backs on zone-read plays. One incorrect dive by a defensive end could mean trouble for Ohio State; Mariota is averaging 10.3 yards per rush and scores a touchdown once every seven rushes when he keeps the ball on zone-reads.

Ohio State RB Ezekiel Elliot
Since the start of November, Elliott ranks third in the FBS with 992 rushing yards, including 450 yards in his last two games.

With Cardale Jones at quarterback, Elliott is averaging a whopping 11.3 yards per rush, including 18.1 yards per rush on 10 zone-read rushes.

What sets Elliott apart is his ability to make plays between the tackles and speed to breakaway when he hits a hole.

He is averaging 7.7 yards per rush inside the tackles this season, second-best among Power 5 players (minimum 100 rushes), and has seven rushes of 40 yards or more (tied for fifth in the FBS).

Why does he matter?
If Ohio State can establish the run, Oregon may have to stack the box to stop Elliott. In the last two games, Ohio State’s passing game has relied heavily on downfield passes (53 percent of their pass yards in that time have been gained on passes of 20 yards or longer), so eliminating a safety could create one-on-one opportunities on the outside for Devin Smith.

Additionally, if Elliott is effective, he could be Ohio State’s best defense. Once the Ducks get on a roll, they are hard to stop; they have scored at least 20 consecutive points in 12 of 14 games this season. By sustaining drives, Ohio State will keep the Ducks’ potent offense off the field and increase their chances for offensive success.

Stat to Know
The team with the rushing advantage may be holding the trophy on Monday night. In the last three seasons, Oregon is 34-0 when it outrushes its opponent and 2-4 when it does not. Similarly, Ohio State lost the rushing battle in two of its three losses during that time.


Mariota's case for best season ever

January, 8, 2015
Jan 8

Steve Dykes/Getty ImagesThe numbers say Marcus Mariota has had a historically great season.
Oregon quarterback Marcus Mariota already has one remarkable performance in the College Football Playoff. And if he has his typical game and leads the Ducks to victory Monday night against Ohio State, a case could be made that his 2014 season will be the best in college football history.

Let’s go by the numbers to break down what Mariota has achieved and what’s at stake in the CFP National Championship Game.

Touchdowns vs. turnovers
56 – Mariota has been responsible for a Pac-12 record 56 touchdowns this season. He has been responsible for at least four touchdowns in 10 games, tied for the most such games in the last 10 seasons with Tim Tebow (2007) and Colt Brennan (2006).

50 – Mariota is on pace to become the first player in FBS history to account for at least 50 more touchdowns than turnovers in one season (56 touchdowns, six turnovers)

0.7 – With three interceptions in 408 passes, Mariota is threatening the record for lowest interception percentage in a season. If Mariota throws at least 23 passes in the national championship game without an interception, he will break the record for lowest interception percentage in a season, held by Boise State’s Kellen Moore (2009).

Heisman Trophy
10 – If Oregon defeats Ohio State on Monday night, Mariota will be the 10th quarterback in the poll era (since 1936) to win a Heisman Trophy and national title in the same season.

90.9 – Mariota received 90.9 percent of possible points in Heisman Trophy voting, second to Troy Smith in 2006 (91.6).

788 – The first-place votes Mariota received for the Heisman Trophy are the third-highest of any player in the history of Heisman voting.

Team and player efficiency
24.9 – Oregon leads the FBS with a +24.9 offensive efficiency this season, meaning the Ducks’ offense contributed almost 25 points per game to the team’s final scoring margin, compared with what an average offense would contribute. That is three points higher than any other FBS team this season.

91.7Total QBR is a complete measure of quarterback efficiency, and Mariota is on pace to post the third-highest single-season QBR since 2004. Neither of the two players with higher QBRs won a Heisman or a national championship.

Total player
4,852 – Mariota has 4,852 yards of total offense this season. His last three seasons rank in the top four in Oregon history. Mariota’s 4,121 passing yards this season alone would rank second in total offense in school history - behind his 4,380 yards last season.

ESPN’s Football Power Index predicts Oregon with a 63 percent chance of winning the inaugural College Football Playoff. The Ducks will again depend on Mariota’s arm and legs. Will he deliver another memorable performance to cap what might be the greatest season in college football history?

Winston leaves mark on FSU, eyes NFL

January, 7, 2015
Jan 7

Rob Tringali for ESPNJameis Winston will head to the NFL next season.
On Wednesday, Florida State quarterback Jameis Winston announced he would declare for the NFL Draft after his redshirt sophomore season. Winston is the No. 6 player on the Scouts Inc. Big Board and is projected to go No. 2 overall in Todd McShay’s first mock draft.

Florida State career
Winston had an impressive career at Florida State. He was 26-1 as a starter, with his loss coming in the Rose Bowl to Oregon last week. His 26 straight wins to begin his career is the longest win streak to start a career in the 21st century.

Winston had 65 career touchdown passes - second in Florida State history - in two seasons. He threw a touchdown pass in each of his 27 career games at Florida State.

Winston’s 2013 season was one for the ages. He won the Heisman Trophy, Walter Camp Award and Davey O’Brien Award, led the nation in Total QBR (89.4) and yards per attempt (10.6) and set an FBS freshman record with 40 Pass TD.

He became the sixth player since 1950 to win the Heisman, go undefeated and win the national title in the same season.

He had a drop-off in 2014, however. Winston had 20 turnovers this season, tied for fourth-most in the nation. His 18 interceptions tied for second-most in the FBS. Only four of the 18 came when Winston was under duress, and none were a result of a dropped ball.

Fifteen of Winston’s 18 interceptions in 2014 came on passes thrown 10 yards or longer. His 15 interceptions on such passes are three more than any other Power 5 quarterback.

A big part of Winston's 2013 success was performing well under pressure. He led all Power 5 quarterbacks by completing 60 percent of his passes when under duress in 2013, throwing for eight touchdowns and two interceptions. This season, he was under duress at a similar rate, but he completed 39 percent of his passes, throwing for two touchdowns and four interceptions.

Winston in the draft
Winston is likely to be a high pick in the draft, as the majority of the teams picking near the top could use some help at quarterback. Each of the teams picking in the top six of the draft finished in the bottom 10 in Total QBR this season.

Being a high pick is no guarantee of success, however. Since 2006, the first year of ESPN's Total QBR, 13 early-entrant quarterbacks have been chosen in the first round. Three have winning records as starters (Andrew Luck, Mark Sanchez and Vince Young). Four have posted career Total QBRs of 25 or worse.

Winston could be the third Florida State quarterback to be drafted in the first round. The first two were EJ Manuel (16th overall in 2013) and Christian Ponder (12th overall in 2011). If that happens, it would be the first time in the Common Draft Era (since 1967) that three quarterbacks from the same school were picked in the first round in a five-year span.

The previous six draft-eligible Heisman Trophy winners went in the first round, five of whom were quarterbacks. Of the five active Heisman-winning quarterbacks in the NFL - Johnny Manziel, Robert Griffin III, Cam Newton, Sam Bradford and Carson Palmer - none has a career record above .500. In fact, until this past weekend, when Cam Newton and the Panthers defeated the Cardinals, none of them had won a playoff game, either.

Top 10 bowl comebacks in last 10 seasons

January, 2, 2015
Jan 2
With two improbable comebacks this bowl season by Michigan State Spartans and Houston Cougars, we take a look at the most unlikely bowl comebacks in the last 10 seasons based on a team’s minimum win probability (lowest chance to win) throughout the game. Two of the top five most unlikely wins occurred in the last two days.

10. Oregon State def. Missouri (39-38) in the Sun Bowl
Dec. 29, 2006: min win prob = 2.7%

Oregon State converted a 4th-and-3 with 47 seconds remaining and then scored a 14-yard touchdown to pull within 1-point of Missouri. The Beavers were successful on a two-point conversion to take the lead with 22 seconds remaining and ended up winning the game, their fourth win by three points or fewer on the season.

9. Colorado State def. Washington State (48-45) in the New Mexico Bowl
Dec. 21, 2013: min win prob = 2.4%

In last year’s New Mexico Bowl, Colorado State scored 18 points in the final three minutes of the game, including a game-winning field goal as time expired. Washington State had two fumbles in those final three minutes to complete one of the worst collapses in a bowl game.

8. Louisiana-Lafayette def. San Diego State (32-30) in the New Orleans Bowl
Dec. 17, 2011: min win prob = 2.4%

San Diego State scored a go-ahead touchdown with 35 seconds remaining to take a one-point lead. When the game seemed all but over, the Ragin' Cajuns drove 49 yards and kicked a game-winning 50-yard field goal.

7. North Carolina def. Tennessee (30-27, 2OT) in the Music City Bowl
Dec. 30, 2010: min win prob = 1.8%

Highlighted by a controversial ending in regulation in which it appeared that time had run out, North Carolina was given another chance to kick a game-tying field goal. The Tar Heels went on to win the game in overtime.

6. Arizona def. Nevada (49-48) in the New Mexico Bowl
Dec. 15, 2012: min win prob = 0.8%

Arizona outscored Nevada 21-3 in the fourth quarter and scored twice in the final 46 seconds to win the New Mexico Bowl. It was the first of two straight improbable New Mexico Bowl victories (see No. 9 above).

5. Houston def. Pittsburgh (35-34) in the Armed Forces Bowl
Jan. 2, 2015: min win prob = 0.7%

Houston recovered two onside kicks and overcame the largest fourth-quarter deficit in a bowl game (25 points) to defeat Pittsburgh on Friday. At its lowest point, when trailing by 25 points early in the fourth quarter, Houston had a 0.7 percent chance to win.

4. Texas Tech def. Minnesota (44-41, OT) in the Insight Bowl
Dec. 29, 2006: min win prob = 0.5%

The Red Raiders trailed by 31 points early in the third quarter before scoring 31 unanswered points to send the game to overtime. Texas Tech won it there, marking the largest comeback in bowl history.

3. Michigan State def. Baylor (42-41) in the Cotton Bowl
Jan. 1, 2015: min win prob = 0.4%

Trailing by 20 entering the fourth quarter, Michigan State had a win probability as low as 0.4 percent before blocking a Baylor field goal with 1:05 remaining that would have put the Bears up by nine. The Spartans scored three touchdowns in the fourth quarter, including a game-winning touchdown with 17 seconds remaining, to complete their improbable comeback.

2. Boise State def. Oklahoma (43-42, OT) in the Fiesta Bowl
Jan. 1, 2007: min win prob = 0.1%

Trailing by seven points under a minute to play, Boise State completed a 50-yard hook-and-ladder play on 4th-and-18 to tie the game against Oklahoma. The Broncos won the game in overtime on the famed "Statue of Liberty" play.

1. Idaho def. Bowling Green (43-42) in the Humanitarian Bowl
Dec. 30, 2009: min win prob 0.1%

Trailing by seven points with 32 seconds remaining, Idaho drove the length of the field for a touchdown and converted a two-point conversion to take a 43-42 lead and win the game.

Stats to know: Ohio State 42, Alabama 35

January, 2, 2015
Jan 2
Boy, o-h boy!

It was a heck of a comeback win for Ohio State that propelled it into a national championship with Oregon, a 42-35 victory over Alabama .

The Buckeyes rallied from a 15-point deficit, their largest comeback victory since Oct. 28, 1989, when they rallied from 31-0 down to beat Minnesota 41-37.

It was the second-highest-scoring game in Sugar Bowl history (77 combined points). The record is the 81 total points scored by LSU (47) and Illinois (34) on Jan. 1, 2002.

The history
It was the first win in four meetings for Ohio State against Alabama. The Buckeyes have now won 12 straight overall, the longest current winning streak in the country now that Florida State’s 29-game win streak has been snapped.

Cardale Jones is the first Ohio State quarterback to win a bowl game without being the team’s season-opening starter since Troy Smith in 2005.

Alabama fell to 0-3 in Sugar Bowls with Nick Saban as head coach. The Crimson Tide fell to 5-2 in bowl games in New Orleans when ranked No. 1 or No. 2 in the AP poll.

Ohio State running back Ezekiel Elliott set a Sugar Bowl record for rushing yardage with 230 yards in this win.

Elliott is the first player to rush for at least 200 yards in a game against Alabama since Cadillac Williams in 2003.

Elliott now has 1,632 rushing yards this season, the fourth-most in a season in school history and the most since Eddie George ran for a school-record 1,927 yards in 1995.

Elliott's 85-yard TD run is the longest rushing play Alabama has allowed under Nick Saban

Ohio State had more rushing yards in this game (281) than the previous four Big Ten opponents against Alabama combined (255).

Stat to remember: Third-down effectiveness
The difference in third-down performance was a big reason Ohio State pulled out the win.

The Buckeyes were 10-for-18 with two touchdowns on third downs. Alabama was 2-for-13 with two turnovers on third down. The Crimson Tide converted 54 percent of their third-down opportunities during the regular season.

Unsung heroes: The punters
Cameron Johnston and JK Scott probably aren’t known outside of Ohio State and Alabama, respectively, but both had huge games for their teams.

Scott, who would have had a good argument for MVP honors if Alabama had won, averaged 56.4 yards on seven punts with a long of 83. Johnston had one shank but was sharp other than that, averaging 46.5 yards with a long of 67.

Did you know?
In each of the past two seasons, the SEC champion blew a 21-6 lead in its bowl game. Auburn did it last season in the BCS Championship against Florida State. Alabama did it in this game against Ohio State.

NCF, Big Ten, SEC

Stats to know: Oregon 59, Florida State 20

January, 1, 2015
Jan 1
An illustrious streak came to an end. A quarterback suffered his first collegiate defeat. And with that Oregon is headed to the national championship game.

The Ducks beat Florida State in a rout, 59-20. It was a record-setting day on a number of fronts.

The history
Oregon broke the previous mark for points in a Rose Bowl, which was shared by 1902 Michigan, 1948 Michigan and 2004 USC Trojans (49). It also broke the record for yardage in a Rose Bowl set by the 2007 USC team against Illinois (633).

Florida State had its 29-game winning streak and six-game bowl winning streak snapped. The 29-gamer was tied for the fourth-longest streak among Power 5 teams in the last 50 years. Quarterback Jameis Winston lost for the first time in his career.

The Seminoles tied their school record for points allowed in a game, matching the 59 they allowed to Auburn in 1985. The 39-point loss is the largest by a defending national champ in a bowl game.

This was the fourth matchup of a team with a 20-game win streak against a Heisman Trophy winner. It’s the first time that the Heisman winner ended up winning that game.

Oregon got its first win in 21 tries against a team ranked first or second by the AP.

The Ducks had their 41st 50-point game in the past 10 seasons.

No other school has more than 30 in that span (Boise State, Oklahoma).

The turning point
The game turned in the third quarter when Florida State turned the ball over four times. The first three of those miscues led to Ducks scores within the quarter, the last led to a touchdown in the fourth quarter.

Oregon's 27 points in the third quarter tied a Rose Bowl record for points in any quarter (also done by Illinois vs. Stanford in the 1951 season and Miami vs. Nebraska in the 2001 season).

Florida State allowed 120 points off turnovers this season (27 coming in this game). Only Connecticut Huskies and Georgia State have allowed more (127).

Heisman Trophy winner Marcus Mariota excelled, throwing for 338 yards.

He got helped big-time by Darren Carrington, who had seven receptions for 165 yards and two touchdowns.

Carrington entered the day averaging 60 yards receiving per game and with only two touchdown receptions all season.

His yardage total ranks fifth in the history of the Rose Bowl.

Did you know?
As bad as it got for Florida State, the Seminoles were nowhere near the record for most turnovers in a Rose Bowl. The still-standing mark is 12, set when Notre Dame had four fumbles and eight interceptions against Stanford in 1925.

Florida State does finish the season with the second-most turnovers by an FBS team (32). Only 2-10 New Mexico State had more (34).


Boise State continues Fiesta magic

December, 31, 2014
The Fiesta Bowl has brought out the best from the Boise State Broncos over the last decade, and Wednesday’s game was no different.

Boise State defeated the Arizona Wildcats 38-30 in another thrilling Fiesta Bowl win.

Boise continues Fiesta magic
Boise State is now 3-0 at the Fiesta Bowl, the second-best record of any team (Penn State 6-0). The Broncos were the underdog in each of their Fiesta Bowl trips and won each game by single digits.

Boise State has now won 12 games for the eighth time in school history, all since 2002. That's two more 12-win seasons than any other program in the country in that span.

Hedrick-Sperbeck connect
Boise State quarterback Grant Hedrick completed 24 of 34 passes for 309 yards, with Thomas Sperbeck serving as his go-to receiver. Hedrick completed 12 of 14 passes and averaged 14.2 yards per attempt when targeting Sperbeck but averaged only 5.5 yards per attempt when targeting all others.

Sperbeck’s career-high 199 receiving yards were tied for the second most in Fiesta Bowl history, trailing only Kansas State's Darnell McDonald’s 206 yards in 1998.

Hedrick completed 14 straight passes to start the game, tying the record for any bowl game (set previously this season by Louisiana-Lafayette's Terrance Broadway in the R+L Carriers New Orleans Bowl). It’s no surprise it was Hedrick who tied the record as he leads all FBS quarterbacks in completion percentage this season (70.8 percent).

Ajayi’s record-breaking day
Jay Ajayi ran for 134 yards and three touchdowns, breaking multiple Boise State records in his final collegiate game.

Ajayi finished with 1,823 rush yards this season, breaking Ian Johnson’s single-season school record, set in 2006 (1,713). With his second touchdown, Ajayi also surpassed Brock Forsey’s school rushing touchdown mark of 26, set in 2002.

More than 78 percent of Ajayi’s rush yards in the Fiesta Bowl came after first contact. For the season, Ajayi had 1,033 rushing yards after contact, 336 more than any other Mountain West player.

Solomon struggles to connect downfield
For Arizona QB Anu Solomon, the Fiesta Bowl was a display of freshman mistakes and future promise. He completed 28 of 49 passes for 335 yards, a touchdown and two interceptions. It was his 11th touchdown pass outside the pocket this season, the most of any Power 5 player.

While he did throw for over 300 yards, Solomon struggled to connect downfield, completing only 2 of 10 passes thrown 15 yards or more. Solomon also took a career-high eight sacks, four more than any other game this season.

Top stats to know: TCU dominates Ole Miss

December, 31, 2014
On paper this looked like an even matchup.

ESPN’s Football Power Index gave the Ole Miss Rebels a slight 51 percent chance to win the game.

The two teams were ranked in the top five in defensive efficiency and the top six in FPI this season.

But once they got on the field, the TCU Horned Frogs picked off Ole Miss quarterback Bo Wallace on the Rebels' first drive and never looked back as they dominated the Chick-fil-A Peach Bowl.

Wallace’s struggles continue
Wallace entered the game having thrown three interceptions without a touchdown in his previous two games. Things got worse on New Year’s Eve as he completed just 10 of 23 passes for 109 yards, throwing three interceptions and losing a fumble.

Wallace was eventually pulled from the game and finished with a career-low Total QBR of 4.7.

TCU’s defense steps up
The Horned Frogs entered the bowl season ranked second in the FBS with a plus-18 turnover margin and 36 takeaways. They added four more takeaways and scored a touchdown on an interception in the end zone.

Ole Miss had 16 drives and put up just three points. Aside from the four turnovers, the Rebels were forced to punt eight times. Seven of those eight punts were the result of a three-and-out.

Aside from a 15-play, six-minute drive that ended in a fumble in the third quarter, Ole Miss did not have a single drive that lasted over two minutes.

Ole Miss finished with 9 rushing yards, its fewest in a game since it had 7 on Nov. 19, 2005, against LSU in a 40-7 loss. The Rebels finished with more rushes for negative yardage (11) than first downs (10).

A day for the record books…
The last time TCU beat an SEC team in a bowl game was back in 1936, when “Slingin’” Sammy Baugh led it to a win in the Sugar Bowl over LSU.

TCU’s 39-point margin of victory is the largest in Peach Bowl history. The previous record was 37 points (LSU over Miami (Florida) in 2005).

It was also TCU’s largest margin of victory in a bowl game in school history (the previous best was a 30-point win over Northern Illinois in the 2006 Poinsettia Bowl).

And, as you might have guessed, it was Ole Miss’ largest margin of defeat in a bowl game in school history (the previous worst was a 32-point loss to Michigan in the Gator Bowl on New Year's Day 1991).

FSU offense, defense bail each other out

December, 30, 2014
Florida State has made a habit of winning close games this season. You've probably heard a lot about the improbable comebacks, the lucky plays, the ability to “turn it on” when they fall behind and all other such things. But one pattern that may have gone under the radar is how Florida State has kept winning because both its offense and its defense has the ability to “bail each other out” -- when one has played relatively poorly, the other played well enough to make up for it and get the win.

According to the team efficiency ratings here, which are in terms of the contribution of that unit to the scoring margin and adjusted for quality of opposition, Florida State’s offense is ranked 16th at plus-13.6 and the defense is ranked 44th at plus-1.9 this season. So both units contribute positively to the scoring margin overall, but the offense has been much better on the whole.

Look at those numbers broken down game by game and you can see that each unit has had its ups and downs around those averages. The bar chart shows Florida State’s offense (red) and defense (green) in each game, going chronologically from left to right. Higher bars are better for both units, with the horizontal lines depicting Florida State’s season average in each category (again, offense higher than defense).

You can see how the red and green bars don’t move together in unison: the Seminoles’ best offensive games are usually their worst defensive games, and vice versa. There is no single game in which the red bar is above the red line and the green bar is above the green line -- in other words, the offense and defense have never performed above their season averages in the same game.

One simple way to summarize this pattern numerically is to break down each unit’s performance by the other unit’s best and worst games, as in the following chart.

To put the numbers in further perspective, the offense played up to the level of Baylor this season (second in FBS) in the defense’s six worst games, while playing as well as Duke (44th in FBS) in the defense’s six best games.

Conversely, the defense played about as well as Utah (14th in FBS) in the offense’s six worst games, but was more like Vanderbilt (77th in FBS) in the offense’s six best games.

Below is a scatter plot where you can see Florida State’s offense and defense in each game plotted against one another (offense on horizontal axis, defense on vertical), with the opponent logos used to denote which game each point represents.

You can see the clear negative correlation between how the units perform on a game-to-game basis. And again, there are no games in which offense and defense both played above the season average (nothing in the top right of the graph). The correlation coefficient, which measures the strength of a linear relationship between two variables on a scale from minus-1 (perfect negative correlation) to 1 (perfect positive correlation), is minus-0.57, indicating a moderately negative relationship between the two units’ performance across games.

While it’s clear this weird negative correlation exists for Florida State, how rare is it? There are actually 12 other FBS teams with as negative of a correlation this season, but only two of those teams finished in the CFP Top 25: Louisville and Kansas State.

The other three playoff teams have very neutral correlations, indicating no real relationship between their offensive and defensive performance on a game-by-game basis. In this sense, Florida State is a bit of an anomaly for a top-tier team.

The Florida State offense and defense have been able to bail each other out enough to keep the team undefeated to this point, but FSU will face its strongest opposition by far in the CFP semifinal against Oregon. It’ll likely take a strong effort from both sides of the ball in the same game for the Seminoles to continue their quest for a repeat national title.


Bowl's 55-48 score wasn't all about offense

December, 23, 2014

Steve Mitchell/USA TodayPaxton Lynch accounted for 7 TD, but the Miami Beach Bowl wasn’t that much of an offensive show.
Memphis defeated BYU in double overtime in Monday’s Miami Beach Bowl, winning by a final score of 55-48. The 103 total points is tied for the third-most in any bowl game ever.

It might seem as if 55-48 has to represent an offensive shootout with little to no defense, but if you look at what happened in the game, it’s clear that the high scoring is not necessarily indicative of good offense.

  • The teams combined for nine turnovers. One was a fumble lost by Memphis’ special teams, but each offense had three interceptions and one fumble lost. Turnovers are huge negatives for offense and huge positives for defense, negating a lot of yardage in terms of their effect on point margin (erasing points and setting up the opponent to score).
  • Both teams had more than 400 yards of offense (Memphis 480, BYU 425), but some of that has to do with how many possessions there were in this game. Partly due to the scoring and partly due to overtimes, each team had 21 drives in the game. So if you look at it in terms of average yards per drive, it’s less than 23. Compare that to the FBS average this season of about 31 yards per drive, and it’s not that great.
  • The teams combined for 15 punts (BYU nine, Memphis six). Combining punts and offensive turnovers, 10 more drives ended with defensive stops (23) than offensive touchdowns (13).
  • The defenses did a great job setting up the offenses to score. In addition to the defensive touchdown (BYU’s pick-six in the fourth quarter), the offenses had 11 combined drives where they started on the opponent’s side of the field. BYU’s five such drives resulted in two touchdowns, two field goals and one turnover; Memphis had six such drives, resulting in four touchdowns, one field goal and one missed field goal.
  • In the first overtime, both offenses moved backward, losing yards. But both kickers stepped up, nailing 45- and 54-yard field goals to force double overtime. Memphis scored a touchdown on its possession in the second overtime, but the game was in fact sealed by another big defensive play – a game-ending interception by Memphis.
Expected points added
Put it all together using expected points added, and the final totals look like the chart to the right (before opponent adjustment, just looking at this game in isolation).

In other words, once you look at all the offense and defense did to contribute to the scoring margin across all plays, Memphis’ offense was slightly above average, and BYU’s offense was actually fairly below average. The 55-48 final score is misleading in terms of how productive the offenses really were: The high point totals were actually more of a result of defenses setting up the offenses and the number of drives in the game.

ESPN's Football Power Index: A look back

December, 18, 2014

Wire service photosAmari Cooper, left, and Marcus Mariota helped their teams meet preseason FPI projections.
With the regular season complete and bowl matchups set, it’s time to look back at the 2014 college football season. Just as teams reflect on their seasons to evaluate their performance, ESPN Stats & Information will do the same for its Football Power Index.

Below is a breakdown of how FPI performed throughout the year. Which preseason projections were correct? How accurate was FPI at predicting games? This is designed to be unfiltered and informative, so if you have questions, feel free to ask them in the comments section below and we will do our best to answer.

As a quick reminder, ESPN’s Football Power Index is a forward-looking system designed to order each FBS team by which is most likely to beat an average team on a neutral field. FPI is intended to measure team strength, not evaluate a team’s résumé for the playoff, to best predict future performance. Once team strength is captured, FPI can be used to go through each team’s remaining schedule to produce game and season projections (expected W-L, chance to win conference, chance to win out, etc.).

Preseason information
FPI was improved this season to allow for preseason projections. Factored into the preseason ratings were prior years’ efficiencies (offensive, defensive and special teams), recruiting data, coaching tenure and information on returning starters.

Where FPI was correct in the preseason
• Preseason top 3: The top three teams in preseason FPI -- Florida State, Oregon and Alabama -- performed as expected. Ohio State was ranked sixth until days before the season when Braxton Miller injured his shoulder and returning starter information was adjusted. The Buckeyes dropped to 12th.

• Preseason W-L projections: Team strength must be weighed with the difficulty of a team’s schedule to accurately predict win totals. The teams ranked fourth through sixth in preseason FPI -- UCLA, Auburn and Stanford -- had some of the toughest schedules in the nation. Therefore, FPI predicted that they would lose more than three games on average (and they did). The five teams that FPI projected for the highest win totals were the only five teams in the nation with 12 or more wins. Similarly, the top five teams from Power 5 conferences in projected win total finished the season as the top five teams in the final CFP rankings.

Along those same lines, FPI projected that Florida State had a 38 percent chance to enter bowls undefeated and no other FBS team had more than a 10 percent chance to win out. The Seminoles are the only undefeated FBS team. The second-most likely team to go unbeaten in the preseason -- Marshall -- had a great chance to accomplish that feat through 11 games.

• Conference projections: Seven of the 10 preseason FPI favorites to win a conference went on to do so, including four of the six that were given the highest chances. The three FPI favorites that didn’t win were in the Big 12 (Oklahoma), Sun Belt (Louisiana-Lafayette) and American (Houston). In the case of the Big 12, Oklahoma (35 percent) and Baylor (33 percent) were very close in the preseason. The Big Ten is an interesting case. FPI favored Ohio State to win the conference without Miller, despite most of the public picking Michigan State.

Where FPI was incorrect in the preseason
• Teams FPI underestimated: Georgia Southern, TCU and Georgia Tech

Sun Belt champion Georgia Southern was transitioning to the FBS, and FPI underestimated the explosiveness of its offense, which finished 23rd in offensive efficiency. FPI projected that the Eagles would finish about 4-8, but they ended the year with a 9-3 record.

After finishing 4-8 last year, TCU’s improvement in 2014 was a surprise to many. FPI had the Horned Frogs 36th in its preseason rankings, which was higher than most but still significantly below their current fifth-place ranking. FPI was high on TCU’s defense (seventh in the preseason) but did not envision its offensive improvement after it ranked 99th in offensive efficiency in 2013. Overall, FPI projected that the Horned Frogs would enter bowls with around seven wins, and they have won 11 games.

Georgia Tech lost 11 starters, including its quarterback, from last year's seven-win team. In the preseason, FPI projected that the Yellow Jackets would win about six games and had a 5 percent chance to win the ACC Coastal Division. Instead, Georgia Tech reached 10 wins for the first time since 2009, and the Yellow Jackets seek their first Orange Bowl win since 1951.

[+] EnlargeKliff Kingsbury
Michael C. Johnson/USA TodayNo defense in the Power 5 had a lower efficiency than that of Texas Tech and coach Kliff Kingsbury.
• Teams FPI overestimated: Michigan, Texas Tech and South Carolina

Michigan was returning 15 starters, including quarterback Devin Gardner, to a team that had four of its six losses come by a combined 11 points last season. The Wolverines were ranked 18th in preseason FPI and were projected to have the 16th-best defense in the country. FPI projected that the Wolverines would have about four more wins than they actually did, marking its biggest miss this year.

Texas Tech was supposed to have a top-10 offense this year, but injuries and turnovers stymied the Red Raiders' air attack. And no Power 5 team finished the regular season with a lower defensive efficiency than Texas Tech. It’s safe to say that preseason FPI whiffed on four-win Texas Tech.

South Carolina was ninth in the preseason Associated Press poll, so FPI was not alone in its overestimation of the Gamecocks. Since South Carolina had a favorable divisional schedule, FPI projected that it had the best chance to win the SEC East and the second-best chance to win the SEC. The Gamecocks finished with three conference wins, four fewer than SEC East champion Missouri.

In-season projections
While the preseason ratings served as the basis for FPI, an important part of the system is that it learns from each game during the season and adjusts appropriately as teams play better or worse than expected. This mechanism allows FPI to be fluid as the season goes on, which improves prediction accuracy from week to week.

The team FPI favored won 77 percent of FBS-only games this season, which is better than the win percentage of the Vegas closing-line favorite. There were 50 games in which FPI and the Vegas line differed on the favorite; FPI went 28-22 (56 percent) in those games, including 17-10 in the final eight weeks.

Interestingly, FPI exceeded expectations in games involving teams that finished the season ranked in the CFP Top 25. Most systems would be expected to correctly predict about 66 percent of such games, but the FPI favorite was 34-12 (74 percent), including 18-2 in the final five weeks.

There were certain teams that FPI had a grasp on and others that baffled the system. There were 10 teams, including Michigan State, Clemson, Washington, Florida State and Texas Tech, for which FPI correctly predicted all of their FBS versus FBS games. Add in another 28 teams for which FPI correctly predicted all but one game and the system had a very good understanding of about a third of the FBS.

Texas Tech is a great example of how FPI adjusts as the season progresses. As noted above, FPI was high on the Red Raiders in the preseason but quickly learned of their flaws and adjusted its in-season projections accordingly. FPI correctly identified the favorite in all 11 of Texas Tech’s FBS games.

On the other end, there were five teams -- Virginia Tech, South Carolina, Akron, Utah and Air Force -- where the FPI favorite lost in more than half of the games. Missouri and Texas A&M also were tough to predict with their fluctuating performances.

We have been able to retroactively apply FPI to the past 10 seasons. Since 2005, FPI has correctly predicted 75 percent of FBS games. It is on pace to have its second-best pick percentage in a season but will finish the year far from its 79.5 correct-pick percentage in 2013.

For those looking for a little bowl advice, FPI projects that Marshall (79 percent), Stanford (79 percent) and Georgia (77 percent) are the most likely teams to win their bowl games, and Navy (60 percent) is the Vegas underdog most likely to win.