The three best dual threat quarterbacks in the NFL right now are Cam Newton, Russell Wilson, and Deshaun Watson. You can make a good case that they were also the three best quarterbacks of week 8.

Watson was the most valuable passer of week 8, throwing for 5 TDs on just 20 dropbacks, with zero interceptions and zero sacks. Wilson was the second most valuable passer of the week, and became just the second player in the last three seasons to have a perfect passer rating while throwing at least three touchdown passes. And Newton led all quarterbacks with 52 rushing yards and a touchdown, led the Panthers to 36 points on 7 scores on 10 drives, and did not take a sack or an interception.

The full week 8 passing stats below. The average ANY/A this week was 6.98; as always, Value is calculated as the difference between each passer’s ANY/A and the weekly average, multiplied by that passer’s number of dropbacks. [continue reading…]

Jameis Winston entered the NFL as a 21-year-old in 2015. Now, midway through his fourth season, Winston still has a number of rookie struggles. He currently has a 6.8% interception rate in 2018, bringing his career interception rate up to 3.2%. Among the 39 passers with 500 attempts since 2015, Winston has the worst interception rate in the NFL.

On the other hand, interceptions are not a very sticky statistic, and a pretty poor way of evaluating quarterback play.  Winston is an aggressive passer who has played on bad teams, so that tends to lead to interceptions.  And young quarterbacks — Winston is still just 24 — sometimes struggle with interceptions, especially on bad teams.

Meanwhile, Winston has been downright remarkable at picking up first downs.  Including sacks, Winston has picked up a first down on 37.3% of his dropbacks since entering the league; that’s the 2nd-best rate in the NFL over that span.  Last year, he led the league in percentage of pass attempts that went for a first down; this year he’s been even better (although he only ranks 3rd).  Including sacks, he has still picked up a first down on 40% of his dropbacks.

Since 2015, Winston ranks 1st in INT rate and 2nd in Passing First Down rate.  No other player ranks in the top 9 in both, although Roethlisberger, Newton, and Fitzpatrick rank in the top 12 in both stats. [continue reading…]

It was not a particularly eventful week in terms of college football upsets. No top-5 SRS teams lost, and only two top-10 SRS teams lost — and both lost to very good teams. Iowa ranked 7th in the SRS last week, and lost on the road to Penn State, who ranked 11th in the week 8 ratings. The annual Florida/Georgia game in Jacksonville featured two top 10 teams, with then-#3 Georgia defeating then-#8 Florida.

Other top teams to lose: #13 Washington, #15 Texas A&M, #16 Purdue, #21 Texas Tech, #22 Missouri, #23 Wisconsin, and #26 Texas. The Longhorns may have been 26th in the SRS, but with a 6-1 record and wins over USC, TCU, and Oklahoma, the Longhorns entered week 8 6th in the AP rankings. A loss to Oklahoma State (hi Doug!) in Stillwater now puts the Big 12 playoff picture into chaos. Next week, West Virginia travels to Austin in a matchup of two of the top teams in the conference.

Alabama was off this week, which opened the way for Georgia and Clemson to be the dominant teams of the week. Both teams demolished teams from Florida: the Bulldogs defeated a pretty good Florida team by 19, while Clemson crushed a bad Seminoles team by 49 points. West Virginia, Arizona, and Utah round out the top 5 teams of the week in terms of single game SRS ratings. [continue reading…]

Against the Jets in week 5, trailing 34-16 with 11 seconds left, Case Keenum threw an interception on the game’s final play.

Against the Broncos in week 1, the Seahawks had the ball with 2 seconds left at their own 11-yard line, down by 3. Russell Wilson threw an interception on the game’s final play.

Against the Panthers in week 3, the Bengals were down 31-21 with 3 seconds left when Andy Dalton threw an interception on the game’s final play.

Trailing 30-14 against the Chiefs, with 9 seconds left, Blake Bortles threw an interception on the game’s second-to-last play.

All of these are examples of meaningless interceptions, at least from a win probability perspective. These teams all had a less than one percent chance of winning the game, and the interception therefore was not meaningful. Tom Brady had one of these, too: the Patriots had a win probability of over 99% when, up 38-0 in the 4th quarter, he threw an interception. That pick was meaningless.

We should not treat all interceptions equally, and with win probability data becoming more mainstream, we are no longer restricted by a game’s box score. A special thanks to Ron Yurko (@Stat_Ron) for providing the raw data. [continue reading…]

Vikings defense end Danielle Hunter has recorded a sack in every game this season, making him just the 8th player since 1982 to record a sack in each of his team’s first 8 games:

[continue reading…]

In 2017, the Houston Texans lost their final six games of the season. Tom Savage started the first of those three games, with T.J. Yates starting the final three. This year, with Deshaun Watson back from injury, the Texans… lost their first three games of the season. So entering week 4, Houston was on a 9-game losing streak. Since then? Houston has won four straight, and with a home game against the Dolphins tonight, have a good chance of running that winning streak to five straight. With a win in Denver next week, the Texans would have followed a streak of 6+ straight losses by, immediately, winning 6+ straight games.

That’s pretty rare: only six teams have ever strung together back to back streaks of 6+ wins and 6+ losses. And if Houston runs the streak to 9 straight wins, the Texans would be the first team to ever have back-to-back 9-game streaks.

The longest streak? That belongs to the San Diego Chargers. In 1987, the Chargers lost the season opener, but responded by winning 8 consecutive games (including three with replacement players). And then San Diego promptly lost 8 straight games, marking the only time a team has ever had back-to-back 8-game winning/losing streaks. For those curious, a 36-year-old Dan Fouts was the starter for 5 of the games during the winning streak, and four of the games during the losing streak, which stretched into 1988.

Just two other times has a team had a 6+ game winning streak followed by a 6+ game losing streak. In 1977, the Don Coryell Cardinals began the season 1-3 before Jim Hart and crew went on a 6-game winning streak, bringing St. Louis’s record to 7-3. From that point on, the Cardinals lost their final four games of the regular season and the team’s first eight games of 1978 under Bud Wilkinson.

Finally, the 2001 Rams under Kurt Warner won their final six regular season games, plus the team’s two NFC playoff games, allowing St. Louis to enter the Super Bowl on an 8-game winning streak. The Rams lost to the Patriots in the Super Bowl, of course, and then began the 2002 season on a 5-game losing streak, with Warner the starter for all but the final game. And at 0-5, the Rams responded with a 5-game winning streak, then under Marc Bulger!

What about teams like the Texans, who followed long losing streaks with long winning streaks? There are just three of those teams, using 6-game minimums.

In 1970, the Cincinnati Bengals had one of the more interesting seasons in NFL history.  With Paul Brown and Bill Walsh as offensive masterminds, the Bengals found a gem in Greg Cook to be the team’s quarterback in 1969.  Unfortunately, Cook suffered what would be a career-limiting injury, and the Bengals brass turned the ball over to Virgil Carter.  You know the story from there, as Carter, Walsh, and Brown helped usher the West Coast Offense into the NFL.  After winning the team’s opener, Cincinnati lost 6 straight games, before going on a 7-game winning streak to end the regular season.  The Bengals lost the team’s only playoff game.

In 1994, the New York Giants began the season 3-0, then lost 7 straight games, and then won the team’s final 6 games. Dave Brown was the team’s quarterback for all but one game.

Finally, we have the 1974 Jets, with a 31-year-old Joe Namath who managed to play every game that season.  New York lost its first game, won its second game, and then won on a 6-game losing streak followed by a 6-game winning streak to end the season.  From weeks 3 through 8, the Jets ranked 25th in passer rating and 17th in Net Yards per Attempt; during the final six weeks, the Jets ranked 3rd in passer rating and 3rd in NY/A.

Previously:

• Week 1 Game Scripts
• Week 2
• Week 3
• Week 4
• Week 5
• Week 6

Your most remarkable Game Scripts-related stat of the week comes from the Saints. If you knew that New Orleans trailed 17-7 after the third quarter, and that Drew Brees was healthy (and effective!), what would you expect the Saints pass ratio to be?

65% 75%? 80%?

Try 44%! New Orleans finished the day with 39 rushes against just 31 pass plays, a 44% pass ratio. How remarkable is that?

The last time the Saints trailed by 7 or more points entering the third quarter, and ran on 55% or more of their plays… was in 1983! In fact, when trailing by 7 or more through three quarters, the Saints have never rushed on 50% or more of their plays in the Drew Brees era! And this was a 10-point third quarter deficit!

In fact, since 2000, there have only been three games where New Orleans trailed by a touchdown entering the 4th quarter and finished the game with more rushes than pass plays:

• A meaningless week 17 game in the 2009 season, when Brees did not dress.

• A 2005 game against the Bears

• This 2004 game against Tampa Bay.

New Orleans finished with 12 passing first downs and 12 rushing first downs, the sort of balanced attack that has rarely been seen for the Saints. Even the ground game was balanced: Alvin Kamara had 4 rushing first downs, Brees had 3 (all on 3rd- or 4th-and-1), Taysom Hill had 3, and Mark Ingram had two.  Remarkably, New Orleans had four 4th-down conversions, and all four game on the ground.
[continue reading…]

This was not a good week to be a rookie quarterback. Josh Allen was on the sidelines due to an injured elbow, and he may have had the best week of any of the top rookie passers.

Josh Rosen was a disaster on Thursday night, as he threw two picks six as his Cardinals were destroyed, 45-10. Even ignoring that two of his three interceptions were returned for touchdowns, Rosen finished with a 0.87 ANY/A average, the worst of any quarterback in week seven.

Sam Darnold was not much better. The Jets had six drives in the second quarter, and picked up zero first downs. For the game, Darnold finished with 45 dropbacks and threw for just 7 first downs, and his 16% passing first down rate was the worst of any quarterback in week seven.

Baker Mayfield was only slightly better: the Browns seven first half drives ended in six straight punts and a turnover on downs when Cleveland took possession at the Tampa Bay 19-yard line. Cleveland had 6 three-and-outs, and the Browns picked up a first down on just 21% of all passing plays: third-worst of the week, ahead of only the Cardinals and Jets.

The top passers this week were Rivers and Mahomes, who have separated themselves as the top two quarterbacks in the AFC this season. The full week 7 passing stats below. On average this week, teams completed 22.1 of 34.0 passes, for 247.4 gross passing yards, with a 93.4 passer rating. [continue reading…]

Four teams saw their SRS ratings drop by at least 4 points from their rating one week ago: Ohio State, Oregon State, Minnesota, and Tulane. In that group, Ohio State was the notable loss, of course: Purdue was 38th in the SRS — so better than people may have realized — but still had a rating 14 points lower than that of the Buckeye. Yet in week 8, Purdue blew out Ohio State 49-20, producing one of the three best performances of the season (joining ALabama’s 62-7 blowout against Ole Miss and LSU’s 36-16 win over Georgia).

Purdue was joined by Virginia, North Carolina, California, and Nebraska as teams that saw their SRS rise by at least 4 points. Below are the full SRS ratings through 8 weeks. As always thanks to Dr. Peter R. Wolfe for providing the weekly game logs. [continue reading…]

Todd Gurley is now 50 games into his NFL career, and he is only getting better. He is averaging an impressive 109.4 yards from scrimmage per game, but he’s been noticeably better since Sean McVay arrived.

As a rookie in 2015, Gurley averaged about 100 yards from scrimmage per game but faded down the stretch, topping the century mark just once in his last seven games. In 2016, in an ugly Jeff Fisher offense, Gurley was extremely consistent, but not in a good way: he did not pick up 109 yards in a single game that season. But since McVay arrived, Gurley has topped his current career average in 16 of 21 games.

The graph below shows the yards from scrimmage gained by Gurley in every regular season game of his career, color-coded by year. [continue reading…]

Which teams are throwing the most to one receiver? Which teams are spreading it around the most? A good way to answer that is by using a Concentration Index. Let’s use Andrew Luck and the Indianapolis Colts as an example. Eric Ebron is the team’s leading receiver with 326 receiving yards, which represents 18.2% of the team’s total 1,796 receiving yards. To calculate a team’s concentration index in receiving yards, you need to square the percentage of receiving yards by each player, and sum the results. For example, 18.2% squared is 3.3%; do that for every player on Indianapolis and the total is 12.6%.

As it turns out, that’s the most diverse passing attack in the NFL. The most concentrated passing attack? That’s in Detroit:

The table below shows the concentration index for each team through six weeks, along with each team’s passing efficiency (as measured by ANY/A). Here’s how you read the top line. Detroit has the highest concentration index. The Lions have 1,404 receiving yards and have averaged 6.17 ANY/A, which ranks 20th in the NFL. Tate is the team’s leading receiver with 31% of the Lions receiving yards, and the team has a concentration index of 23%. [continue reading…]

Below is a scatter plot showing the Game Scripts (on the X-Axis) and pass ratio (on the Y-Axis) of every game so far this season. I have also provided a best-fit line:

[continue reading…]

Previously:

• Week 1 Game Scripts
• Week 2
• Week 3
• Week 4
• Week 5

Entering week 6, the Cowboys had Game Scripts of -6.0 against the Panthers, +10.1 against the Giants, -7.8 against the Seahawks, +1.6 against Detroit, and +0.2 against the Texans.

Entering week 6, the Jaguars had Game Scripts of +3.6 against the Giants, +12.2 against New England, -1.5 against Tennessee, +8.2 against the Jets, and -13.7 against the Chiefs.

So color me surprised to see Jacksonville get throttled on Sunday against the Cowboys in Dallas, losing 40-7 and finishing with a -17.3 Game Script. It wasn’t quite a Vikings against the Bills level upset, but Jacksonville was a 3-point favorite; it is pretty unusual for a 3-point dog to lead by 21+ points at halftime and win by 30+ points; in fact, the last time it happened, the Jaguars were on the other side of things.

The full week 6 Game Scripts are below: [continue reading…]

In week 1, the average over/under across 16 games was 44.8. In 9 of those games, the over hit.

Week 2 was when the passing explosion began; in 8 games that week, the teams combined for 48+ points, and in the other 8, teams failed to top 41 points. Nearly every over/under is between 41 and 48 points, so the O/U didn’t matter much this week: in the 8 games that produced 48+ points, the over hit, and in the other 8 games, the under hit.

But that doesn’t mean Vegas wasn’t aware of the passing explosion. The average over/under in week 2 was 45.0, but it jumped to 46.1 in week 3. Once again, we had a 41/48 split: no team scored 42, 43, 44, 45, 46, or 47 points in week 3. So the over/under didn’t matter: 8 games had 48+ points scored and went over, and 8 games had 41 or fewer points and went under. But Vegas had already begun raising the line.

How much? Week 4 had an average over/under of 46.7! And yet the over was still the good bet, as 9 of 15 games went over. This despite the rare event of two games producing 50+ points and still going under! In week 4, 10 out of 15 games combined for 49 or more points, in the greatest passing week in NFL history.

So in week 5, the over/under average jumped to 47.3…. and the over still hit more often than not! There were 8 games with 46+ points scored (7 had 50+ points scored), and the over hit in all of them. The other 7 games all went under, including two games with 44 points scored.

And last week the over/under average was 46.8, perhaps artificially lower because the Saints and Lions (who had 50+ point over/under lines in week 5) were on bye. And yet in week 6, 9 out of 15 games went over.

The table below shows the weekly over/under results this year. The Under has never been a winner, and the books seem to be doing all they can to make over/unders high enough to get people to bet the under: [continue reading…]

The Buffalo Bills have the worst passing attack in the NFL, and it is not particularly close. The NFL is averaging 6.5 Adjusted Net Yards per Attempt, the 30th-best passing offense is averaging 4.5 ANY/A, and the 31st-best passing attack is averaging 4.1 ANY/A.  The Bills? They are averaging 2.0 ANY/A and have reached the cellar by being bad at every facet of the passing game.  The Bills rank last in the league in completion percentage, passing yards, passing touchdowns, passing touchdown rate, and yards per attempt. The Bills also have the worst sack rate *and* the worst interception rate in the NFL.

In 1941, the league average completion percentage was 44.3%; the 2018 Bills are completing passes at a 50.6% clip.

In 1941, the average NFL team gained 122 passing yards per game; the 2018 Bills are gaining 123 passing yards per game.

In 1941, NFL teams threw an interception on 10% of passes; the 2018 Bills have thrown an interception on 7.3% of passes.

But if you pick any year more recent than 1941, you might think Buffalo was a below-average passing team.  The league average completion percentage has been over 50.6% in every season since 1961. When it comes to interceptions, 1975 and 1971 are the only post-merger seasons where the league average more than 1.5 interceptions per game.  And in passing yards per game?  It’s been higher than Buffalo’s current average in every season beginning in 1942.

If you want to use yards per attempt, you have to go back to 1935 to find a season where Buffalo’s current 5.39 Y/A average would be above average.   The last time the league average ANY/A was below 2.0 was in 1938, and the NFL has never had a NY/A average as poor as the 2018 Bills.

So yes, Buffalo’s passing performance this season has been otherworldly.  And it got worse in week 6, as Josh Allen and Nathan Peterman combined to gain just 129 passing yards on 31 dropbacks, while throwing two interceptions.  And with 90 seconds left in the 4th quarter of a tie game, Peterson threw a brutal pick six that swung the game to Houston.

Below are the week 6 passing stats, with some familiar names at the top.

And in addition, here are the passing stats to date. I am using 15 dropbacks multiplied by the number of weeks in the season to determine whether a passer has qualified. The table has 58 rows, but for ease of reading, I have defaulted to only showing the top 10.

On Sunday against the Panthers, Washington quarterback Alex Smith threw a touchdown pass to tight end Vernon Davis. Over 12 years ago, the two connected for the first touchdown reception of Davis’s career, back when both players were with the 49ers.

How rare is it for a quarterback to throw a touchdown pass to a player, and then throw a touchdown pass 12 years later to the same player on a new team? Well, it’s never happened before. The closest was with Randall Cunningham and Cris Carter; the duo connected for a touchdown pass in November 1987 with the Eagles, and last connected for a touchdown pass in September 1999 with the Vikings, a span of nearly 12 years.

The table below shows, through 2017, all QB/Receiver pairings where the duo connected for touchdown passes on different teams.  I have used “different teams” liberally here (e.g., including the old Browns and Ravens as different teams), but you can consider them the same team if you like.

Stabler and Casper hold the record for combined touchdowns on two teams, at 45, although most of those were in Oakland. Bledsoe/Glenn, Cunningham/Carter, Cutler/Marshall, and — if you want to include them, Testaverde/Jackson and Testaverde/Alexander — are the only pairs to connect for double digit touchdowns on two different teams.

It was an eventful week of college football, with 6 teams ranking in the top 16 falling in week 7:

• #2 Georgia was blown out by LSU, 36-16.

• #6 West Virginia lost to giant killer Iowa State; over the last 53 weeks, the Cyclones have beaten three Big 12 teams that ranked in the top 6: Oklahoma and TCU in 2017, and the Mountaineers 30-14 on Saturday.

• #7 Washington, which had a loss to Auburn in the season opener (and which only looks uglier in retrospect, as the Tigers have gone 2-3 against the rest of the FBS since then), had a perhaps fatal loss to Oregon on Saturday afternoon. The Huskies lost in overtime, 30-27, in Eugene.

• #8 Penn State, a week after a narrow home loss to Ohio State, was upset 21-17 at home against Michigan State.

• #15 Wisconsin was obliterated by Michigan in Ann Arbor, 38-13. The 25-point differential was the worst by the Badgers in a true road game in over a decade.

• #16 Miami lost to Virginia, 16-13.

In addition, Notre Dame narrowly beat Pittsburgh (19-14), Central Florida barely escaped Memphis (31-30), and Texas, Texas Tech, and Texas A&M all escaped with one-score wins. [continue reading…]

Doug Williams was a better passer with the Bucs than he gets credit for, particularly in 1979. He easily led the NFL in sack rate, with an absurdly low 1.7%, when the league average was 7.5%. He also averaged 14.7 yards per completion, the 4th-best rate in the league. But Williams completed only 41.8% of his passes, which was bad even for a 2nd-year quarterback in 1979.

The Bucs ranked dead last in completion percentage (42.2%), with the 27th-best team in completion percentage over five percentage points higher at 47.4%. But as I noted last week, passing first down percentage is a more important metric than completion percentage. And because Williams had a remarkable sack rate and was a downfield thrower, the Bucs picked up a first down on 30.7% of all passing plays. In 1979, that ranked a respectable 12th in the 28-team league.

The Bucs were 11.9% below average in completion percentage but 1.2% above average in passing first down percentage. That’s a difference of 13.1%, which is the largest difference between any team’s completion percentage and passing first down rate since 1970. The full results below: [continue reading…]

Let’s compare two groups of 2018 receivers.

Group A

Adam Thielen
Odell Beckham
DeAndre Hopkins
Zach Ertz
Julio Jones

Group B

Brandin Cooks
George Kittle
Mike Evans
Tyreek Hill
Cooper Kupp

Which group of receivers would you rather have going forward? I think you know the answer to that question.

The receivers in Group A are the top 5 receivers this season as measured by one simple metric: yards multiplied by targets.

The receivers in Group B are the top 5 receivers this season as measured by another simple metric: yards divided by targets, based on a the players who rank in the top 50 this season in targets.

Here’s how to read the table below. Brandin Cooks has played 5 games this year and has seen 33 targets and gained 452 receiving yards. If you multiply his yards by his targets, he has 14,916, and his Yards x Target rank is 28th. On the other hand, he has averaged 13.7 Yards per Target, which ranks 1st in the league through five weeks. [continue reading…]

Previously:

• Week 1 Game Scripts
• Week 2
• Week 3
• Week 4

The Buffalo Bills got rid of Tyrod Taylor, spent a first round pick on a quarterback, and are somehow more run-heavy! In 2018, the Bills had the strongest rush identity in the NFL, in large part because of Taylor. Yet on Sunday, Buffalo produced the most run-heavy game of the year, rushing more than twice as often as passing in a 13-12 win against the Titans. The only time the Bills did that last year was in the snow storm against the Colts.  In week 5, Josh Allen had 20 dropbacks and gained just 78 yards; Allen rushed 4 times for 19 yards and a touchdown, LeSean McCoy had 24 carries for 85 yards, and Chris Ivory had 14 carries for 43 yards.

The most pass-happy team of the week is also the most pass-happy team from 2017: the New England Patriots. Against the Colts, the Patriots led 7-0 after 6 minutes, 14-0 after 18 minutes, 24-3 at halftime, and 38-17 midway through the 4th quarter. And yet Tom Brady finished the day with 44 pass attempts, while Patriots running backs had just 20 carries. This game was the 5th time in the last 10 years that the Patriots had 44+ pass attempts despite leading by 14+ at both halftime and the end of the game; over that same period, the rest of the NFL had done it just four times. That’s why New England — with the 4th-best Game Script of the week and the highest Pass Ratio of any team with a positive Game Script in week 5 — was the most pass-happy team of the week.

The full week 5 Game Scripts are below: [continue reading…]

It was not a good day for Blake Bortles and the Jaguars in week 5. He finished the game with a 3.65 ANY/A, going 33/61 for 430 yards, with 1 TD, 4 interceptions, and 5 sacks for 29 yards. Even those numbers overstate things: At halftime, he was averaging 0.92 ANY/A, and after 40 minutes, he was averaging just 1.86 ANY/A. With some good passing numbers when the game was out of hand, Bortles bumped his ANY/A to 4.40, before it dropped to 3.65 when he threw a meaningless interception on the final play of the game.

One interesting way to show how the game progressed is by using a bubble chart, where the size of each bubble represents the scoring differential, the X-Axis represents time elapsed in the game, and the Y-Axis shows adjusted net yards. Each bubble represents one attempt. As you can see, a lot of Bortles’ production came when the game was out of hand (and win probability would be an even better metric to use than scoring differential here, as the “larger” bubbles late in the fourth quarter were also when the game was out of hand due to time remaining as much as scoring differential):

[continue reading…]

Jets running back Isaiah Crowell had a remarkable game on Sunday against the Broncos. He made the most of his 15 carries, producing a 77-yard rushing touchdown, along with carries of 54, 36 and 15 yards. He finished the day with 219 rushing yards, setting a new Jets franchise rushing record in the process.

Do you know what else he did? He averaged 14.33 yards per carry, the highest single-game YPC average by any player in history with at least 15 carries. That made me wonder: what are the highest single-game yards per carry averages at all carry levels?

Joey Galloway holds the record for yards per carry in a game, at 86.0. He did that on one carry, as you could probably guess (and what a carry it was). Up the minimum threshold to 2 or more carries, and Brian Mitchell is your record-holder: he averaged 52.2 yards per carry on 2 carries for 105 yards. At 3+ carries, Cordarrelle Patterson was the record-holder, with a 3-102-34.3 performance … until Browns running back Nick Chubb broke that mark in week 4 of this season against the Raiders with 3 carries for 105 yards.  At 4+ carries, Latavius Murray holds the record with a 28 yards per carry average.  And so on. [continue reading…]

Completion percentages continue to skyrocket, and part of the reason for the increase is the rise of low value completions. As a result, the best way to measure the consistency of a passing game isn’t with completion percentage, but with passing first down percentage. That’s defined as simply:

Passing First Downs / (Pass Attempts + Sacks)

As noted last week, the Giants and Texans provided good real world examples for 2018. Through four weeks, Houston ranked 25th in completion percentage but 8th in passing first down percentage. Meanwhile, the Giants ranked 2nd in completion percentage but 22nd in passing first down percentage. My proposition is that passing first down percentage is a better reflection of the passing game and a more meaningful statistic than completion percentage; and that’s why, through four weeks, Houston ranked 15th in points/game, while the Giants ranked 29th.

But does that argument hold up over time? How do we *know* that Passing 1st Down Rate is more important than Completion Percentage? Here’s a simply study I did.

1) Calculate the completion percentage relative to league average for every team since 1970.

2) Calculate the passing first down percentage relative to league average for every team since 1970.

3) Identify teams that ranked two percentage points better than league average in one metric and two percentage points worse than league average in the other metric.

4) Compare those teams. [continue reading…]

It’s time, once again, to release the first edition of the college football Simple Rating System ratings. While it’s too early to put too much weight on these ratings, they help to at least begin framing the discussion of which are the most impressive teams in college football. As a reminder, here is the methodology:

1) For each game not played at a neutral site, 3 points are given to the road team. After that adjustment, all wins and losses of between 7 and 24 points are recorded exactly as such. This means that a 24-10 road win goes down as +17 for the road team, -17 for the home team.

2) With one exception, wins of 7 or fewer points are scored as 7-point wins and losses of 7 or fewer points are scored as 7 point losses. So a 4-point home win goes down as +7 (and not a 1) and a 1-point home loss is a -7 (and not a -4). The one exception is that road losses of 3 or fewer (and home wins of 3 or fewer) are graded as ties. So a 21-20 home victory goes down as a 0 for both teams.

3) Wins/Losses of more than 24 points are scored as the average between the actual number and 24. This is to avoid giving undue credit to teams that run up the score. So a 75-point home win goes down as a 48-point win.

Once we have a rating for each team in each game, we then adjust each result for strength of schedule. This is an iterative process, where we adjust the ratings hundreds of times (to adjust for SOS, you have to adjust for the SOS of each opponent, and the SOS of each opponent’s opponent, and so on.) in Excel. Then we produce final ratings, where the SRS rating is the sum of the Margin of Victory and Strength of Schedule in every week. [continue reading…]

Eli Manning ranks 2nd in the NFL in completion rate with a remarkable 74.2% average. Houston’s Deshaun Watson, on the other hand, ranks 27th out of 34 qualifying passers with a 62.2% completion percentage. But that’s a bit misleading, because as I’ve written about before, first down percentage is a much more important metric than completion percentage.

For Watson, 62 of his 92 completed passes have gone for a first down, meaning just 33% of his completed passes didn’t pick up a first down. For Manning, just 53 of his 112 completions have gone for a first down, which means a whopping 53% of his completions didn’t pick up a first down.

Completion percentage is supposed to measure how consistent a quarterback is at making positive plays. A better formula than (completed passes/pass attempts) is (first downs gained via the pass / [pass attempts + sacks] ). In other words, the numerator should be passing first downs, not completed passes, and the denominator should include sacks. [continue reading…]

Previously:

• Week 1 Game Scripts
• Week 2
• Week 3

Minnesota, Indianapolis, and Pittsburgh all have franchise quarterbacks. Kirk Cousins, Andrew Luck, and Ben Roethlisberger are three of the nine quarterbacks with salary cap hits in excess of $23,000,000 in 2018. Perhaps, then, we shouldn’t be too surprised to see those three teams all choose to pass on over 75% of plays in week 4.

And it’s not like week 4 was an anomaly: these three teams lead the league in passing attempts (including sacks) per game so far in 2018. The Colts are averaging a whopping 51.2 passing plays per game (boosted by another pass-happy performance last night in week 5), the Vikings 50.5 passing plays per game, and the Steelers 48.75. The fourth-place team in this metric is Baltimore, and Joe Flacco is another one of the $23M salary cap hit club.

I noted in the offseason that you can break down the league’s 32 teams into a few simple buckets:

• Half of the league (including Indianapolis, Minnesota, Pittsburgh, and Baltimore) had a quarterback under a “franchise money” type contract

• 12 of the league’s teams are using quarterbacks on rookie contracts, including the Jets, Bills, Cardinals, and Browns who are now starting 2018 first round picks.

• 4 teams, all in the AFC, are stuck with veteran quarterbacks who don’t appear to be franchise quarterbacks. Of this group, Ryan Tannehill, Blake Bortles, and Andy Dalton are all playing well: their teams are all 3-1 and they have slightly above-average passing stats, which would count as exceeding expectations. The fourth player is Case Keenum, and you can make the case that the Broncos are the only team that doesn’t have its 2019 quarterback already on its roster. In other words, 16 teams have a franchise quarterback (or a quarterback with a franchise quarterback salary), 12 teams have a quarterback on a rookie contract and at least some optimism for a long-term future (with perhaps Tampa Bay and Dallas being the sole exceptions), and 3 teams have a middling option who is actually playing pretty well.

The full week 4 Game Scripts data below: [continue reading…]

The table below shows the main statistics for the top running backs from week 5 of the 2017 season through week 4 of the 2018 season:

[continue reading…]

Through four weeks, teams have rushed 3,181 times. They have thrown 4,587 passes and been sacked 320 times, which means teams have rushed on 3,181 out of 8,088 plays. That’s a run ratio of just 39.3%, which would set a new record. This makes some sense, of course: given that this has been the most efficient passing season in NFL history, you would think teams are shifting more to the pass than the run. The graph below shows the percentage of running plays for each season in NFL history since 1950; as you can see, it’s not 1956 anymore.

[continue reading…]

As a rookie, Jared Goff threw 5 touchdowns on 205 pass attempts. In week 4, he threw 5 touchdowns on 33 pass attempts.
As, Mitch Trubisky threw 7 touchdowns on 330 pass attempts. In week 4, he threw 6 touchdowns on 26 attempts.

Both joined the exclusive 17/17 club of 17 Adjusted Yards per Attempt on 17 passes. ^[1]While I prefer using ANY/A to AY/A, the PFR Game Finder only has AY/A as a default search option, hence my use of AY/A there., making a remarkable three such performances in September 2018.

Meanwhile, there is Josh Allen, who became just the 19th quarterback since 2008 to finish a game with a negative ANY/A on at least 30 pass attempts.

And then there’s Matt Ryan, who for the second straight week found himself on the losing side of a passing shootout. He became just the 8th quarterback since the merger to average 11.0 ANY/A or better, throw for 400 passing yards, and lose.  The Falcons join the 1966 Giants as the only teams to score 36+ points in consecutive regular season games… and lose both games.

Note that it was a terrible week for AFC East quarterbacks not named Tom Brady, and actually not even a good week for Brady (because week 4 was absurd, Brady’s 6.97 ANY/A average was actually below-average). Below are the passing stats for week 4: [continue reading…]