The NFL Draft is a good way to measure how the league values particular positions. Last year, it seemed as though the running game was back in vogue: after quarterback, running back was perhaps the most highly valued position in the 2018 Draft, and a lot of draft capital was used on non-pass rushing linebackers.

Which positions did NFL teams focus on in the 2019 NFL Draft? We can use the Football Perspective Draft Value Chart to answer that question pretty easily for the first 224 picks (all picks after that have been excluded, since they have a draft value of zero).  This was a draft for interior defensive linemen and edge rushers, but even more notable was the absence of draft capital spent on interior offensive linemen.

Consider that 3 out of every 22 players on a football field, at all times, is a guard or center.  All else being equal, you would expect 14% of all draft picks and all draft capital to be spent on interior offensive linemen.  That means 1 out of every 7.3 draft picks should be an offensive guard or center, but in the first 73 draft picks, there were just 5 interior offensive linemen selected.  Just 8% of the first 100 picks were guards or centers, and overall, only 8.4% of all draft capital was spent on guards and centers. Here’s the full list of draftees. [continue reading…]

Josh Rosen was terrible as a rookie, and his statistical performance is not debatable. He was the worst passer in the NFL by ANY/A and also the worst quarterback according to ESPN’s Total QBR metric. Statistically, Rosen was awful, and there’s little more to add.

Now, does he have excuses? Absolutely. As Bill Barnwell wrote, Rosen faced some pretty rough circumstances last year, particularly when it came to his offensive line.

Other rookie quarterbacks have started their careers behind porous offensive lines, of course, but not to this extreme. By the end of last season, the Cardinals had lost all five of their starting linemen to injuries. During the second half of 2018, the five linemen who took snaps most frequently protecting Rosen included a pair of rookies (third-rounder Mason Cole and seventh-rounder Korey Cunningham), a player signed off Minnesota’s practice squad (Colby Gossett) and a pair of veterans who were cut by teams and almost immediately stepped off the street and into Arizona’s starting lineup (Oday Aboushi and Joe Barksdale). It’s one thing to have a relatively untalented line, but the Cardinals were starting guys who barely knew the playbook at times.

Now, given that we know Rosen was terrible as a rookie, how likely is he to still turn out to be a good quarterback?

The Eight That Turned It Around

Since 1967, there have been 8 quarterbacks who were drafted in the first round, played more than a handful of snaps as a rookie, were very bad, and then one day became a good quarterback.

They are: Terry Bradshaw, Troy Aikman, Eli Manning, Matthew Stafford, Jared Goff, Donovan McNabb, Bert Jones, and Alex Smith.

All 8 quarterbacks had era-adjusted passer ratings of under 50 as a rookie; they averaged a collective 39.1 era-adjusted passer rating, slightly worse than what Rosen (40.5) averaged in 2018. All 8 quarterbacks finished at least 2.00 ANY/A worse than average; collectively, they were 3.07 ANY/A worse than league average, slightly worse than Rosen’s -2.79 RANY/A in 2018. [continue reading…]

The Miami Dolphins have been looking for the next Dan Marino since the day he retired. In that pursuit, the franchise has a habit of thinking all quarterback woes can be solved with a second round pick. And while that has been flawed thinking so far, it’s hard to find any criticism with what GM Chris Grier did last night.

In 2003, Miami traded a 2004 second round pick to Philadelphia for A.J. Feeley, who lasted just one year with the Dolphins and went 3-5 as a starter.

In 2007, Miami used a second round pick on John Beck from BYU, who also lasted only one year in South Beach and went 0-4 as a starter.

In 2008, Miami used a second round pick on Michigan’s Chad Henne. The Wolverine product stuck in Miami for four years and went 13-18 as a starter.

In 2009, Miami — fresh off of its success with the Wildcat offense — used a second round pick with West Virginia superstar running quarterback Pat White. He never worked out in the NFL at any position, lasting only one year in the NFL and finishing with under 100 total career yards. [continue reading…]

In March 2013, the Steelers let Mike Wallace leave in free agency for Miami, which would give Pittsburgh a compensatory draft pick at the end of the 3rd round of the 2014 Draft. That seemed to encourage some bad drafting behavior. A couple of months later, during the 2013 Draft, Pittsburgh traded its own 2014 3rd round pick to Cleveland for the 111th pick in the Draft to grab safety Shamarko Thomas. Pittsburgh traded its own 3rd round pick next year for the Browns 4th round pick that year because the team needed Thomas.

Well, Thomas ended up busting, and the Steelers used the compensatory pick on Dri Archer, another bust. Pittsburgh wound up trading the 83rd pick in the 2014 Draft for the 111th pick in the 2013 Draft, which is rarely a smart move. For Cleveland, the team used that 83rd pick to move up in the 1st round and get Johnny Manziel, so the story doesn’t have a happy ending there, either. And the actual 83rd pick turned out to be Louis Nix III, which means everyone tied to these transactions (including the Dolphins) with a pair of snake eyes.

Why that context when talking about the 2019 Draft? Well, the Steelers are going to get a 3rd round compensatory pick in the 2020 Draft after letting Le’Veon Bell leave for the Jets in March 2019. That seemed to encourage Pittsburgh to freely give away its own 2020 3rd round selection…

1)
Steelers trade: 20th overall, 52nd overall (2nd round), 2020 3rd round pick
Broncos trade: 10th overall

Pittsburgh traded up Michigan inside linebacker Devin Bush, while Denver would take Iowa TE Noah Fant with the 20th pick. How much did Pittsburgh give up?  Let’s assume the 2020 3rd round pick is valued at the 96th pick in this Draft, which is the last pick in the 3rd round (so this puts in some discount for losing a future pick rather than a current one).  By my draft value chart, it means that Pittsburgh paid 153 cents on the dollar, sending 30.4 points of AV for 19.9 points of AV.  By the traditional draft value chart, Pittsburgh gave up 104 cents on the dollar, a more reasonable sum.

The reality is that the Steelers likely viewed the 2020 3rd round pick as free money, courtesy of losing Bell.  The price to move up was not out of the ordinary, but the ordinary transaction is a bad one for the team moving up.

2)

Packers trade: 30th overall, 114th overall (4th round), and 118th overall (4th round)
Seahawks trade: 21st overall

By my chart, Seattle did a solid job here, forcing the Packers to pay 141 cents on the dollar.  By the traditional chart, the Seahawks were basically looking to get whatever they could to trade down, as Green Bay paid only 93 cents on the dollar!  It is rare for the team trading up to pay so little, so this is a strong sign that it was a buyer’s market for moving up (is that a reflection of teams getting smarter or the talent distribution in this particular draft?)

Green Bay moved up for Maryland safety Darnell Savage, while Seattle would trade this pick again an hour later. This was probably a good trade for both sides, as the Packers were able to get the player they wanted while only giving up a pair of 4th round picks, and the Seahawks added a few more swings at the plate later on.

3)

Eagles trade: 25th overall, 127th overall (4th round), 197th overall (6th round)
Ravens trade: 22nd overall

Philadelphia moved up to draft Washington State tackle Andre Dillard, the second offensive tackle off the board. This was likely a necessity, as Philadelphia jumped Houston with the 23rd pick. On my chart, the Eagles gave up 127 cents on the dollar, while this was a near perfect match on the traditional chart (99.7 cents on the dollar). This is a trade that makes a lot of sense: Philadelphia didn’t have to overpay by much to make a trade that matched their needs perfectly.

4)

Redskins trade: 46th overall (2nd round), 2020 2nd round pick
Colts trade: 26th overall

Washington moved up to get Mississippi State edge rusher Montez Sweat, who was the 40-yard dash superstar this year. If we value Washington’s 2nd round pick as equal to the 64th pick in this year’s draft, the Redskins gave up 132 cents on the dollar for Sweat, who perhaps fell in the first round due to a misdiagnosed heart condition. By the traditional chart, this would be sending 101 cents on the dollar.

So this is a great trade for Washington if you view Sweat as a top-20 player and the 2020 2nd round pick as the 64th pick in this year’s draft. The concern, of course, is that the Redskins might be very bad next year. If this turns out to be, say, the 36th pick in the draft, and you apply no time value discount, the traditional chart would say Washington paid 140 cents on the dollar while my chart would say Washington paid 158 cents on the dollar.

It is always risky for a bad team to trade a future pick, but Sweat at least profiles as the sort of player you can understand trading up for, especially since the premium wasn’t that significant.

For Indianapolis? This was another smart way to add value.

5)
Giants trade: 37th overall (2nd round), 132nd overall (4th round), 142nd overall (5th round)
Seahawks trade: 30th overall

The Giants picked up the 132nd pick in the Odell Beckham trade and the 142nd pick in the Damon Harrison trade. Both of those trades were hard to defend on the merits, but now New York has muddied the ultimate analysis for those trades! By sending those two picks, along with the Giants early 2nd round pick, New York was able to get Georgia cornerback Deandre Baker.

By my chart, the Giants paid 140 cents on the dollar, but on the traditional chart, it was a trade of just 98 cents on the dollar.  Again, teams often (mistakenly) view extra picks as free picks: here, at least for Giants fans, the price to move up was not very high.

6) Falcons trade: 45th overall (2nd round), 79th overall (3rd round)
Rams trade: 31st overall, 203rd overall (6th round)

By my chart, Atlanta sent 127 cents on the dollar to move up; by the traditional chart, Atlanta sent 106 cents on the dollar. The Falcons grabbed Washington offensive tackle Kaleb McGary, giving up two solid picks in the process. Still, this was not an egregious overpay, which is something we have seen in past years as teams try to trade back into the end of the first round. Last year, the Ravens moved up to the 32nd pick for Lamar Jackson, and in doing so, gave up the 52nd pick and a 2019 2nd rounder and also moved down 7 slots in the 4th round.  Atlanta at least paid a cheaper price than that, but this was still a good way to grab some extra value by Los Angeles.

Kyler Murray is going to be drafted tonight, likely very early in the 2019 Draft. At 5’10, he is really short for a first round quarterback: like really short.

Davey O’Brien (yes, that Davey O’Brien), standing 5’7, is the shortest quarterback ever selected in the first round. He was the 4th overall pick in the 1939 Draft. There have been no quarterbacks taken in the first round who were 5’8 or 5’9, and only five were standing exactly 5-foot-10: Ted Marchibroda (yes, that Ted Marchibroda), Travis Tidwell, Ernie Case, Boley Dancewicz, and Frankie Albert. All of these players were drafted in 1950 or earlier.

The modern draft era begins in 1967, when the AFL and NFL joined forces for the common draft. Since then, the shortest quarterbacks drafted are 6’0 Michael Vick and Johnny Manziel. And only 6 quarterbacks who were 6’1 went in the first round: Baker Mayfield, Jim McMahon, Clint Longley, Bob Griese, Rex Grossman, and Cade McNown. There have been no quarterbacks who were 5’11 and drafted in the first round, and tonight, Murray will become the first 5’10 quarterback drafted in the first round since 1950.

The graph below shows the distribution, by height, of quarterbacks selected in the first round since 1967. [continue reading…]

Yesterday, I took another look at the draft value chart and the appropriate values we should assign to each draft pick. One conclusion was that the value of all draft picks has increased, as more AV is going to players on rookie contracts.

Today, I want to specifically examine the claim that since the NFL instituted the rookie wage scale as part of the 2011 CBA, teams are giving more playing time and production to players on rookie contracts. For all graphs today, I will be separating players into two categories: players who are in their first 4 seasons will be graphed in red, and veterans in their 5th seasons or later will be graphed in black. [continue reading…]

It has now been seven years since I created the Football Perspective Draft Value Chart (longtime readers know that at Pro-Football-Reference, I created something similar 11 years ago).

You can see all the values for the Football Perspective Draft Value Chart here and there’s a link on the home page to those same values embedded in the FP Draft Value Calculator. The FP chart was derived from analyzing drafts from 1970 to 2007, and by looking at the marginal Approximate Value produced by players in each of their first five seasons. In short, I gave players credit for all AV produced above 2 points of AV each year. If a player had AV scores of 0, 4, 1, 7, and 10, that translates to 15 marginal points. To create the Draft Value Chart, I simply calculated the average number of marginal points produced by each draft pick from ’70 to ’07 and smoothed out the results.

But now, with 7 more years of data, we can look at the drafts from 2008 through 2014, and see how those players have faced in their first 5 seasons. How has the FP Chart done since then?

When analyzing results over a small period, there will obviously be outliers. One out of every 7 players taken with the 75th pick doesn’t turn into Russell Wilson, one out of every 7 players taken with the 154th pick doesn’t turn into Richard Sherman, and so on. The 92nd pick, with T.Y. Hilton, Trai Turner, Cliff Avril, Joe Barksdale, Shawn Lauvao, and Jerraud Powers was extremely good at getting consistent starters (Stedman Bailey being the exception).

And even the number one picks have been very good over this stint: Jake Long, Matthew Stafford, Sam Bradford, Cam Newton, Andrew Luck, Eric Fisher, and Jadeveon Clowney is a pretty good stretch; JaMarcus Russell just missed the cut. Here are the average results from these 7 drafts in terms of marginal AV produced by these players in their first five years:

Now, how does that compare to the Football Perspective Drat Values? And what would the results from ’08 to ’14 look like if we smoothed the results? I’m glad you asked! I’ve reprinted the results below, but added the FP Draft Value Chart line in red, and a smoothed, best-fit line of the ’08 to ’14 results in blue.

On first glance, I have two immediate takeaways.

One, the FP Draft Value Chart looks pretty good: it’s directionally correct, and holds up over time and a small sample.  It doesn’t have the super steep decline of the JJ Draft Value Chart, which has the 3rd pick worth twice as much as the 11th pick, and the 11th pick worth twice as much as 35th pick.

The other takeaway is that almost all the picks are more valuable, and that is not a function of small sample size: it’s a function of the rookie wage scale. Players taken in the 3rd, 4th, and 5th rounds are now expected to contribute, because the bar has been lower.  If a 4th round pick costs a fraction of what a veteran costs, teams will be more willing to give that player a chance to produce even if he isn’t very good.  This artificially inflates AV, since AV is tied to metrics like starts and games played.  Teams are giving more starts and snaps to players on rookie contracts because of the rookie salary cap, and that leads to more AV — and distorts the draft value chart a bit.

My overall suspicion is the success of first overall picks from ’08 to ’14 distorts the steepness of the graph — it doesn’t appear like the 2nd and 3rd picks are doing any better than they used to — and that is likely due to small sample size (although the first overall picks since ’14 also seem pretty good so far, too!).   And on a relative basis, I am not sure much has changed in the draft value chart world.  But I do think it’s fair to acknowledge that draft picks are more valuable than they used to be and my Draft Value Chart implies, and that’s worth thinking about when teams trade multiple picks for one pick.

As always, please leave your thoughts in the comments.

In 2018, there were three teams that made huge strides in their passing game, and all three involved turnover at quarterback.

In 2017, the Colts averaged 5.06 ANY/A, 25th-best in the NFL, with Andrew Luck missing the season due to injury. Last year, with a healthy Luck, Indianapolis finished 10th in ANY/A with a 6.90 average.

In 2017, the Packers finished 30th in ANY/A with a 4.66 average as Aaron Rodgers missed over half of the season due to injury; last year, with Rodgers back, Green Bay averaged 6.58 ANY/A, 13th-best in the league.

But the biggest jump from ’17 to ’18 came from the Cleveland Browns. After adding Baker Mayfield, the Browns jumped from last in ANY/A at 3.63 to 17th with a 6.25 average.

Yesterday, I looked at the teams that saw the biggest year-over-year declines in passing efficiency. Today, the opposite: which teams made the biggest ANY/A improvement each year? That’s what is shown in the table below. For example, in 2017, the Rams made the biggest jump in ANY/A. The year before, the Rams averaged 3.98 ANY/A, worst in the league, behind mostly Case Keenum (although a rookie Jared Goff was miserable as a starter in 7 games). The next year, Los Angeles averaged 7.47 ANY/A 4th-best in the league, as Goff made an enormous leap.

In 1993, the Bears finished 2nd to last (behind Washington) in ANY/A at 3.35, with Jim Harbaugh struggling at quarterback. In 1994, with Steve Walsh at quarterback, Chicago finished 17th in ANY/A with the biggest improvement (+1.87 ANY/A) in the league at that metric.  And then in 1995, with Erik Kramer , Chicago jumped another 1.98 ANY/A; not only was that the biggest jump from ’94 to ’95, it also made the Bears the top passing team of 1995. In a span of two years, the Bears went from averaging 3.3 ANY/A to 7.2 ANY/A, increased their touchdowns from 7 to 29 and their average completion from 9.9 yards to 12.2 yards, while seeing their interceptions call from 16 to 10 and sacks drop from 48 to 15.

That’s one of two times a team had the biggest ANY/A improvement in back to back years.  The other time involved Dan Marino and the Miami Dolphins; Miami made a huge jump going from not Marino to a rookie Marino in ’83, and then another big jump going from rookie Marino to HOF Marino in ’83.

What stands out to you?

There were 5 teams that experienced a decline of at least 1.00 Adjusted Net Yards per Attempt last season. The Lions, with Matt Stafford, dropped by 1.27 ANY/A, while the Cardinals (-1.32) and Bills (-1.41) saw declines after switching to rookie quarterbacks. The Jaguars experienced the scary half of the Blake Bortles roller coaster last year, as the team’s passing game declined by 1.43 ANY/A.

But it was Washington — who evicted Kirk Cousins — that saw the biggest decline, a drop-off of 1.56 ANY/A. The Redskins problems were compounded by the injury to Alex Smith, as Washington averaged 4.82 ANY/A for the year but Smith averaged 5.81 ANY/A.

It made me wonder: which teams have had the biggest decline in their passing games each year? The table below shows that for each year since the merger.

Some notes:

12 times the biggest dropoff came from the team that ranked 1st in ANY/A the year before, which makes some sense, along with 10 more that ranked 2nd in ANY/A. Three teams went from average to horrible, which is even harder to pull off: the 1988 Patriots, the 1998 Eagles, and the 2006 Raiders.  And a special nod to Mark Rypien and the Redskins, who led the NFL in ANY/A in 1991 at 8.33, had the biggest decline in ANY/A from ’91 to ’92 when the Redskins ranked 11th with 5.31 ANY/A, and then again suffered the biggest decline in ANY/A from ’92 to ’93 when the team averaged 3.18 ANY/A, ranking last in the league.

On Thursday, the NFL schedule was released. After that happens, one thing I like to do is to measure how much rest each team has relative to its opponents each week (although Brian Burke beat me to the punch this year). In past years, some teams got really screwed when it came to extra rest, but that wasn’t the case this year.

The Lions have the most friendly schedule this year when it comes to rest. With the exception of a week 7 game on Sunday following a week 6 game on Monday Night Football, Detroit doesn’t have any games this year when it played a game more recently than its opponent. The Lions don’t play any team coming off of a bye or a Thursday night game, and Detroit gets extra rest following its own bye, following its Thanksgiving game, and potentially two extra days in week 17 (the Lions might play on Saturday in week 16, while Detroit’s week 17 opponent, Green Bay, plays on Monday night in week 16).

On the other hand, you have the Patriots. New England has three consecutive games against opponents coming off of a bye: the Browns have a bye in week 7 before traveling to New England in week 8, the Ravens have a bye that week before hosting the Patriots in week 9, and both the Eagles and Patriots have a week 10 bye before facing off in Philadelphia in week 11. In addition, the Texans have a Thursday night home game in week 12, and 10 days later, host the Patriots.

The table below shows the amount of extra rest each team (and its opponent) has this year. Note that all 5 week 16 games that are possible Saturday games are considered Saturday games. The table is sorted from most favorable to least: e.g., the Lions face opponents with 12 days of fewer rest than them, the Chargers face opponents with 10 fewer days of rest, etc.

The “Opp Byes” column shows how many games each team has against teams coming off of byes. Obviously this totals to 32, but it is not evenly distributed. The Eagles and Patriots each play three teams coming off of byes, while nine teams face zero teams coming off of byes.

As you can imagine, heavier players fare much worse in the 3-cone drill, and taller players have a slight advantage, too. Here was the best-fit formula from the 2019 combine:

Expected 3-Cone Drill = 7.4183 – 0.0287 * Height (Inches) + 0.0081 * Weight (Pounds)

Michigan defensive back David Long, who posted the fastest (but not the best) time in he dominated in the short shuttle, finishing in 6.45 seconds, the fastest time in the drill. Given his dimensions — 71 inches, 196 pounds — he’d be expected to complete the drill in 6.97 seconds. Therefore, Ford finished the drill in 0.52 seconds better than expected, the best adjusted performance in this drill.

Below is a chart showing the expected 3-Cone Drill based on various heights and weights: [continue reading…]

Earlier this week, I wrote about the passing breakdown on short throws to both the left and right side of the field.

What about on deep throws? As it turns out, there’s not much of a difference in terms of either quantity or quality, with one notable exception. Last year, there was a nearly perfect 50/50 split on deep throws between being to the left side of the field and the right side of the field.  And the completion percentages were nearly identical, too, at 40%, as were the yards/attempt averages.  One big difference, though, was in touchdown rates.  And this is actually all consistent with what we found in the 2017 season, too: a 50/50 split between left and right side, the same completion percentage, but a much better TD rate on throws to the right side.

So, I suppose, all else being equal, you want to throw more deep passes to the right side of the field.  What’s interesting, though, is someone like Marcus Mariota, who threw 31 passes to the deep left side but just 11 to the deep right.  His top wideout, Corey Davis, caught 6 of 8 passes to the deep right, and was targeted 9 times on the deep left side.  But unless he was throwing to Davis, Mariota rarely threw to the deep right side of the field. [continue reading…]

Every year, I publish a color-coded version of the NFL schedule the night it is released. Tonight is that night.

Download the Excel file here

Some notes:

As usual, the games are color-coded based on time: Thursday (the Thanksgiving slate is week 13, with Detroit/Chicago repeating as the early game, Dallas/Buffalo in the afternoon, and Atlanta/New Orleans repeating at night) games are in light red, Sunday games at 1PM have no color, Sunday afternoon games are in orange, Sunday night games are in green, and Monday night games are in blue. In addition, in week 16, three of five possible matchups currently listed as TBD will be scheduled for Saturday. Those games are all in red, with white font.

There are five international games, with four in London. The Bears play “at” Oakland in week 5, Carolina plays “at” Tampa Bay in week 6, the Rams lose a home game to the Bengals in week 8, and the Jaguars “host” the Texans in week 9. In addition, the Chargers and Chiefs will play in Mexico City on Monday Night Football in week 11. That game is color-coded in blue for Monday Night, but with yellow font for international. Yes, my schedule grid has an easter egg.

Enjoy, and let me know if you spot any errors in the comments.

Last year, I wrote about the distribution of passes across NFL fields, both horizontally (left/middle/right) and vertically (short/deep).

Today I want to do a quick update on two of those 6 boxes: passes to the short left and passes to the short right parts of the field. Historically, passes that are short and to the left side of the field have been slightly more effective for teams. Despite that, teams throw slightly more passes to the short right side of the field than the short left. Did that hold true for 2018?

Yes and yes. There were 5,079 passes marked as “short right” in 2018, and 5,508 attempts marked as “short left.”  And once again, passes that were thrown short and to the left were slightly more effective. Passes thrown short and to the left were completed 73.0% of the time, and had a slightly lower interception rate and slightly higher yards per completion rate, too.  In the aggregate, it does appear that throwing short and to the right is better than throwing short and to the left

So we have here a bit of an inefficiency, at least on the surface.    It seems as though teams should be throwing more short left passes than short right passes, but the opposite is happening.  What if we look at the individual data — are some teams throwing short passes more often to the left, and other teams are throwing significantly more often to the right?

I sorted all quarterbacks with at least 50 passes that were either short left or short right, and then noted which quarterbacks had the largest disparity between those.  On one side, you have  Tom Brady, C.J. Beathard, and Jared Goff.  On the other side you have Blake Bortles, Kirk Cousins, and Eli Manning.

Anyone want to guess which trio of quarterbacks was throwing short left at a disproportionately high rate?  If you guessed the quarterbacks coached by Bill Belichick, Kyle Shanahan, and Sean McVay, you are correct.

Now, I don’t know enough to say that there is a definite market inefficiency that can be exploited. But after seeing the results from this table, I am now more inclined to think that there is one.

Russell Wilson has been with the Seahawks for 7 seasons, and in 5 of those years, there has been a large discrepancy between how well Seattle has passed and how often the Seahawks have passed.

In 2018, the Seahawks ranked 7th in Adjusted Net Yards per pass Attempt and 32nd in number passing plays (pass plays plus sacks).

In 2015, the Seahawks ranked 3rd in ANY/A and 27th in pass plays.
In 2014, the Seahawks ranked 8th in ANY/A and 32nd in pass plays.
In 2013, the Seahawks ranked 5th in ANY/A and 31st in pass plays.
In 2012, the Seahawks ranked 6th in ANY/A and 32nd in pass plays.

In general, there isn’t much of a correlation between pass efficiency and pass quantity. You might think the best passing teams would pass more frequently, but game scripts force the best passing teams to pass less frequently as a counterbalancing force.

So how unusual have the Seahawks been under Wilson? The graph below shows all team seasons since 2012, and each dot represents one team. The X-Axis shows where each team ranked in ANY/A and the Y-Axis shows where each team ranked in number of pass plays. [continue reading…]

Last year, I wrote a two-part series on how teams were using more highly drafted players. In 2017, 50% of all passes came from players selected in the top 32 of the draft, but I suspected that 2018 could be even more tilted in favor of highly drafted players. The reasons I suspected all came true, namely:

• Baker Mayfield, Sam Darnold, Josh Allen, and Josh Rosen all entering the NFL and potentially getting a significant number of pass attempts (all topped 300, as it turned out).
• Andrew Luck and to a lesser extent, Ryan Tannehill returned from injury.
• Patrick Mahomes became a starter, replacing 250th pick Trevor Siemian as an NFL starter, once you net out the Alex Smith (KC to WAS), Kirk Cousins (WAS to MIN), and Case Keenum (MIN to DEN) musical chairs. Deshaun Watson also became a full-time starter in 2018, but he actually falls below the median line as the 12th pick.

Thanks to those new starters, plus returning starters Eli Manning, Jared Goff, Matthew Stafford, Cam Newton, Jameis Winston, Alex Smith, Mitchell Trubisky, Carson Wentz, Marcus Mariota, Matt Ryan, Blake Bortles, Philip Rivers and 101 pass attempts from Blaine Gabbert – and yes, those 101 attempts were necessary — 2018 was a record-setting year.  Over half of all pass attempts in the NFL came from players drafted in the top 10 just the second time that’s happened since 1967 (the year of the common draft).
[continue reading…]

The Vikings and Falcons were two of the most disappointing teams in the NFL in 2018. They happened to be the two teams with the worst fumble recovery rates last year, which is largely driven by luck. On the other side, the Seahawks and Rams had the best two fumble recovery rates, and were two of the most overachieving teams in football last year.

The table below shows the fumble numbers for each team last year when they were the ones with the football. Seattle and Washington had the best fumble recovery luck in this state of the world (i.e., teams on offense). The Seahawks and Redskins each had 18 fumbles, and only lost 4 fumbles. That’s a 78% offensive fumble recovery rate, the best in the league. The average team recovered 57% of their own fumbles. The final column, therefore, shows the number of own fumbles a team recovered over expectation. For Seattle and Washington, they each recovered 3.8 more fumbles than we would expect, since they had 18 fumbles (we would have expected them to recover 10.2 own fumbles, but actually recovered 14). The full list below: [continue reading…]

The 20-yard shuttle is the Combine’s approach to measure an athlete’s agility, short-range explosiveness, and lateral quickness. Here’s the description from NFL.com:

The athlete starts in the three-point stance, explodes out 5 yards to his right, touches the line, goes back 10 yards to his left, left hand touches the line, pivot, and he turns 5 more yards and finishes.

As you can imagine, heavier players fare much worse in this metric, and taller players have a slight advantage, too. The best-fit formula from the 2019 Combine using height and weight as inputs is: 4.13 -0.0125 * Height (Inches) + 0.00485 * Weight (Pounds). In other words, for every 20-21 pounds a player weighs, he would be expected to take an extra tenth of a second to complete the drill. Ohio State defensive end Nick Bosa is 6’4 and weighs 266 pounds; that’s a formula for just being average in this drill. But he wound up completing the workout in just 4.14 second, but we would have projected Bosa to take an extra 0.33 seconds to finish, which means he is your 2019 Short Shuttle champion. [continue reading…]

For the last week, I’ve posted about the career passer ratings for quarterbacks after adjusting for era. Today is a simple data dump of the single season passer ratings.

Below are the era-adjusted passer ratings for every player in every season since 1932.  Here’s how to read the table below, which is fully sortable and searchable.  Sid Luckman has the best single season, playing in the NFL for Chicago in 1943.  That season counted for 11.58% of his career pass attempts (useful if you want to calculate a player’s career passer rating), as he threw 202 passes, completed 110 of them for 2,194 yards with 28 TDs and 12 INTs.  That was enough attempts to qualify for the passer rating crown; his actual passer rating was 107.5, and his Era Adjusted Passer Rating was 135.0, the best ever. [continue reading…]

On Saturday, I looked at the era-adjusted leaders in completion percentage. On Sunday, I did the same for yards/attempt, on Monday, I analyzed the era-adjusted leaders in touchdown rate, and yesterday continued the analysis but for interception percentage.

I thought it would be helpful to have all the information in one place, so that’s what today’s post is.  Here’s how to read the table below.
Otto Graham threw 2,626 pass attempts, and played from 1946 to 1955. He is in the Hall of Fame. Based on the passer rating formula — where 1.00 represents league average (a 66.67 era-adjusted passer rating), and a 1.50 in each category translates to a 100.00 passer rating — Graham scored a 1.40 in completion percentage, 1.53 in yards/attempt, 1.25 in touchdown rate, and 1.53 in interception rate. If you add those four numbers and divide by 6 — yes, this is exactly how passer rating is calculated! — you get 95.2, which is Graham’s era-adjusted passer rating. The full results are below.
[continue reading…]

On Saturday, I looked at the era-adjusted leaders in completion percentage. On Sunday, I did the same for yards/attempt, and yesterday, I analyzed the era-adjusted leaders in touchdown rate.  Today, we continue the analysis but for interception percentage.

Here’s a look at the interception rate in each year since 1932.  As you can see, there is much more variation (and in a much more straightforward manner) than there was with TD rate:

[continue reading…]

On Saturday, I looked at the era-adjusted leaders in completion percentage. Yesterday, I did the same for yards/attempt; today, we continue the analysis but for touchdown percentage.

Here’s a look at the touchdown rate in each year since 1932:

[continue reading…]

Yesterday, I looked at the era-adjusted leaders in completion percentage. Today, we’ll do the same thing but with yards per attempt.

The traditional passer rating formula measures Y/A by taking a passer’s Y/A, subtracting 3.0, and dividing the result by four. This makes sense when the average Y/A is around 7.0; in that case, 7.0 minus 3.0 equals 4.0, and dividing that by 4 gives a result of 1.00. But when the average passer isn’t averaging 7.0 yards per attempt, this formula isn’t so great. The graph below shows the average Y/A for all passers since 1932:

[continue reading…]

Regular readers know that I have spent some time over the past few years adjusting passer rating for era. One valuable part of the methodology is that we can also adjust each of the four component parts — completion percentage, yards per attempt, touchdown percentage, and interception percentage — for era.

Let’s take completion percentage. The passer rating formula measures completion percentage by taking a passer’s completion percentage, subtracting 30%, and multiplying the result by five. This made sense when the average completion percentage was around 50%; in that case, 50% minus 30% equals 20%, and multiplying that by 5 gives a result of 1.00.

To adjust for era, we replace “30%” in that formula with “league average minus 20%.” So in 2018, the league average completion percentage was 64.9%, which means we would use 44.9% for this formula. Drew Brees completed 74.4% of his passes; if we subtract the baseline from his result, we get 29.5%. Multiply that result by 5, and Brees gets a completion percentage score of 1.48 for 2018.

If we do this for every quarterback in every season of his career, and then weight each season by his number of pass attempts, we can get career grades. This is one way to come up with career completion percentages adjusted for era.

The overwhelming champion in this regard is Sammy Baugh, who led the NFL in completion percentage 8 times during the decade of the ’40s. As recently as 1975, Baugh was still 4th all-time in career completion percentage, and less than 1% off of the leader. Baugh has a rating of 1.58, which means on average he was better at completing passes relative to his era than Brees was in 2018.

The top passers in measuring completion percentage this way are Baugh followed by a who’s who of the completion percentage kings: Len Dawson, Otto Graham, Steve Young, Joe Montana, Sid Luckman, and Drew Brees.

The bottom 5? Rex Grossman, Jay Schroeder, Doug Williams, Mike Pagel, and the man at the very bottom of the list is… Derek Anderson. [continue reading…]

[continue reading…]

Two years ago, I wrote a 6-part series describing how to adjust passer rating for era. I posted the career results in Part V, and the whole series is background reading for anyone who wants to learn how to adjust passer rating for era.

Last year, I updated those numbers based on the 2017 results. Earlier this year, I posted the 2018 single-season results, and today, I am going to update the career ratings.

Here’s how to read the table below. Otto Graham threw 2,626 passes, and played from 1946 to 1955. His actual passer rating was 86.6, but his era adjusted passer rating was 95.2, the best in pro football history. The final column shows whether a player is in the Hall of Fame, is a HOF lock (attributed to five players), is not in the Hall of Fame, or has never been eligible for the HOF. [continue reading…]

The broad jump is a good way to measure a player’s all-around athletic ability. As a rule of thumb, the drill is heavily biased in favor of lighter players (who can jump farther since they weigh less), but it is also biased in favor of taller players, who have longer legs. Therefore, to adjust for weight and height, we use the following formula, based on the actual 2019 results:

Projected Broad Jump = 110.31 + 0.63 * Height (Inches) – 0.164 * Weight (Pounds)

Here’s a graph showing the expected broad jump results for a player based on a variety of different heights and weights.

Last year, Virginia Tech safety Terrell Edmunds (now with the Steelers, drafted 28th overall) posted the best broad jump. This year, it was safety Juan Thornhill of Virginia — who also posted the best vertical jump — who was the broad jump champion.

At just six feet tall, Thornhill wouldn’t be expected to dominate this event, but he did, jumping a whopping 141 inches. That’s tied for the second-most over the last two decades, and easily the best by a player 6’0 or shorter. The full results below.

Yesterday, I looked at rushing success rate for individual running backs. Today, I perform the same analysis for running backs, but at the team level (and ignoring runs by non-RBs).

Here’s how to read the table below. The Rams led the NFL in rushing success rate by running backs last season. Los Angeles RBs had 363 carries (after removing 3rd or 4th and long runs that did not pick up a first down) and 228 of them were successful, a 62.8% conversion rate. That was the best rate in the NFL. As noted yesterday, Todd Gurley was great (60.2%), but the other Rams running backs had even higher rates. It was truly a remarkable rushing attack in Los Angeles last year, at least until the NFC Championship Game and the Super Bowl. [continue reading…]

This past week, I’ve looked at the reasons why I think yards per carry is an overrated and misleading statistic. It’s just as, if not more valuable, to examine how often a running back is successful, and yards per carry tells us nothing about the distribution of a rusher’s performances.

Today, I want to study running back success rate. What do I mean by that? It’s simply the number of successful running plays divided by the total number of running plays; in other words, it’s the rushing analog of completion percentage. How am I calculating this metric?

Let’s start with the denominator: which rushing plays are included? All rushing plays are included but with one exception: I have discarded all runs (a) on 3rd or 4th down, (b) with greater than 5 yards to go, and (c) where the running back failed to get the first down. If a team calls a run play on 3rd-and-6, I am not going to fault the running back. I will simply discard the play. However, if he actually picks up the first down on 3rd-and-15, I will count the play. Only 3% of rushing plays were excluded using this, but it just “feels” like the right thing to do. [continue reading…]