Posts by Douglas Dolitsky

More Than A Feeling: Are Rushing Plays Becoming More Extreme?

Last Sunday, rookie running back Todd Gurley burst onto the NFL scene with a monster 19-carry, 146-yard day on the ground. Given his 7.7 yards per carry mark for the game, you would expect that Gurley did most of his damage with very good six- to nine-yard runs, mixed in with a couple of stuffs and

Last Sunday, rookie running back Todd Gurley burst onto the NFL scene with a monster 19-carry, 146-yard day on the ground. Given his 7.7 yards per carry mark for the game, you would expect that Gurley did most of his damage with very good six- to nine-yard runs, mixed in with a couple of stuffs and a few monster plays.

Turns out that you would be wrong. In fact, not a single one of Gurley’s runs fell within the six- to nine-yard range. Actually, four of his carries, or 21.1 percent, went backwards. Another five, or 26.3 percent, resulted in a gain of two yards or less. Another five went for three to five yards. The remaining five all went for at least twelve yards.

This feels like the norm in the NFL today. It feels like the five-yard run is an endangered species, and running backs are becoming increasingly dependent on the 20-plus-yard runs to keep their yards-per-carry numbers respectable.

But as Peyton Manning, who has managed to lead the Denver Broncos to a 5-0 record without any feeling in his right fingertips, would probably tell you, “It’s not all about the feeling.” So are NFL rushing plays really trending in a more extreme direction, or is that just our imagination?

The following charts illustrate the distribution of yards-per-rush-play in each of the past four seasons (including this one):

Image title

Image title

Image title

Image title

At quick glance, each of the charts seem to have just about the same shape. However, a closer look seems to imply that the 2015-16 graph has some key differences.

Let’s zoom in on the three- to six-yard range:

Image title

Rushes for each yardage value between three and six have been less frequent this season than in each of the previous three. Overall, rushes of three to six yards are down to 32.04 percent, three percent less than the 35.05 percent average of the 2012-15 seasons.

All of those rushes have to go somewhere. This season has thus far marked a four-year high in rushes for negative yards, no gain, nine yards, and ten or more yards. These numbers confirm that our feelings were correct. Take that, Manning!

The increase in the rate of “extreme” rushes is not the only trend the NFL is seeing in the running game. This season has also seen a four-year low in both carries per game and yards per carry:

Image title

It is very possible that all of these trends are related. As teams run less and less, running backs are given fewer opportunities. This puts pressure on running backs to make the most of the carries they do get. Therefore, rather than running in a way that will maximize the expected yards, they are running in a fashion that will maximize their chances at a homerun. When they manage to evade the defender or break a tackle, it’s off to the races. However, they also leave themselves susceptible to being bottled up near the line of scrimmage.

Of course, this is just one potential theory. Feel free to interpret the numbers as you will!

Read More 451 Words

Quarterforward: Has Andy Dalton Taken His Game To The Next Level?

There's a reason that the NFL plays a new season each year. It is true that many things in the NFL tend to be constant, such as Arian Foster going down with an injury, the Indianapolis Colts beating up on the AFC South, and Tom Coughlin's face turning red when the New York Giants make a mistake. However,

There’s a reason that the NFL plays a new season each year. It is true that many things in the NFL tend to be constant, such as Arian Foster going down with an injury, the Indianapolis Colts beating up on the AFC South, and Tom Coughlin’s face turning red when the New York Giants make a mistake. However, there are also so many changes from one season to the next that we can’t help but come back for more. We have reached a point in the 2015 season where we can easily identify some differences from last year.

How are starting quarterbacks playing this season, when compared to last season’s performance? The following chart shows the QBRs of the 24 quarterbacks who have qualified in each of the past two seasons, sorted in order of the QBR differential from 2014 to 2015. The 2015 QBR is reflective of numbers through Week 6.

Image title

Quite predictably, three of the top seven most-improved quarterbacks, Blake Bortles, Derek Carr, and Teddy Bridgewater, were rookies last season. We expect players to struggle in their rookie seasons, as they transition into the NFL, but we also expect them to improve the following season, when they are a bit more comfortable.

It is also not surprising that amongst the top seven quarterbacks, three are playing for new teams this season. Ryan Fitzpatrick is now with the New York Jets. Brian Hoyer has replaced Fitzpatrick with the Houston Texans. Josh McCown has replaced Hoyer with the Cleveland Browns, after spending last season with the Tampa Bay Buccaneers. 

When quarterbacks switch teams, they find themselves in situations with new coaching staffs, new offensive systems, and new sets of weapons to play with. Therefore, it is very likely that they will experience major swings in performance. The only other quarterback on the list who is playing for a new team is Nick Foles, who was traded from the Philadelphia Eagles to the St. Louis Rams this offseason. Notice that he also finds himself on an extreme end of the list, with the third worst differential.

After accounting for the sophomores and the team-switchers, we are still left with one other quarterback in the top seven: Andy Dalton of the Cincinnati Bengals. Dalton seems to have shown a drastic improvement this season, leading the 32 qualified quarterbacks in QBR, after ranking 21st out of 30 last season. In fact, the worst single-game QBR Dalton has posted through six games this season, which was 67.8 versus the San Diego Chargers in Week 2, is 13.9 points better than his overall QBR last year!

How much of this increase in QBR can be attributed to Dalton’s performance, as opposed to external factors? While Dalton is playing for the same team as last season, with the same head coach, in Marvin Lewis, and the same offensive coordinator, in Hugh Jackson, there are still other factors to consider. 

For example, can the improvement be attributed to a soft schedule? The following chart illustrates the difficulty of Dalton’s schedule via the Football Outsiders’ Pass Defense DVOA of the Bengals’ opponents this season:

Image title

The average rank of 18.5 out of 32 teams suggests that these teams are collectively slightly below average in pass defense. However, the schedule has not been nearly easy enough to make a mediocre quarterback look like an MVP.

Maybe Dalton’s performance has looked better due to an improved arsenal of weapons. There are five players who have caught double-digit passes from Dalton this season – A.J. Green (35), Tyler Eifert (28), Marvin Jones (24), Giovani Bernard (18), and Mohamed Sanu (16). None of those players has missed a game for the Bengals this season.

While each of those players was technically on the Bengals last season, some of them were forced to miss games due to injury. Both Green and Bernard sat out three games in 2014, while Eifert suffered a dislocated elbow in Week 1 that cost him the rest of the season and Jones didn’t play at all due to foot and ankle injuries. Eifert has been a monster this season, ranking ninth amongst tight ends in receiving yards per game, and tied for first in receiving touchdowns. Jones has been very solid as the number-two wide receiver, ranking 37th amongst wide receivers in yards per game, and tied for 15th in touchdowns.

If Dalton has really improved, his receivers should have better numbers across the board. Otherwise, it is likely that the Bengals’ improved passing game is merely a result of better health for the members of the receiving corps.

Three players have played significant roles in the Bengals’ receiving game in each of the past two seasons: Green, Bernard, and Sanu. When targeting those players in 2014, Dalton averaged 7.99 yards per attempt; this season he is averaging 9.38 on such throws. That number would actually increase to 9.52 if we used the same target distribution as last season.

Analyzing Dalton’s completion percentage tells a similar story. In 2014, Dalton’s completion percentage on passes targeting one of those receivers was 0.615. This season, it is 0.711, or 0.706, if we use last season’s target distribution. Clearly, Eifert and Jones are not solely responsible for the Bengals’ improved passing attack.

We are only seven weeks into the NFL season, and there is still a lot of football to be played. We don’t know if the 6-0 Bengals will continue to win every game they play. What is pretty clear, however, is that, to this point in the season, they have benefited from Andy Dalton’s superb play.

Read More 865 Words

Don’t Get Robbed: Why Rob Gronkowski Should Not Be Your First Round Draft Pick

Rob Gronkowski is the top fantasy tight end in the NFL. In fact, it's not particularly close. 

If we go by ESPN standard scoring, he scored 178 fantasy points last season, despite not playing in the last game of the season. To put that in perspective, that is 30 points more

Rob Gronkowski is the top fantasy tight end in the NFL. In fact, it’s not particularly close. 

If we go by ESPN standard scoring, he scored 178 fantasy points last season, despite not playing in the last game of the season. To put that in perspective, that is 30 points more than the next best tight end, Antonio Gates, and 41 more than third-place finisher Jimmy Graham. Gates is suspended for the first four games of this coming season. Graham is moving from a New Orleans Saints team that passed the ball 62% of the time last season (fourth in the NFL) to a Seattle Seahawks team that did so 46% of the time (last). The gap between Gronkowski and the field is quite large.

Judging by Gronkowski’s average draft position, it is pretty clear that most fantasy football participants would agree. In Yahoo leagues, Gronk is being selected with the 8.1 average draft position (ADP), the seventh highest such value amongst all players. Gronk’s ADP in ESPN leagues is a more moderate – but still very high – 14.4, the fifteenth best.

It is in your best interest to pass on Gronkowski at those junctures in the draft. The most obvious reason is his injury history. While Gronk has consistently put up elite fantasy numbers when healthy, due to injury he did miss a combined fourteen games in the 2013 and 2014 seasons.

However, while the injuries are surely concerning, they should not drop Gronk more than a few spots in the rankings. Every player comes with question marks and most injuries are unpredictable. And while quarterback Tom Brady‘s likely suspension will curb Gronk’s potential for the first four games, Gronk is a talented enough player to still thrive with a lesser QB.

The primary justification for dropping Gronk in the fantasy rankings has absolutely nothing to do with his expected performance. Rather it is the application of a very simple economic concept: opportunity cost.

For those of you who don’t know what that is, opportunity cost is the value of the opportunity that one foregoes when they choose one option over another. In other words, it is what you are missing out on.

If you expend a late-first or early-second round pick on Gronk, or any tight end for that matter, you are missing out on the opportunity to take an elite running back or wide receiver. The inverse, however, is not true; that is, if you draft a running back or wide receiver with that selection, you are in no way precluded from securing a great tight end later on.

Let’s take a look:



ESPN Tight End Average Draft PositionImage title






Yahoo Tight End Average Draft PositionImage title

As you can see, in either league, you can grab the second best tight end, Graham, with a late-twenty pick. A very solid tight end like Martellus Bennett should be available in the 60s. Heck, you could even wait until the very end of the draft and nab Gates. Sure, you will have to make due with another tight end for four weeks, but you would be getting three-quarters of a season’s worth of a great tight end for next to nothing.

So how do these tight ends compare to the other players available at those spots? For each of the top ten tight ends by ADP for each site, I will take a look at the top running back and wide receiver available at that point. I’ll also throw Gates in there for good measure, as he is likely going to be starting for his draftee’s team eventually. Along with each player, I included their ADP positional rank.

ESPN

Tight End Running Back Wide Receiver
Rob Gronkowski (1) Jeremy Hill (10) Odell Beckham Jr. (4)
Jimmy Graham (2) Mark Ingram (11) Mike Evans (12)
Greg Olson (3) Todd Gurley (19) Sammy Watkins (19)
Travis Kelce (4) T.J. Yeldon (24) Brandon Marshall (25)
Jason Witten (5) Giovani Bernard (25) Brandon Marshall (25)
Martellus Bennett (6) Giovani Bernard (25) Brandon Marshall (25)
Julius Thomas (7) C.J. Spiller (28) Mike Wallace (28)
Zach Ertz (8) Tre Mason (39) Breshad Perriman (39)
Jordan Cameron (9) Bishop Sankey (40) Breshad Perriman (39)
Dwayne Allen (10) Bishop Sankey (40) Breshad Perriman (39)
Antonio Gates (13) Andre Williams (45) Anquan Boldin (47)


Yahoo

Tight End Running Back Wide Receiver
Rob Gronkowski (1) DeMarco Murray (6) Dez Bryant (2)
Jimmy Graham (2) Frank Gore (13) DeAndre Hopkins (13)
Travis Kelce (3) Latavius Murray (18) Brandon Marshall (21)
Greg Olson (4) Latavius Murray (18) Brandon Marshall (21)
Martellus Bennett (5) Andre Ellington (21) Vincent Jackson (27)
Zach Ertz (6) Joseph Randle (22) Vincent Jackson (27)
Jason Witten (7) Jonathan Stewart (23) Vincent Jackson (27)
Jordan Cameron (8) Jonathan Stewart (23) Jeremy Maclin (28)
Julius Thomas (9) C.J. Spiller (25) Mike Wallace (31)
Owen Daniels (10) Chris Ivory (33) Michael Floyd (37)
Antonio Gates (11) Darren Sproles (37) Pierre Garcon (46)

In either league, the opportunity cost of drafting Rob Gronkowski is an elite (top 10) running back or wide receiver. Unfortunately, this will likely leave you in a situation where you are starting running backs or wide receivers who you would rather have on your bench (such as Bernard or Jackson), and if your starters are on a bye week or injured, you may even have to start RBs or WRs whom you would rather not have on your team at all (such as Williams or Garcon).

Or you can take the alternative option. You can select a very good running back or dominant receiver instead of Gronk, and still end up with a solid starting-caliber tight end falling into your lap much later on.

I’m not arguing that Rob Gronkowski is not a great fantasy tight end. I’m not even arguing that he is overvalued. However, at the very least, his tight end peers are being more undervalued than he is.

So while it may be tough to pass on having a guy that parties as hard as Gronk on your fantasy team, your team will likely look better because of it.

Read More 895 Words