Geno Smith facing long odds

— -- The New York Jets are sticking with quarterback Geno Smith -- for now -- despite a disappointing start to his second NFL season. It's a logical move as long as there's a chance Smith will improve with experience. Even if backup Michael Vick were to provide a short-term upgrade, he's 34 years old and hasn't been durable throughout his career. There is no long term with Vick.

The concern for the Jets is that quarterbacks over the past eight-plus NFL seasons have almost never improved significantly after their first 16 starts. ESPN analyst and former NFL executive Bill Polian joined me in breaking down some of the findings back in June. We singled out the 41 quarterbacks who made their first 16 starts from 2006 through last season. The ones who succeeded ultimately almost always played well early.

With Smith having passed the 20-start mark Sunday, and having posted a 27.6 QBR through four games after a 35.9 QBR last season (50 is average), that is ominous news for his chances to be a long-term success as an NFL quarterback.

I've updated the project to show how those 41 players performed from their 17th starts forward. The results bring into focus the long odds facing Smith and the Jets as they prepare for Week 5 and beyond. It's also instructive to see which quarterback is beating the odds and what factors could explain his success.

Parameters and initial findings

Polian and I sought a statistical standard for QB excellence over time to serve as reference point. We found that Peyton Manning, Tom Brady, Drew Brees, Aaron Rodgers and Matt Ryan were the only players with at least 60 starts over the past four years and a 65.0-or-higher score for those games on the 100-point Total QBR scale. That 65.0 score became our standard for excellence. This was also the point ESPN Stats & Information established as a general cutoff for Pro Bowl-caliber play.

Seven of the 41 QBs who made their first 16 starts since 2006 scored in that 65-plus range for those initial starts. Most have continued to play well, health-permitting. The list includes Ryan, Tony Romo, Robert Griffin III, Russell Wilson, Colin Kaepernick, Rodgers and Andrew Luck. Philip Rivers was next on the list with a QBR score around 60.1 for his first 16 starts (62.4 since then).

The 15 players with the lowest QBR scores through their initial 16 starts flamed out almost without exception. Working from the bottom up, that list includes Blaine Gabbert, Brandon Weeden, John Skelton, Kellen Clemens, JaMarcus Russell, Bruce Gradkowski, Mark Sanchez, Tarvaris Jackson, Brady Quinn, Smith, Kevin Kolb, Sam Bradford, Matt Moore, Tim Tebow and Matthew Stafford.

Keep Stafford's name in mind. He is the one player out of the 41 to improve significantly from the level he showed during his first 16 starts.

There were 19 other players falling between the top seven and the bottom 15 during their initial 16 starts. Of those, Joe Flacco is the only one to make meaningful gains over his initial 16 starts. That list, from top to bottom, includes Rivers, Derek Anderson, Matt Cassel, Nick Foles, Vince Young, Chad Henne, Cam Newton, Jay Cutler, Jake Locker, Ryan Tannehill, Shaun Hill, Andy Dalton, Matt Leinart, Jason Campbell, Josh Freeman, Christian Ponder, Trent Edwards, Flacco and Colt McCoy.

Stafford appears to be the exception

The chart below shows Stafford with a 59.8 QBR score for the 50 starts he has made since his initial 16 starts; that is a 48.4 percent improvement. For reference, consider that 52.2 is the composite QBR score for all starting quarterbacks since 2006. Peyton Manning leads the NFL at 80.1 during that span. Sanchez ranks last among qualifying quarterbacks during that span (38.5).

Stafford entered the NFL with an especially poor supporting cast in Detroit. The team had gone 0-16 the season before his arrival, and this extreme talent deficiency put Stafford at a disadvantage early. But as the Lions have built up the roster, Stafford's production has risen. Unlike the Jets' Smith, who was a second-round pick, Stafford was a No. 1 overall talent. Also unlike Smith, Stafford now benefits from a diverse offensive arsenal featuring the game's best wide receiver in Calvin Johnson. The Jets aren't close to surrounding Smith or any quarterback with that kind of talent, and Smith isn't as equipped as Stafford to get the most from that kind of supporting cast.

Beyond Stafford, Flacco has improved from a 43.1 QBR score over his first 16 starts to a roughly average 53.8 in 97 regular-season and playoff starts since then. Rodgers is next on the list, improving from a Pro Bowl-caliber 66.4 for his first 16 starts to an MVP-caliber 75.9 since then.

From there, none of the 24 players with 16 initial starts and at least 10 subsequent starts has improved more than 4.8 points on the 100-point scale. Miami Dolphins backup Matt Moore, with nine starts since his initial 16, barely missed the cut. His QBR score has jumped from 39.5 in his first 16 starts to 63.2 thereafter. Keep that in mind if Dolphins coach Joe Philbin wavers in his commitment to starter Ryan Tannehill (50.4 QBR score in first 16 starts, 45.1 since then).

A closer look at Geno Smith

We hear a lot about developing quarterbacks, but there hasn't been much statistical evidence lately suggesting players improve significantly. Setting aside stats, ESPN analyst Jon Gruden said last week he thought Smith had shown signs of improvement early in the season.

"When I see him complete almost every pass in the opening game and he has got the Packers down 21-9 and he is going in for another score when his arm gets hit and there's an interception, I see progress," Gruden said a week ago. "The Jets are asking Smith to do a lot. They are asking him to run a West Coast offense that Marty Mornhinweg ran with Andy Reid and Donovan McNabb. ... I heard some of the same criticisms of McNabb when he came out of Syracuse that I'm hearing about Smith now. McNabb had to adjust from the freeze option that he ran at Syracuse."

Smith ranked 32nd out of the 41 players, with a 35.9 QBR score during his first 16 starts. He completed 55.8 percent of his passes, with 12 touchdowns and 21 interceptions. Sanchez put up nearly identical numbers during his first 16 starts. His QBR score went from 34.6 over his first 16 starts to 39.4 in 52 subsequent charts. The team around Sanchez deteriorated, and that did not help him. Smith is also suffering from below-average weaponry. He has four touchdowns with five interceptions and a 27.6 QBR score through four games this season.

Buffalo's EJ Manuel, the only starter with a lower QBR score than Smith's 12.3 in Week 4, lost his job Monday. If the Jets feel as though Vick gives them a better shot at contending in a vulnerable AFC East, they could use that as justification for making a change. While that would seem short-sighted, precedent says Smith will likely remain about what he has been to this point.

Notes

• Long and short of it for Giants: Eli Manning and the New York Giants' offense have outperformed expectations over the past couple weeks in a stark reversal. What has changed? We can talk about Manning adjusting to a new system or Larry Donnell emerging as a viable target, but collecting nine opponent turnovers over the past two games has changed everything.

The Giants were last in average starting field position during their first two games, beginning inside their own 21-yard line on average. They were first in that category in the two subsequent games, beginning at their own 40, on average. Last season, the average Giants drive started just outside the 27-yard line, about a yard worse than average.

• Early read on new QB: Austin Davis, the St. Louis Rams' new starting quarterback, has put up better numbers than Sam Bradford produced for any three-game stretch to open a season. Davis has completed 72.3 percent of his passes while averaging 8.0 yards per pass attempt with a 93.1 passer rating and 66.7 Total QBR score.

Bradford's best numbers in those categories through three games were 64.2 percent completions (2012), 7.0 yards per attempt (2012), 88.7 passer rating (2013) and 41.9 QBR score (2013). Two caveats: The Rams' supporting cast is better now than it typically was for Bradford, and the team has shown a stronger commitment to running the ball from the beginning of the season.

• The Rob Ryan file: With the New Orleans Saints' defense struggling and coordinator Rob Ryan taking some heat, I revisited a chart I put together last year showing how Ryan's defenses had performed when he was with Oakland (2004-2008), Cleveland (2009-2010) and Dallas (2011-2012). There's also a column showing where his Saints defenses have ranked in various categories since the start of last season. The categories are ordered from best to worst rankings for the Saints. The expected points added (EPA) category is the one I like the most because it adjusts for situations, as explained here.

Ryan generally has not been associated with top-notch defensive personnel. I thought New Orleans was questionable in that area heading into last season, but the Saints wound up performing well on defense. The team appeared to make some upgrades this past offseason. I'll take a closer look at the Saints a little later in the week, but in the meantime, here is that chart: