Should Michael Vick Run More Against the Blitz?

Check out this chart, numbers courtesy of Pro Football Focus. It shows the breakdown in Michael Vick’s performance when he dropped back to pass against the blitz, in 2010 and then 2011. The chart is stacked so that you see how all the individual slivers add up to 100 percent:

What do we see? Sacks, down slightly. Touchdowns, down slightly. Other completions, up from 35 percent to 44 percent. Regular incompletions, down 3 percent. All good — except for the touchdowns.

Then there’s Vick’s runs and his interceptions. Granted, juxtaposing these two stats isn’t necessarily fair play. But there’s an interesting correlation, whereby Vick cut his scrambling in half from 14 percent to 7 percent of plays against the blitz while his interception rate on those plays jumped from 1.6 percent to 5.6 percent.

On one hand, calming down in the pocket and passing under pressure is an important skill to be an elite quarterback. On the other hand, maybe Vick would get himself into less trouble if he allowed himself to run a little bit more and forced his throws a little bit less.

Has Michael Vick Already Peaked?

Last week, Ron Jaworski went on SportsCenter and talked about Michael Vick’s potential in 2012:

“Vick has shown he is capable of throwing the ball exceptionally well from the pocket,” said Jaworski. “His overall throwing skill set can be top five in the league. His objective in 2012 must be to play that way more often. It becomes an availability issue. You can’t be an elite NFL quarterback if you can’t be counted on every single week.

“I am really excited to see Michael Vick in 2012. A more disciplined player will result in fewer turnovers. I would not be surprised if we’re getting ready to see the best year of Vick’s ten-year career.”

While Jaws supposedly watched every 2011 snap of Vick, his conclusions seem half-baked, especially compared to Sheil Kapadia’s epic breakdown of Vick’s game for the Eagles Almanac (plug alert!). For example, Sheil noted all of Vick’s injuries came on hits in the pocket, not because he was running around. Availability seems to be less of an issue (an NFL team loses its starting QB, on average, for three games a season) than accuracy and decision-making.

Regardless, Jaws’ sentiments are those I think a lot of fans hold. Vick had an amazing season in 2010, then fell back to Earth in 2011. But, we are told, his first full offseason as the starter with Marty Mornhinweg and Andy Reid will push him back to the top. It’s not a crazy opinion, but it is an optimistic outlook, and one that I’m not sure there’s any more evidence for than that Vick has already peaked, and he’ll never again reach that height.

In the spirit of a series of posts I did about Donovan McNabb two years ago (yeesh, that long ago?), I put together a new graph showing where Vick ranked on key statistics, during the years he was the main starter. By focusing on the rankings, rather than the stats themselves, we can see how well Vick has done compared to his peers — since the last ten years has resulted in a better passing environment pretty much across the board. QB Rating is slightly bolded because it’s more of an aggregate indicator than a separate statistic.

Obviously, Vick’s passing career has been less than stellar overall. Through 2006, he was below average across the board, and especially terrible in completion percentage. He went on his two year hiatus, came back as a wildcat threat in 2009, and resurrected triumphantly in 2010.

The question of “what’s next?” remains, though. There was a big drop off from 2010 to 2011, approximately from Aaron Rodgers-level to Jay Cutler-level. Was that a temporary bump in the road, soon to be righted after an offseason of hard work? That’s certainly the conventional wisdom right now.

The other answer, while not necessarily disastrous, is significantly less Super Bowl-worthy. That possibility suggests that Vick simply regressed to his career mean. His numbers were still up from 2006, but one might attribute that to slightly improved play and a much better supporting cast. Vick never had DeSean Jackson, Jeremy Maclin, Brent Celek, and LeSean McCoy in Atlanta.

Ultimately, there’s no such thing as a prediction engine for player performance. Vick may really benefit from these months of personal tutoring, allowing him to overcome the blitz-happy adjustments many teams made against him last year. Or, at 32 years old, it may be too late for him to completely change his ways. Jaws’ statement about Vick’s talent and potential are the same things people have been saying about him for the last decade.

One’s personal expectation probably has a lot to do with your thoughts on Mornhinweg and Reid’s mentor skills. In my experience, you doubt their abilities to coach and gameplan at your own risk. But I also wonder how much this ballyhooed “first full offseason as a starter” will really make a difference. Vick has been playing at a more mediocre rate since the end of the 2010 season, and we’re supposed to believe the three of them haven’t had time until this offseason to go over that?

For now, among the cautiously optimistic analysts, consider me more cautious than than optimistic.

Photo from Getty.

Is LeSean McCoy Better Than Brian Westbrook?

Thank you everyone who has already bought the Eagles Almanac 2012! I’m really proud of the work we’ve done on this book, and I hope you all enjoy it.

This week, I’m going to share a series of smaller graphics and other posts based on the work in the Eagles Almanac. For those of you who bought it, hopefully this will provide an opportunity to discuss some of the findings (since that’s difficult on an ebook). And for those who haven’t, you’ll see what you’re missing.

Below is a simple chart from my article, which was a detailed examination of LeSean McCoy’s running style and the areas he can still improve. The question this chart poses, as the title suggests, is whether Shady is a better back than Westbrook, using Football Outsiders’ year-by-year rushing plus receiving DYAR (defense-adjusted yards above replacement). Your answer probably varies from “so far” to “not yet.”

Dion Lewis and the Eagles Kickoff Return Futility

When you looked at the 2011 Eagles roster as a whole, it was plain that the team had enough overall talent to win more than eight games. But over and over again they demonstrated that they were less than the sum of their parts, failing at all the little parts of the game, like tackling and holding on to the ball.

Another one of these factors was kickoff returns. Dion Lewis, the rookie running back who didn’t return kicks in college, took nearly all of the Eagles returns. And he was (predictably) dreadful at it.

As you can see above, out of 47 players who returned at least 10 kickoffs, Lewis ranked 43rd. His 21.7 yards per return was 3.3 yards from the median, which cost the Eagles an extra 100 yards (over 30 returns) they surely could have used.

Lewis’s longest kickoff return was a mere 33 yards, again 43rd in the NFL. Whatever vision or explosiveness or speed that an kick returner needs to break through for a big gain, he didn’t show it last year.

Note: Long returns do skew the average kick return results, obviously. In some ways that’s good — achieving that is part of what makes a good returner. However, a returner could be overall below average but have one big return that skews everything upward. For the record, taking out the top return does little to improve Lewis’s numbers relative to everyone else.

It’s worth pointing out that this isn’t a problem solely related to the rookie from Pitt. Lewis is only the latest in a string of poor kick returners, who you can see at right. He was actually better than Jorrick Calvin. Quintin Demps, back in 2008, was the last above-average returner the Eagles employed. It would be nice to see the Eagles try to correct that in 2012, although I won’t be holding my breath.

Photo from Getty.

Touchbacks: Returning More With Less

Last year, I projected that the new kickoff rules — mainly the new default kick spot at the 35, rather than 30 yardline — would result in a huge increase in touchbacks. The math from 2010 showed that touchbacks would likely go from 16 percent to almost 40 percent of all kicks.

Turns out I was close, but not high enough. According to what I compiled directly from play-by-play data, 44.5 percent of all kicks were touchbacks in 2011. But let’s go a little bit deeper.

Just how much did those extra five yards help the kickers? Check out the graphs at right.

In 2010, only 38 percent of all kickoffs made it to the end zone. With an extra five yards of distance, NFL kickers took advantage. They kicked 81 percent of all kicks into the end zone in 2011.

But that’s not all. With that kind of increase, we would expect even higher levels of touchbacks than just 44 percent. So what kept it down? Apparently, the returners.

The chart above plots the percentage of kickoffs that became touchbacks by where they landed in the end zone. Obviously, as the kicks got deeper, returns became more rare.

The odd thing is that the end zone kicks were returned much more frequently in 2011 than in 2010. The first two yards are still almost always returned. But under the old rules, returners frequently didn’t bring out kicks that were just 3 or 4 yards deep. By 5 yards into the end zone, 72 percent were touchbacks.

This past year, that changed. Returners actually brought out more than half of all kicks as deep as 7 yards into the end zone. There were still more touchbacks overall, and a higher percentage of kicks into the end zone became touchbacks with the added distance. But returners took more risk than before, even knowing that the coverage teams also had five yards less to make up.

Perhaps kick returners felt the need to justify their continued presence on the roster. Kneeling down over and over would drive many a fierce competitor to take unnecessary risks. It will be interesting to see how that changes in year two after the kickoff changes.

Photo from Getty.

Charting the Eagles 2011 Playoff Odds

For the Eagles to win the division and make the playoffs, they need to win out against the Cowboys and Redskins, and hope the Giants lose this week and beat the Cowboys next week.

That’s still an unlikely string of outcomes, but it’s much better playoff chances than the Eagles have had in weeks. Football Outsiders head honcho Aaron Schatz was gracious enough to provide archives of their week-to-week playoff odds, and I’ve compiled the Eagles numbers here:

You can follow along the Eagles ups and downs, from the heights of perceived Super Bowl glory (18 percent chance of winning it all after week one) to the brief resurgence following week eight, to getting eliminated from wild card consideration completely.

Amazingly, the current 15 percent chance is the highest they’ve had in seven weeks. Still a long shot, but it at least makes this weekend interesting — a fan luxury we wouldn’t have if any of the other NFC East teams were any good.

Worst NFC East Ever?

It’s tough to watch this year’s NFC East. Eagles fans have long prided themselves on the fact that their team plays in one of the toughest divisions in the league. This year, that just isn’t true.

I charted the Football Outsiders overall team efficiency DVOA for each NFC East team since 2000. The results are below:

So it turns out, on a purely average DVOA basis, that this is not the worst NFC East in the last decade. In fact, this year’s group is barely below the 2010 average.

Perhaps the real reason this year’s group seems so awful has more to do with the lack of even one good team. To this point, among all “best teams in the NFC East,” this year’s Giants are by far the worst. In every year other than 2000, at least one team has posted a DVOA score of at least 20 percent, and the 2011 Giants would be no better than third in the division in any other year since 2005.

The way they’ve played this year, the Eagles should not have even the 3 percent chance they have of winning the division. Widespread mediocrity will do that for you.

On Aussies and Adjusted NFL Punting Statistics

One of the less heralded but more puzzling moves the Eagles made this offseason was letting punter Sav Rocca walk in free agency. By almost every conventional metric, Rocca had perhaps his finest season.

At age 37, in his fourth year in the NFL, Rocca had the highest punting average, the fewest touchbacks, and the most  punts landing inside the 20 of his career. Yet, despite his performance, Rocca was allowed to take his talents down to Washington, and the Eagles picked up undrafted rookie Chas Henry out of Florida.

Perhaps the move wasn’t so much about Rocca as it was about Bobby April starting fresh with young specialists who he can coach up. Still, the move was at least a minor head-scratcher, considering the veteran’s career year in 2010.

So, naturally, I began to wonder if Rocca’s season was actually as good as we thought. Basic punting statistics are particularly unreliable because they rely heavily on the situation. When a punter is backed up in his own territory, he gets to stretch his leg as far as he can, achieving distance above everything else. However, if a player has more punts from near midfield, he will have to kick for more accuracy and less power. The net punting average will depend heavily on this distribution of situational punts.

Derek Sarley of Iggles Blog tackled this problem last offseason. He compiled the play by play data and came up with a situational-based “optimal” punting scale. It works like so:

Yline = Line of scrimmage for the punting team.  Number 1-99 (theoretically) from their own goal line.

Punt = Punt distance

Return = Return distance

Result = Yline + Punt - Return

Optimal Result = “IF ( Yline < 40 , Yline + 50 ,  90).”  In English, if the line of scrimmage was between the 1 and 39  yard lines, I made the optimal result a 50-yard net punt.  If it was on  the 40-yard line or beyond, I called the optimal result a change of  possesion on the 10-yard line.  There are opportunities for further  refinement here, but as you’ll see in a minute, it won’t matter that  much once we start comparing apples to apples.

Difference = Actual result - Optimal result.

Obviously, “Optimal” doesn’t necessarily mean perfect. But it works to show about the top 10 percent or so of all punts, accounting for field position. All punters average results below the optimal point, but the better punter they are, the closer they come. You can see the full results of the top 36 punters in the NFL in 2010 (compiled and tabulated painstakingly from play by play data) by clicking on the thumbnail picture at right.

Note that ranking the punters by difference from optimal sometimes confirms preconceived notions of greatness, and sometimes it debunks them. This statistic shows that Dallas’s Matt McBriar, for instance, deserved his Pro Bowl nod. Oakland’s Shane Lechler, on the other hand, seems less impressive when you normalize the punting situations.

Overall, Rocca was ranked 11th in the league last year, good enough for the top third. However, we can dig a little bit deeper by comparing Rocca’s stats to those of the average punter at each point on the field. See the chart below with splits from line 1-20, 21-40, and 41+ (small sample size warnings for Rocca’s numbers apply):

As you can see from the “Difference” column, Rocca’s punts had better results than the average at every point on the field. He wasn’t an elite punter, by any means, but there wasn’t much cause to let him go either. In other words, he was in about the same place as David Akers: above average but not substantially so.

This supports my theory that clearing out special teams was as much about setting a new tone and bringing in players April could teach as actually improving the results. Still, I’m interested in any other theories or lessons you can take away from this data.

Photo from Getty.