09 Jan 2007
Last week we looked at our game charting data for defensive backs from the wild card games. I figured it was a good idea to check out the top two seeds in each conference, since we left them out last week.
The data now consists of Weeks 1-15, and while there are some holes, we've tried to get as many games as possible involving the eight remaining playoff teams. Standard caveats, of course: this data is unofficial, compiled by volunteers, and it is hard to always see who was in coverage when using TV tape. Some definitions:
• Yards per pass: Simply yards per pass attempt with this defender listed as the main one in coverage.
• Stop rate: The percentage of plays with this player in coverage that fell short of 45 percent of needed yards on first down, 60 percent on second down, or 100 percent on third down.
All rankings are out of players with at least 30 charted passes.
I've written about this each time I've looked at the game charting data this year
. While it isn't quite as extreme as it was early in the season, there's a colossal difference between Samari Rolle and Chris McAlister. Rolle's stop rate of 43% puts him among the 10 worst cornerbacks. He's also allowed 11.0 yards per pass. That's the highest of any cornerback in the league except one: Kevin Dockery, the Giants nickel back who I mistakenly referred to as a safety last week.
McAlister, on the other hand, has a stop rate of 62%, which ranks seventh among all cornerbacks (he's tied with Champ Bailey). He's allowed 7.1 yards per pass, which is about the league average. I'm guessing that McAlister has a high stop rate despite an unimpressive yards per pass because Baltimore opponents are often in third-and-long thanks to sacks and stuffed running plays. McAlister has stopped 18 of 24 charted passes on third down. Rolle has stopped 8 of 18.
As far as the other players, Ed Reed's numbers are actually really awful, a 29% stop rate and 14.2 yards per pass on 21 charted passes. Reed had the highest stop rate of any safety in 2005, so I don't know if this is just limited data, or our charters screwing up the identity of defenders on those passes down the middle, or Reed being used differently in coverage, or some combination of all three. Everyone pretty much agrees that Reed has played very well this year, and I think the safety and linebacker numbers from the charting project are shakier than the cornerback numbers, since corners are much more likely to be in obvious man coverage.
The big difference here is not in the numbers but in the number of targeted passes. Charles Tillman has a better stop rate than Nathan Vasher, 61% to 56%, but he gives up slightly more yards per pass, 5.9 to 5.4. Both players are above average in both stats. But we have 88 charted passes with Tillman as the main defender, and only 48 with Vasher as the main defender. We've charted more passes at Tillman than any other defender in the league, and while you might think this is because some teams have incomplete data this year, Tillman was second behind Ike Taylor in 2005, and Vasher was way behind Tillman last year as well.
Ricky Manning comes out with a lower stop rate and more yards allowed per pass (46%, 7.7). Also, he's a BLEEP .
Like I said, linebacker numbers aren't quite as trustworthy as cornerback numbers. But Lance Briggs has a 62% stop rate, second behind Zach Thomas, and allows 4.6 yards per pass, second behind Keith Bulluck. He was near the top in both categories last year too. Briggs is one of the top five linebackers in the number of passes charted with him as the primary defender.
OK, this one is really strange. When I first looked at charting data, Fred Thomas came out as one of the top guys in the league. Then he got injured, and when he came back he was getting super-toasted by everybody he tried to cover. Now his numbers are worse than those for both Mike McKenzie and the other cornerback in New Orleans, who I bet you can't even name.
Even weirder is the split between New Orleans before and after the bye week, which is roughly when Thomas was injured.
Mike McKenzie: 25 charted psses, 36% stop rate, 10.2 yards per pass.
Fred Thomas: 39 charted passes, 67% stop rate, 3.3 yards per pass.
Other Guy: 21 charted passes, 48% stop rate, 6.3 yards per pass.
Mike McKenzie: 30 charted passes, 77% stop rate, 4.8 yards per pass.
Fred Thomas: 36 charted passes, 33% stop rate, 14.9 yards per pass.
Other Guy: 30 charted passes, 77% stop rate, 4.8 yards per pass.
"Other Guy" is eight-year veteran Jason Craft. Yes, his numbers from Weeks 8-15, based on the data collected as of today, are the exact same as McKenzie, and both are far better than they were before the bye week, and Thomas is far, far worse. Honestly, I have no idea what the heck is going on here. Craft's charting numbers were horrific in 2005. In 2006, based on data so far, he's tied with Pac-Man Jones for the best stop rate among all cornerbacks (65% for the whole year).
This one is pretty interesting, especially since the secondary was supposed to be the weakness of the Chargers. Drayton Florence has one of the best stop rates in the league, 63%, and Quentin Jammer's stop rate is just 47%. But they allow roughly the same yards per pass (6.1 for Florence, 6.5 for Jammer), because Florence allows double the yards after catch.
Nickel cornerback Antonio Cromartie also has a 63% stop rate. He's allowed 8.4 yards per pass, which doesn't look good until you see that the average pass we've charted against him is 20.3 yards in the air. That's the highest of any cornerback with at least 30 charted passes by a good THREE YARDS. (Fred Thomas, by the way, is second. He really has been super torch-o-licious over the second half of the year.)
Donnie Edwards, like Lance Briggs, is one of the top five linebackers in the number of passes charted with him as the primary defender.
Possibly the closest Super Bowl matchup in history also poses the question: how much does it mean when certain aspects of an NFL team improve dramatically in the second half of the season?