11/18/07—Two Disturbing Trends Continue vs. Browns
I’ve mentioned several times over the last few weeks:
• The Ravens are a mirror image with respect to last year in that their blitzes have been beaten regularly
• The Ravens are giving up a mess of yards on 1st down passing plays, but are otherwise very stingy
I’m sad to report both of these trends continue.
Against the Browns, the Ravens YPP allowed by number of pass rushers were:
3: 7 plays/50 yards, 7.1 YPP
4: 17/63, 3.7 YPP
5: 14/156, 11.1 YPP
6: 2/-6, -3.0 YPP
So rushing 4 or fewer the Ravens allowed the Browns 4.7 YPP (season 5.5), but 9.4 YPP (season 8.1) with 5 or more rushing.
Should the Ravens stop rushing 5+?
First of all, you can’t become predictable in the pass rush any more than you can pass or run all the time on offense, so the Ravens will continue to mix it up. In addition, if Rolle and CMac are both back this week, this may well be a game where we let the corners play man. More men in the backfield is generally very good vs. the run (that’s something I’m sure you’ve observed watching the Ravens the Ravens the last 9 years, but it also deserves a statistical demonstration before you accept it as dogma). Additionally, Phillip Rivers has succumbed to the rush with 12 INT’s and 10 fumbles despite being sacked just 13 times in 10 games. Furthermore, with no Pryce, if the Ravens do not bring extra pressure, this is a game where Rivers may be able to order dinner in the pocket. This is one game where I hope Ryan ignores the 2007 numbers and plays 2006 defense.
I mentioned the trend of getting burned on 1st down several weeks ago, and the situation has not improved. Versus the Browns, the Ravens allowed 149 net passing yards on 16 1st down pass plays (9.3 YPP). For the season they have allowed 990 yards on 103 1st down pass plays (9.6 YPP). That’s compared to 3.9 YPPA on all other plays. Of all yards surrendered for the season, 34% have come on 1st down passes which have accounted for just 18% of the non-kneel snaps.
What’s causing the Ravens to get burnt so much in these situations?
Without having carved up the data further to corroborate, I’ll tell you what I’d guess might be some of the issues and how I might model each:
• The Ravens, particularly without Pryce, have a non-stellar pass rush crew on 1st down that has afforded opposing QB’s more time to throw. This one seems obvious, but ideally, I would check sacks and QB hits relative to the number of pass rushers to verify.
• The Ravens have gambled a lot on 1st down with 5+ man rushes used primarily to stop the run. These have not caused enough havoc in 2007 to make up for the times they are burned. To confirm, I’d check the distribution of pass rushers on 1st down relative to other downs.
• Other teams are doing a good job of noticing the Ravens tendencies and using schemes that either pick up the blitz or maximize downfield threats. This would require significantly more work, since I’d really need to check blocking schemes vs. each number of pass rushers. I do not have this currently recorded for opposing offenses, but I think it’s something I want to add next year.
• The unstable situation in the secondary has left no shortage of targets. One could build a model for this by assigning a “total secondary quality” score to each alignment. Something like 3 for each starter, 2 for a backup or player out of position, and 1 if the player was 2 deep on the depth chart. We could then add up the total for all DB’s, allowing nickel and dime sets naturally to have higher scores since they are intended to stop the pass.