Writers of Pro Football Prospectus 2008

21 Nov 2005

2005 Quick Reads: Week 11

Here's the latest edition of Quick Reads, featuring the best and worst DPAR ratings of the week. Somebody had the most valuable game of the season yesterday, but it wasn't Peyton Manning or Carson Palmer. It was Drew Brees. The Sage Rosenfels joke originally read "After watching this game, Reform Judaism reconsidering stance on patrilineal descent" but I decided that this was even more obscure than the Bobby Engram Scientology joke from a couple weeks back.

Posted by: Aaron Schatz on 21 Nov 2005

41 comments, Last at 22 Nov 2005, 8:25pm by Gut of the quantifier


by Sean (not verified) :: Mon, 11/21/2005 - 5:32pm

Argument for drafting Matt Leinart, exhibit A:
After Brees, the next best performances were turned in by, in order- Peyton Manning, Michael Vick, Carson Palmer, Byron Leftwich and Steve McNair. (Joey Harrington is actually next!) Apparently, there seems to be something to this whole, get good quarterback play from a franchise quarterback thing.
Argument against drafting Matt Leinart, exhibit A:
The following players all had statistically better days than Eli Manning, the last QB to go #1 overall (who actually suited up to play yesterday)- Drew Brees, Jake Plummer, Tom Brady, Marc Bulger, Mike MicMahon, Kurt Warner, Aaron Brooks, Trent Dilfer, Ken Dorsey.
There's more than one way to skin a cat.

by PerlStalker (not verified) :: Mon, 11/21/2005 - 5:39pm

I can't belive the Jets' rotten luck with QBs. I can see it now: "NFL Surviver: NY Jets QB" by NFL Films. The last guy standing (or at least kinda upright) gets to be the starting QB for the next week.

by Michael (not verified) :: Mon, 11/21/2005 - 5:41pm

You seem to have a personal vendetta against Eli with these similarity scores.

by RS Gray (not verified) :: Mon, 11/21/2005 - 5:49pm

We need some sort of formula to explain the fact that Michael Vick's DPAR is inversely proportional to Falcons' victories. All of Vick's supporters want to rebut any criticism of Vick with "Scoreboard" so now of course Vick has his three best games and the Falcons' go 1-2. Somehow, both sides are wrong.

by Sean (not verified) :: Mon, 11/21/2005 - 5:50pm

Actually, the Jets rotten luck is fairly easy to explain- it's called the offensive line. When you have five quarterback injuries in the course of ten games, it's safe to say that your offensive line is getting people hurt.

Really, it's David Carr who has had the rotten luck- he should have been knocked out for the season some time ago, thus saving him another six weeks of pounding.

by Slippery Pete (not verified) :: Mon, 11/21/2005 - 5:56pm

If the Jets have rotten luck with QBs, what does that make the Bears? At least the Jets got two good years out of Chad Pennington.

by Sophandros (not verified) :: Mon, 11/21/2005 - 6:01pm

RS Gray,

Or, as anyone who has ever played a TEAM sport knows, often players in certain positions receive too much credit for a win and too much criticism for a loss. In football, that position is often quarterback. BTW, it was relieving to see King mention Corey Simon's records in high school and college. You don't hear talking heads discuss a lineman's record as a starter.

In the case of the Falcons, the two losses were you mentioned were not Vick's fault, just as several of the wins attributed to him were despite him rather than because of him. For example, they used to play solid defense. Yesterday, they didn't. Their special teams have come up with plays in the past. That's just two other areas that have nothing to do with Vick, yet were huge reasons for Atlanta's success in the past.

If winning games or winning Super Bowls were the measure of great QB play, then Brad Johnson is as good as Peyton Manning this year. Are you willing to make that argument? By that logic, Trent Dilfer and Mark Rypien are better quarterbacks than Dan Marino and Jim Kelly...

by B (not verified) :: Mon, 11/21/2005 - 6:02pm

I get quoted on a national website, and it's a joke about lesbians. My mom will be proud.

by Mike B. (not verified) :: Mon, 11/21/2005 - 6:16pm

Re: #5 -

Really, it’s David Carr who has had the rotten luck- he should have been knocked out for the season some time ago, thus saving him another six weeks of pounding.OK, you just made me laugh so hard I almost passed out. 'Cause it's SO true...

by RS Gray (not verified) :: Mon, 11/21/2005 - 6:45pm

#7 - that was exactly my point.

by mshray (not verified) :: Mon, 11/21/2005 - 6:49pm

Michael, it's not a personal vendetta if you were here the first time Aaron did it. A few weeks ago Eli had some highly anomolous stats regarding accuracy & yards & at the behest of a reader he ran some comparisons. At that time Eli hadn't started 16 games so Aaron promised to re-run them when he had, to see if the anonmolies held up. And at the time Eli was the toast of the town & the Giants were highly ranked. Not a vendetta.

by Kibbles (not verified) :: Mon, 11/21/2005 - 7:03pm

Hey, Aaron, a little bit off topic, but is there any way you guys could come up with an adjusted success rate stat this offseason? I notice that Edgerrin James is leading the league in success rate, which is extremely impressive until I reflect on Indy's past schedule. Sure enough, James is only 11th in DVOA, while the guys ranked second and third have played actual NFL teams and as a result are ranked 1st and 4th in DVOA. Nothing fancy, just take the success rate and subtract the average success rate against the defenses faced (for instance, if a defense allows an average success rate of 42%, and a guy is successful on 50% of his runs, then it'd be an 8% adjusted success rate). Something like that, so you can compare the Mike Andersons and the Edgerrin Jameses of the world?

by Flux (not verified) :: Mon, 11/21/2005 - 7:18pm

I suppose this is the question you'll get the most of from non FO readers, but how does Steve Smith catch 14 for 169 yards against the #4 DVOA bears pass defense and not make the top 5? I didn't see the game, (I'm in the Bay Area; we get nothing but Raiders and 49ers every week. Yes, it's hell.) but did Carolina throw him like 25 passes, or what?

by charles (not verified) :: Mon, 11/21/2005 - 7:38pm

oh god, taylor jacobs, redskins fans like myself are not at all surprised he popped up on the worst receivers list.

by Mike Singleton (not verified) :: Mon, 11/21/2005 - 7:51pm

#13, I'm not sure on the exact total of passes thrown Smith's way but I'm pretty shocked at him not cracking the top 5 either. He had a huge catch on 4th and long to extend the drive at the end of the game after all. Jake threw his way quite a bit, but he still put up a pretty respectable % of catches. Of course, it's hard to catch passes that are thrown out of bounds your way or halfback passes that fall 20 yards short (Smith could have easily had 2 TDs on the day but Jake and Foster underthrew him on both occasions).

But yeah, maybe Aaron can explain it better.

by BHW (not verified) :: Mon, 11/21/2005 - 7:54pm

Throwing to receivers not named Chad Johnson this week, Carson Palmer was 18-25 for 146 yards (5.84 yards per attempt).

Going back two weeks, Palmer was 14-24 for 157 yards (6.54 Y/A) to non-Chad Johnson.

Going back about a month, he was 19-24 for 137 yards (5.71).

I'm getting those off of Quick Reads columns ... is the data for how many times a guy has been the intended receiver recorded anywhere?

Oh, I guess I can get it off of your page ... for the year, Palmer is 169-238 for 1,623 yards (6.82 Y/A), against an overall 230-331 for 2,620 (7.92).

Not sure how many attempts when Smith's way yesterday ... it might be interesting to see how the average QB does throwing to WR/TE who aren't his best receiver.

by Michael (not verified) :: Mon, 11/21/2005 - 8:16pm

I have been following this site since the beginning. Ill admit the vendetta accusation was harsh now that I think about it. It was in an article on Drew Brees, about his first two full seasons, that Aaron introduced similarity scores last year. He then used them to point out Eli's terrible play in his first few starts. This is the reason people have been asking about them this year. I felt last year it was far too early in Eli's career to try and use these comparisons and probably still is now. The most similar quarterbacks have been Vinny Testaverde (after last season), Steve Walsh(early this season), and Jay Schroeder(now). I interpreted Aaron pointing this out each time as shining an unfair negative light on Eli's development. Im sorry for making this an issue and I guess it would be best if we just dropped it.

by BlueStarDude (not verified) :: Mon, 11/21/2005 - 8:25pm

"Believe it or not, our system says Bledsoe is the year's highest-rated quarterback inside the red zone: 25-for-38 with 12 touchdowns, only two sacks, and zero turnovers."

Great! Now if he could only get us there more often, we could be rolling on into Detroit!

by zip (not verified) :: Mon, 11/21/2005 - 8:32pm

Yeah, I second the call for Steve Smith's rating from last week. Surely he must have been 6th or something? I guess he had a few catches that were short outs that got shut down, I guess those probably don't help.

Anyway, he had like 169 yard receiving, and Delhomme only threw for 235 or so, so he was kind of important to the offense. Very curious to see how he did.

by Richie (not verified) :: Mon, 11/21/2005 - 8:43pm

7 of 10 for 14 yards by Sage. That's impressive.

by Richie (not verified) :: Mon, 11/21/2005 - 8:59pm

I literally wonder if I could walk into an NFL game next week and complete 7 of 10 for 14 yards.

2 yards per completion.

On the other hand, he threw 2 picks - so only one of his passes hit the ground!

by Paytonrules (not verified) :: Mon, 11/21/2005 - 9:14pm

The Steve Smith thing - I watched the game.

I believe it's because he had two legitimate big catches - one big pass in the first quarter and that big catch on the last drive. There were a lot of passes caught that got nowhere, and while he was definitely the Panthers best player you never felt like he was dominating the game.

Think about it this way - how many times have you seen a game where a reciever had 14 catches, 169 yards an 0 TDs?

If only 2 of the throws to him were actually useful this is a perfect example of the statistics on footballoutsiders.com demonstrating an "empty" statistical performance. He didn't impact the outcome, he didn't produce points, because the Bears managed to 'contain' him, although the traditional measures don't show it.

by JonL (not verified) :: Mon, 11/21/2005 - 10:19pm

RE: #14

Actually, I think most Redskins fans are surprised he made it onto the field long enough to make any list.

by Jerry P. (not verified) :: Mon, 11/21/2005 - 10:53pm

"If only 2 of the throws to him were actually useful this is a perfect example of the statistics on footballoutsiders.com demonstrating an “empty� statistical performance. He didn’t impact the outcome, he didn’t produce points, because the Bears managed to ‘contain’ him, although the traditional measures don’t show it.'

Definitely agree. If anyone wants to see why putting up great "traditional" stats doesn't mean you'll have the best performance in DPAR take a look at this link:


The pertinent question was from Count Blah. Enjoy.

by mshray63 (not verified) :: Mon, 11/21/2005 - 11:28pm

Hey Michael, no biggie. We both been here for a while, I just figured you might have missed the part where Aaron promised to come back & review it again. In the interim Elisha decided to have what Peter King called "the worst game of his life" (which I questioned over on the MMQB thread).

BTW, am I the only one who thinks calling him Elisha is fun?

by Michael (not verified) :: Tue, 11/22/2005 - 12:04am

Tiki calls him Elisha all the time. Its an interesting name.

by thad (not verified) :: Tue, 11/22/2005 - 12:38am

Rypien in 1991?
Sure I would take him in a heartbeat over Kelly and Marino.

by charles (not verified) :: Tue, 11/22/2005 - 1:07am

yeah i swear rypien played kelly in the super bowl that year right? i mean they played head to head in that game? and who won again? who won mvp for that game? can someone refresh my memory?

by thad (not verified) :: Tue, 11/22/2005 - 1:33am

It wasn't just the Super Bowl.
Rypien as i am sure you recall, had a fantastic year start to finish.

by Joon (not verified) :: Tue, 11/22/2005 - 2:01am

aaron, i have a question about brees. i'm sure his rating was so high because buffalo had been #2 against the pass, but given that they were so awful against the run and decided to stack the box to stop tomlinson, shouldn't brees be adjusted upwards somewhat less? he had a great game, for sure, but i feel like putting up those numbers against the #2 pass D and #32 rush D is less impressive than, say, if he had done it against the #4 pass D and #3 rush D (the bears, say). perhaps there is an appropriate weighting of pass D and rush D that should be used to make the opponent adjustments? maybe something like 80/20? because it's clear that pass D and rush D aren't independent.

by tim (not verified) :: Tue, 11/22/2005 - 3:20am

doesn't it seem success rate for running backs is more dependent on the quality of the offensive line than the actual skills of the running back?

by Kibbles (not verified) :: Tue, 11/22/2005 - 3:23am

Re #30: The problem with that is that, theoretically, the Bills have known all season that they're studs against the pass and rubbish against the run. So they should have been stacking the box all season. And that still hasn't stopped them from being such a highly ranked pass defense.

It's not like Brees is the only QB with a great running game that the Bills have faced all season.

by Spoilt Victorian Child (not verified) :: Tue, 11/22/2005 - 3:28am

yeah i swear rypien played kelly in the super bowl that year right? i mean they played head to head in that game? and who won again? who won mvp for that game? can someone refresh my memory?
Who could forget? The moment that sticks in my mind is when Rypien scrambled away from Kelly's rush, heaved the ball downfield, got Kelly to bite on the slant (while the ball was in the air, no less), and made that great sideline grab at Kelly's 6. Rypien of course would go on to score the go-ahead touchdown on a double reverse.

by jebmak (not verified) :: Tue, 11/22/2005 - 4:10am

#33 nicely put

I can't believe that Gus isn't getting any credit. Sure, Sage only threw for 14 yards, but Gus was 4/18. It was definately a team effort.

by NF (not verified) :: Tue, 11/22/2005 - 4:31am


Free Cleo Lemon!

by James, London (not verified) :: Tue, 11/22/2005 - 8:18am

I get quoted on a national website, and not one person I know has any interest in the NFL.
Never mind...

by MCS (not verified) :: Tue, 11/22/2005 - 9:37am

Actually James, since you're here, wouldn't this be considered an international website?

by Joon (not verified) :: Tue, 11/22/2005 - 9:55am

Re: #31

no. maybe for power running conversion rate (3rd and 1 or 2, 4th and 1 or 2, 3rd/4th and goal from the 1 or 2), the OL matters more than the RB. but success rate includes all runs, adjusted for down & distance. so it's any time you get 40% of the way to the sticks on 1st down, 60% on 2nd down or 100% on 3rd or 4th. i think. the numbers may be a little off, but that's the idea.

Re: #32

yeah, i know. but that doesn't change the fact that when a team plays defense, it has to defend both the rush and the pass. so putting up big numbers against a D which is great against both is more impressive and more difficult than putting them up against a team that can only do one. maybe not by a lot, but it obviously matters. i feel pretty strongly that adjusting QB performance for opponents rush D as well as pass D is more accurate and would lead to better correlation with future QB performance. maybe this is not true, but i'd like to see aaron or somebody run the numbers.

likewise, RB performance should be adjusted (if only slightly) for opponent pass D, not just opponent rush D. like if tennessee has to play a lot of nickel to help out their struggling rookie CBs, it clearly opens up the middle for opposing RBs.

the awkward part about it is that teams obviously have a choice to make about how they defend. sometimes they will stack the box, and sometimes they won't. so i'd guess that buffalo (to go back to the original game that got me thinking about it) decided to make stopping LDT a priority, and their pass D suffered for it just this one week. but on a larger scale, it seems fairly clear to me that during the course of an entire game, teams will balance their strategies out somewhat, sometimes trying to stuff the run, sometimes playing back to prevent the pass. that's why i think that both of the components of the opponents' defensive rating should be used to adjust for QB/RB performances. again, though, we'll only really know when aaron has time to try it out and see if it improves the model's predictive ability.

by Art (not verified) :: Tue, 11/22/2005 - 10:03am

Aaron, what was Steve Smith's DPAR?

He had 14 catches vs a good Chicago defense.

by ElJefe (not verified) :: Tue, 11/22/2005 - 1:20pm

Re: 27-29

Fun stat about Rypien in 1991. In 16 games, he was sacked 9 times. In the 14 games he played against teams other than the Eagles, he was sacked twice. Two sacks in 14 games! There are a lot of guys who can play quarterback in the NFL if they're not going to be hit.

In years prior to 1991, and subsequently, Rypien was kind of a turnover machine because when he got hit, he fumbled. A lot. Not much of a problem in 1991.

Actually, checking the stats I am incorrect. He still managed to fumble 9 times in 1991. But ... an improvement over the 14 fumbles in 14 games in 1989. :)

by Gut of the quantifier (not verified) :: Tue, 11/22/2005 - 8:25pm

James, London

I noticed old bean.