Writers of Pro Football Prospectus 2008

12 Sep 2006

2006 Quick Reads: Week 1

Here's the latest Quick Reads at FOXSports.com with PAR for all quarterbacks in Week 1 plus the top running backs and receivers. Join us as we celebrate Football Outsiders vs. National Jump to Conclusions Week. Sorry this is late, I forgot to send it to FOX on Monday. Yes, I'm an idiot.

Posted by: Aaron Schatz on 12 Sep 2006

46 comments, Last at 16 Sep 2006, 1:17am by Sean_C

Comments

1
by Pat (not verified) :: Wed, 09/13/2006 - 12:20am

I think there's a mistake with McNabb's rushing PAR. McNabb's only rush was a rush for 10 yards. On first and 10. That gained a first down.

The other 3 rushes were kneels. For some idiotic reason, they got listed in the play-by-play as rushes rather than kneels.

2
by admin :: Wed, 09/13/2006 - 12:21am

Yep, usually I catch that and manually change. Missed it this time. Thanks for the note.

3
by Yaguar (not verified) :: Wed, 09/13/2006 - 12:27am

Vernon Davis as least valuable is the most interesting stat there. I saw only the highlights, so I was under the mistaken impression that he did pretty well.

4
by jim's apple pie (not verified) :: Wed, 09/13/2006 - 12:33am

Did Rivers' 11 attempts put him under a minimum threshold? Where would have been if he was included?

5
by birk (not verified) :: Wed, 09/13/2006 - 12:40am

The Quick Reads are usually written and posted before the Monday Night games. Brunell, B Johnson, Brooks, Rivers, and Walter all go unlisted, so I'm sure this is the reason.

6
by Bobman (not verified) :: Wed, 09/13/2006 - 1:36am

Jeez, I am so tired of seeing Payton Manning's name sandwiched between Rex Grossman and Alex Smith. Enough already, just get them a triple bunk in Canton and be done with it!

Nice Styx reference in Eli's entry, but Aaron, aren't you too young to remember when they were actually sorta cool?

This is a great table, because based on all the glowing praise of Carson Palmer, I'd have assumed he'd be higher.

What I REALLY need to see is where Aaron Brooks falls in the hierarchy. Jerry Porter was IMing me jokes during the game and I missed a few key plays.

7
by Eric (not verified) :: Wed, 09/13/2006 - 1:55am

Did anyone else notice that Brad Johnson threw a left-handed pass on Monday night... and was praised for it by Joe Theismann? If throwing lefty is a bad idea for Plummer, it should also be a bad idea for Johnson, regardless of impending sack, shouldn't it?

8
by Richie (not verified) :: Wed, 09/13/2006 - 2:06am

Nice Styx reference in Eli’s entry, but Aaron, aren’t you too young to remember when they were actually sorta cool?

South Park reinvigorated that song about 10 years ago.

9
by D (not verified) :: Wed, 09/13/2006 - 2:18am

"Players receive bonuses when they play well against good defenses, and they don't get rated as world-beaters when they shred the 49ers."

Not a huge deal, but that should probably be taken out of the "How PAR Works" section at the end until the opponent adjustments are in place.

10
by Noble (not verified) :: Wed, 09/13/2006 - 2:47am

3: I actually watched part of the game and Davis had a reception (the 5 yarder I believe). Based on commentary, you couldn't tell he was having a bad day, let alone worst of all WRs and TEs.

11
by Darrel Michaud (not verified) :: Wed, 09/13/2006 - 3:48am

#6, #8: The song was also in the first episode of Freaks and Geeks.

12
by sm (not verified) :: Wed, 09/13/2006 - 4:26am

Is Eli Manning's -2.7 Rushing PAR entirely from that one fumble?

13
by Kibbles (not verified) :: Wed, 09/13/2006 - 4:28am

Re #7: Did anyone else notice that Brad Johnson threw a left-handed pass on Monday night… and was praised for it by Joe Theismann? If throwing lefty is a bad idea for Plummer, it should also be a bad idea for Johnson, regardless of impending sack, shouldn’t it?
Announcers and columnists are nothing if not inconsistant. If you watch, you'll notice that lots of QBs throw lefty passes from time to time (or righty passes if they're naturally lefthanded). Trent Green threw a left-handed interception two years ago just 3 weeks after Jake Plummer's, in fact. And Plummer had a left-handed TD that season, too. The problem is, that doesn't fit into the stereotype of Plummer being way more inconsistant and mistake-prone than the rest of the league, and Trent Green being oh-so-reliable, so columnists don't run with it.

It shocks a lot of people to hear, but in the season when "Jake the Mistake" tossed 20 INTs as a Bronco... Trent Green had 17 (3rd most in the league, iirc). And yet, somehow, Plummer was a walking mistake and Green was one of the most underrated QBs in the league.

14
by Malene, cph (not verified) :: Wed, 09/13/2006 - 6:58am

so, I don't regret taking Tiki in the PAR fantasy league anymore ;-)

15
by The Aristocrats (not verified) :: Wed, 09/13/2006 - 8:49am

so, um, what do you think about pennington's shoulder?

anybody know what his rating was after that game?

16
by billvv (not verified) :: Wed, 09/13/2006 - 10:43am

Re: Pennington. You'd think you'd hear more chomping on crow than you will. Nobody will admit they were wrong but you'll sure hear about their successful picks!

17
by Slippery Pete (not verified) :: Wed, 09/13/2006 - 10:55am

#13: LOL! Teh anser is obvious. Trent Green has swagger, heart, and nver throws MOMENTUM killing INTs. That's the key. Avoid MOMENTM killng INTs. And any rtard nows that left handed passses cna only wrk if it comes from a sorce of grt swegger.

On a slightly more serious note, criticizing Jake Plummer for throwing a left handed pass in Sunday's game has to be the stupiest thing to criticize him for. He actuallly completed his left handed pass! It's not like it was an interception. Of the endless things to criticize about him, a completed left handed is actually probably worthy of praise instead of criticism.

18
by Led (not verified) :: Wed, 09/13/2006 - 11:58am

billv: To paraphrase the immortal Winston Wolf, let's not start sucking Pennington's stick yet. I hope he keeps it up (and I do think he'll continue to play well), but it's a long season and not every team will be covering Coles with Reynaldo Hill.

19
by Wanker79 (not verified) :: Wed, 09/13/2006 - 12:12pm

Adding onto 18:

Plus it seems kinda foolish to be all upset that people somehow thought that a QB coming back from 2 rotator cuff surgeries in consecutive years may somehow have problems returning to form.

If Pennington goes through the year with no ill effects of his shoulder problems, good for him (and I really hope he does since the J-E-T-S, Jets, Jets, Jets are my AFC squad). But why is it such a travesty that, especially before the season, people doubted that it'd happen?

20
by paytonrules (not verified) :: Wed, 09/13/2006 - 12:24pm

Curious. You mention Ahman Green's quality DPAR - but almost half his yardage came in the 4th quarter when the Bears were playing the pass exclusively. In the first quarter he couldn't convert first downs when they were needed, hence the early three-and-outs.

Aren't these metrics supposed to account for that? Or do they only account for the success/failure of a play regardless of game context. Green did get a few first downs, but they were largely meaningless.

21
by Kal (not verified) :: Wed, 09/13/2006 - 12:37pm

I'm waiting until week three before eating crow. That should give at least some ability to be statistically relevant. I mean, come on - the Texans and the Titans?

22
by rk (not verified) :: Wed, 09/13/2006 - 1:56pm

I think Charlie Batch's rush PAR seems low. He ran for 12 on 3rd and 13 (they then converted on 4th and 1) and 11 on 1st and 10. He had 4 kneel downs and a run ob at the LOS that I believe counted as a sack.

23
by CA (not verified) :: Wed, 09/13/2006 - 1:58pm

Re: 7, 13

Against the Bears, Favre nearly attempted a pass while sitting on the ground. In response, the announcers laughed about how much fun he was having out there.

24
by Crushinator (not verified) :: Wed, 09/13/2006 - 2:17pm

Re; Chad Pennington

Chad Pennington playing well isn't a surprise. He's a good QB. The Jets were a fringe playoff team going into last year before taking 59830298 injuries to their QB position. He should look competitive at least.

The real worry is just if he has any range left. I remember early last year, the Jets had gloated during their playoff run he really had a season-ending injury but he managed to play through it. Then last year, all his passes seemed to lose 15 yards.

The Titans aren't a good test.

25
by MRH (not verified) :: Wed, 09/13/2006 - 2:26pm

It shocks a lot of people to hear, but in the season when “Jake the Mistake� tossed 20 INTs as a Bronco… Trent Green had 17 (3rd most in the league, iirc). And yet, somehow, Plummer was a walking mistake and Green was one of the most underrated QBs in the league.

Plummer had a pretty good year in 2004 despite the 20 INTs. He threw for 27 TDs and 7.8 ypa, good numbers. His comp pct was only so-so at 58.2%. DPAR was 72.5 and DVOA 11.5%.

Green matched Plummer's TD numbers. Since Plummer threw 30-some fewer passes, that means Jake was throwing TDs at a higher rate than Green. Of course, it also means those 20 INTs were at a higher rate than Green's 17. Also, Green threw for 8.3 YPA and 66.4% comp - I think the higher comp rate is significant since the Vermeil/Martz/Saunders offense is typically a more down-the-field passing game than the WCO, Shanahan version. DPAR/DVOA seem to agree, as Green's 114.3/27.7% were much better than Plummer's.

For their careers, Green's numbers look better than Plummer's.

Green: 60.8% - 7.7 ypa - 150 td - 93 int
Plummer: 57.2% - 6.7 ypa - 150 td - 151 int. (Plummer has about 700 more career att).

Over the six years of DPAR/DVOA stats, Green has had significantly better stats three times, Plummer once, and two years they were close.

I'd say Trent Green has been a better QB, and under-rated, but I am a Chiefs fan. Oh, and Green certainly took plenty of critcism for his 24 INT season in 2001, at least in the KC press.

And, Plummer DOES get unfairly criticized. He certainly kills the Chiefs at Mile High (insert corporate name here), especially with bootleg plays. I would love to see DEN go with Cutler for this week's game. It won't happen since unfortuantely Shanahan is smarter than the press critics of Plummer.

26
by Dennis (not verified) :: Wed, 09/13/2006 - 2:36pm

Re #24 The real worry is just if he has any range left.

He threw 3 or 4 deep passes and they were all right on the money. That's what was so encouraging about his performance.

27
by Will Allen (not verified) :: Wed, 09/13/2006 - 2:47pm

Maybe I missed it, but qb PAR or DPAR doesn't yet adjust for drops, does it? Perhaps this isn't feasable, but I thought of this while watching the Vikings game, when, if Troy Williamson had just caught his first two drops (they should have been routine catches, while the third was more difficult) Johnson would have had about a 7% higher completion percentage, about eighty yards more, and more than likely another td throw. It would be interesting to see Johnson's PAR for that game, and his PAR adjusted for those two drops. Then again, PAR can't adjust for bad passes caught due to a tremendous effort by the receiver, so maybe it all comes out in the wash.

28
by Will Allen (not verified) :: Wed, 09/13/2006 - 2:51pm

I wish nothing but the best for Pennington, but the list of guys who made their living with their shoulders, had rotator cuff procedures, appeared to be recovered, and then had everything go to hell again, is kinda' long. I wonder if they are cutting back his throws during the week.

29
by RW (not verified) :: Wed, 09/13/2006 - 3:26pm

Re: #13 - I've watched every Chiefs game for the last 5 years at least - can you give me a cite on when Trent Green threw a lefty INT. Because that's one thing that I'd definitely remember.

30
by Wanker79 (not verified) :: Wed, 09/13/2006 - 4:20pm

Re:27

Giving a QB credit for a dropped pass probably wouldn't be right. But maybe that play could just be completely ignored for the purposes of QB stats. In other words, treat those first 2 Williamson drops like then never happend. Johnson attempted 2 fewer passes that night; Johnson had two fewer plays counting against him as unsuccessful. The drops would definitely still count against Williamson, and I'd think they'd probably still count against the Vikings' offensive efficiency, just not against Johnson.

31
by rk (not verified) :: Wed, 09/13/2006 - 5:40pm

It's a "Jump to Conclusions mat". You see, you have this mat, with different CONCLUSIONS written on it that you could JUMP TO.

32
by jetsgrumbler (not verified) :: Wed, 09/13/2006 - 5:52pm

ok, i know it was only a win over lowly titans, so i shouldn't get carried away. but the JETS definitely looked better than i expected.

i am all for your advanced statistics, but please explain to me how JETS and Titans played one game against each other, the JETS won, and they are ranked 9 spot lower in DVOA?

after last week, i don't care what any stats say, the JETS are at least better than Titans, Oakland, and Green Bay.

33
by jetsgrumbler (not verified) :: Wed, 09/13/2006 - 6:13pm

my bad, i saw that you have since put out week 2 dvoa that puts things in order.

SORRY!

34
by Wanker79 (not verified) :: Wed, 09/13/2006 - 6:17pm

Re: 32

You should probably have noticed that the column you're referring to is titled "Proj + Week 1" and that the very first paragraph of the Week 2 DVOA page states...

"This year we introduce a new feature of the DVOA ratings: An early-season rating that combines the preseason projection with the results of early games to give us a better prediction of how each team will rank at the end of the year. Can you guess how much of this rating is based on the first week’s games?"

The reason the Jets are ranked 31st while the Titans are ranked 22nd is because the Jets were projected to be abismal this year while the Titans were only projected to be merely very bad.

If you look at the column titled "Total VOA" (which takes into account only Week 1 results and are not adjusted for opponent), the Jets are 7th with +36.9% VOA and the Titans are 28th with -53.9% VOA

35
by Matt Weiner (not verified) :: Wed, 09/13/2006 - 6:35pm

Re: 22 I think Charlie Batch’s rush PAR seems low.

I had the same thought, but I'm guessing that the goal-line fumble counts against his rush DPAR. That'll hurt ya. Gamebook has him listed with 7 rushes; 4 were kneels, 2 the scrambles you mentioned, I think the seventh was the fumble (not running OB at the LOS which it looked like they counted as a sack).

36
by DWL (not verified) :: Wed, 09/13/2006 - 7:09pm

If you’re going to remove the drops from calculations, then perhaps we should eliminate the following as well:

• when the receiver mistakenly runs an “in� pattern while the QB throws “out�

• when the ball soared on the QB due to a random factor such as slipping out of his hand,

• when the QB gets hit just as he throws and the ball falls incomplete

• when “some other random out of the QB’s control event� occurs

Eventually all QBs will complete ~100% of their passes (I’m sure Theisman would buy off on this system).

What should happen with a QB is

• either the receiver finally figures out how to catch the passes thrown to him (or he gets benched), or

• the QB figures out that maybe not throwing as often to said WR is a good idea and said receiver should be the final option in his reads

This growth/insight on the QB’s part should result in an improved rating over time and the QB who doesn’t figure this out, probably doesn’t deserve a better rating.

37
by Sophandros (not verified) :: Thu, 09/14/2006 - 10:36am

Announcers and columnists are nothing if not inconsistant. If you watch, you’ll notice that lots of QBs throw lefty passes from time to time (or righty passes if they’re naturally lefthanded). Trent Green threw a left-handed interception two years ago just 3 weeks after Jake Plummer’s, in fact.

Brett Favre threw a pass backward in preseason, and nothing was said, even though it resulted in a turnover. At least Bill Simmons has said that his recent play has made him into the white Aaron Brooks.

38
by Wanker79 (not verified) :: Thu, 09/14/2006 - 11:54am

Re: 36

I don't see why it's so ridiculous (like most of the examples you cited) to not count perfectly thrown passes against a QBs ratings. Afterall, if the whole point of DVOA and DPAR is to try to put numbers to the value and production of a QB, penalizing him for making a perfect throw doesn't really seem in line with that goal. The ball slipping out of a QB's hands is his fault. Getting hit as you release is the QBs fault (they should have better pocket-presence). A wideout letting a perfectly throw ball slip right through his hands is not the QBs fault.

It's definitely a deficiency of the WR, and it's definitely a deficiency of the total Offense's efficiency. And it should definitely still count against those two ratings. And I'm not saying that it's some travesty that those first two Williamson drops count against Brad Johnson. I'm just suggesting that it may be something to look at that may improve the system. Is that really such a bad thing?

39
by Jets fan (not verified) :: Thu, 09/14/2006 - 12:45pm

Hey #32 -- I'm right there with you. very pleasantly surprised. Barlow looks like he might just become a force to be reckoned with, an offensive line with 2 rookies gave up zero sacks. Pennington put on a clinic, at one point 9 out of 10... and the defense looked like they came prepared and well adjusted to the 3-4. And you can't give the pundits too much credit. They all said the bucs would beat the ravens, they all said the panthers would win, and they all said pennington would be weak. They were 100% wrong right across the board.

40
by DWL (not verified) :: Thu, 09/14/2006 - 3:14pm

RE: 38
A totally independent number representing QB effectiveness (i.e. removing all receiver error/variance) is theoretically possible (?); however, QBs throw to receivers on almost every play, so QB performance is never truly independent of receiver and other outside (ridiculous or not) factors/variance.

If drops are going to be eliminated from the QBs’ rating, perhaps all balls that were virtually uncatchable, but were miraculously (or spectacularly) caught by the WR, should be eliminated from the QB’s reception total, as the receiver is the one who contributed more too the play’s success. Similarly, a throw that is tipped by a defender, but ends up in the receiver’s hands should be removed as well. Neither of these was perfectly thrown, but each was caught anyway (sort of the anti-drop). Also, in situations where a defender drops an easily intercepted ball or the WR strips the ball from the defender preventing an interception, the pass should be counted as an interception against the QB for ratings purpose.

If receiver play is going to be factored in, eliminating the receivers’ drops oversimplifies the problem; instead, a value adjustment for receiver quality (in much the same way strength of defense is factored into other ratings) for each receiver would be more appropriate.

41
by Wanker79 (not verified) :: Thu, 09/14/2006 - 4:10pm

perhaps all balls that were virtually uncatchable, but were miraculously (or spectacularly) caught by the WR, should be eliminated from the QB’s reception total

Similarly, a throw that is tipped by a defender, but ends up in the receiver’s hands should be removed as well.

I'd agree with both of those adjustments. More so with the latter than the former because I guess it's possible to argue in many cases that the wideout put himself in the position where a spectacular catch was necessary (ie, the ball was in the right place but the WR ran a poor route...etc). But there's no way a QB should be credited for anything positive resulting from a pass being deflected at the line (negative results are a completely different story, though since stuff like identifying throwing lanes and pass release point fall on the QB).

I really like your suggestion of a value adjustment factor. You would have to limit the adjustment to only account for drops that are irrefutably the WRs fault. It would probably end up being fairly insignificant in most cases, but seeing Johnson's ratings penalized for those first 2 Williamson drops is just annoying me for some reason (and I an Eagles fan who really couldn't care less about the Viking or Brad Johnson).

42
by DWL (not verified) :: Fri, 09/15/2006 - 10:08am

The definition of “irrefutably the receiver’s fault� (ItRF) introduces a subjective measurement, which opens the door to increased error variance to the QB rating equation. A more efficacious strategy would be to have as unbiased as possible receiver adjustment that adjusts the rating based on what receiver is thrown to. This “receiver adjusted value� would take into account the likelihood that the receiver makes a catch of any ball thrown in his direction (i.e it would account for ItRF drops, for receivers who are more/less prone to being separated from the ball when hit, poor route running, etc.).

In related news, what about YAC? In some instances YAC is due to the QB’s ability to perfectly deliver the ball (dreaded W. Coast offense), but in others, it is the effort of the receiver (maybe Reggie Bush someday) that makes the play work.

43
by Wanker79 (not verified) :: Fri, 09/15/2006 - 10:31am

I would only limit it to ItRF because there's really no reliable way to determine if the QB threw to the wrong route or the WR ran the wrong route (for example). I would err on the side of caution in cases where it isn't blatantly obvious, and still count those plays against everyone.

As for worrying about adding a hint of subjectivity, correct me if I'm wrong but I'm pretty sure that the game-charting project accounts for alot of pseudo-subjective results.

44
by B (not verified) :: Fri, 09/15/2006 - 12:12pm

The game charting project lists each incomplete/interecepted pass as Overthorwn, underthrown, thrown ahead, thrown behind, defensed, dropped, thrown away, tipped at line, out of bounds, alligator arms, WR tripped an fell, Hit in Motion, and Miscommunication (the last one would be for wrong route).

Speaking of the game-charting project, I charted the 2nd half of the Bills/NE game last night, and I just want to post something about the Bills Playcalling. On the series with the safety, both pass plays had the QB drop 8 yards to the end zone. It wasn't a case of Losman scrambling backwards to avoid pressure, he dropped back as soon as the ball was snapped. I don't know if it was play-calling or just a really bad descision by Losman, but it happned twice, and with Losman throwing the ball away to avoid a sack on first down, you'd think the coaches would call a shorter drop, or a bootleg or something on 3rd down.

45
by Wanker79 (not verified) :: Fri, 09/15/2006 - 3:33pm

I think I'm going to send this little discussion over through the contact form and see what Aaron thinks. Either way, it'll be interesting.

46
by Sean_C (not verified) :: Sat, 09/16/2006 - 1:17am

Is PAR based on the game charting project or on the NFL official play by play? I thought the official NFL material, but maybe things have changed?

It would be interesting to see how PAR results calculated using both methods might differ over the course of a season. I'm assuming, of course, that PAR calculations based on the game charting strive to split success points between QB and reciever on a situational basis and not some fixed ratio.
If a reciever lets a perfectly thrown, catchable ball bounce off his hands and into the arms of a defender (a la Braylon Edwards versus the Saints), he should get the lions share of the penalty points (not Charlie Frye, who may only deserve 10% of the penalty for putting a bit too much zip on the throw).

RE the game charting project: do we get to see some mid-season results at some point, or does it all get saved up for the next Prospectus?

I definitely like what this site is doing - keep at it guys!