PFF on the Niners-Bengals, Smith

[Z-Comeback]

Alex Smith has right read on Philly defense - NFC West Blog - ESPN

OK, then. Next question: How has Smith fared when teams rush four or fewer defenders relative to when they send more? And what about when teams send DB pressure?

So far this season, Smith ranks 21st in NFL passer rating (81.2) and 25th in Total QBR (43.7) against four or fewer rushers. His NFL rating is 115.0 with an 80.1 QBR against five or more.

Smith is a sensational 8 of 8 passing for 106 yards, one touchdown and perfect ratings -- both NFL and QBR -- when teams rush at least one defensive back this season. But the Eagles have pressured with their secondary only six times through their first three games, so Smith might have few opportunities to build on his fast start in those situations.

The Eagles' change in approach would appear to work against Smith overall.

After watching the video of the 1st half offense vrs Bengals on this thread

http://www.sportshoopla.com/forums/san-francisco-49ers/46801-video-9ers-offensive-plays.html

I tried to pay attention to which plays the Bengals only rushed 4.

(2:00 in to the video) Smith hits Davis on the wheel route as the Bengals only send 4 and protection holds up nicely.

(2:18) they line up in Shotgun with 5 Wide I believe, the Bengals send 4, protection seems to hold up nicely but Smith for some reason decides to immediately dump off to Hunter for a minimal gain. Poor decision here?

(3:00) 3rd and 5, the Bengals send 4 but Niners OL blows it with a false start

(3:14) Very next play (3rd and 10) the Bengals send 4 again, protection breaks down, Smith does a nice job of avoiding the pressure and looks to head up field but then seems hesitant and bounces a pass to an open receiver.

(3:50 on video), Bengals send 4 and Smith throws a nice lob to Davis down the sideline which Davis then drops.

(4:37) 3rd and 5, Bengals send 4, looks like Smith reads the play well and gets the ball out on time...but makes a very poor throw to Morgan on the out.

(6:43) 3rd and 14 after an Anthony Davis penalty, Bengals send 4, protection doesn't do very well, Smith is sacked, but he did have enough time to get rid of the ball.

(8:15) 1st and 15 after another false start, Bengals send 4 and Smith dumps off again but bounces it incomplete.

(8:34) 3rd down and 8, Bengals send 4, Smith hits Hunter short, just as protection is collapsing and Hunter turns up field and shoves ahead for the first down.

(9:00) 1st Down, Bengals send 4, Smith hits Crabtree for 8 yards along the sideline.

Shortly after this on 3rd and short yardage, the Bengals send 4 and the Niners decide to run up the middle with Gore. Didn't really care for that play call but oh well.

 

[Crimsoncrew]

I think you are right but not being able to see the math makes it hard to put much faith in a formula.

That's fair. I don't know all that much about it, but that's true of a lot of these statistical measures, including PFF's system, FootballOutsider's DPAR, etc. They aren't gospel and shouldn't be taken as such, but they can shed some light on things. I haven't scrutinized the Cincy game, but it seemed like they didn't blitz all that much. I know that Smith struggled much more against a four-man rush than against the blitz against Dallas, as Dallas backed off the blitz in a big way in the second half.

 

[Ray_Dogg]

NFL - Total Quarterback Rating shifts the way we see the position - ESPN

Enjoy.

 

[MW49ers5]

Of course it will not show in the passer rating because it is different from the QBR rating...


"... QBR allocates the points added by every play in an NFL season to each of the players involved, every play. On completed passes, for example, it splits credit among QBs, receivers and blockers, depending on factors such as whether the quarterback was under duress, where he threw the ball, how far it traveled and how many yards the receiver gained after the catch. QBR splits the blame for sacks on quarterbacks and offensive linemen and attributes QB fumbles to QBs. Further, QBR weights every play by its clutch value -- its contribution to a team's chances of winning, given the score of a game, not just to scoring points..."

This provides some insight - Thanks, Four!

The part about 'clutch value' probably played a big role in his numbers due to the Dallas game. I hope the formula for this mysterious rating is published at some point. Thus far, it sounds like a highly subjective rating.

 

[MW49ers5]

Fixed. Good news for you, you'll be waiting for a long time.

As usual, you criticize but offer nothing constructive. It's a pretty tired act.

You want some constructive criticism, here is some constructive criticism:

Pull your head out of your ass BEFORE you post!

There, there is your constructive criticism.

Now, if you'll excuse me I have to go play with my dick; but, don't worry, I'll be thinking of you the whole time.

 

[sayheykid1]

Of course it will not show in the passer rating because it is different from the QBR rating...


"... QBR allocates the points added by every play in an NFL season to each of the players involved, every play. On completed passes, for example, it splits credit among QBs, receivers and blockers, depending on factors such as whether the quarterback was under duress, where he threw the ball, how far it traveled and how many yards the receiver gained after the catch. QBR splits the blame for sacks on quarterbacks and offensive linemen and attributes QB fumbles to QBs. Further, QBR weights every play by its clutch value -- its contribution to a team's chances of winning, given the score of a game, not just to scoring points..."

That sounds pretty subjective and we have no idea how sacks are factored in.

 

[sayheykid1]

That's fair. I don't know all that much about it, but that's true of a lot of these statistical measures, including PFF's system, FootballOutsider's DPAR, etc. They aren't gospel and shouldn't be taken as such, but they can shed some light on things. I haven't scrutinized the Cincy game, but it seemed like they didn't blitz all that much. I know that Smith struggled much more against a four-man rush than against the blitz against Dallas, as Dallas backed off the blitz in a big way in the second half.

I can agree with that.

 

[MW49ers5]

NFL - Total Quarterback Rating shifts the way we see the position - ESPN

Enjoy.

Thanks Ray!

So, according to the video, this QBR is base 100 and comes with the following interpretations.

A rating over 75 is considered MVP candidate.
A rating between 65-75 is considered a Pro-Bowl candidate.
A rating between 35-64(?) is considered average.
A rating below 35 is terminable.

 

[Flyingiguana]

sometimes there is this thing called overanalysis

 

[Crimsoncrew]

You want some constructive criticism, here is some constructive criticism:

Pull your head out of your ass BEFORE you post!

There, there is your constructive criticism.

Now, if you'll excuse me I have to go play with my dick; but, don't worry, I'll be thinking of you the whole time.

I don't doubt it.

 

[MW49ers5]

sometimes there is this thing called overanalysis

Analysis Paralysis - is what it is called, you are 100% correct

 

[Z-Comeback]

That sounds pretty subjective and we have no idea how sacks are factored in.

Can't remember where I read it but I believe they know that the sack rating is insufficient and that adjustments will be made.
But based on that admission alone, it kind of throws the whole QBR idea into question.

 

[SY8goat]

100% wrong!

That supports the fact that sacks (especially those taken by QB's who don't correctly read & react to the blitz) are not considered as part of the QB rating. Think Aaron Rodgers in '09. I'll bet his 'five or more' rating was just as high.

This is from one of Sando's blog posts today:

So far this season, Smith ranks 21st in NFL passer rating (81.2) and 25th in Total QBR (43.7) against four or fewer rushers. His NFL rating is 115.0 with an 80.1 QBR against five or more.

That would tend to support the idea that the bigger problem is the OL making the correct blocks, not managing the blitz. I'm certainly not thrilled with his work against the blitz, but I think the far bigger problem is our OL simply getting outplayed.

Why the heck are guys talking about the blitz when they are only RUSHing 4? No one is "blitzing."

How many rush the passer in a 4-3 base D? 4

How many rush the passer in a 3-4 base D (depending on the scheme)? Usually 4, right?


One stat I would like to know, is Smith's completion percentage against 7 in coverage (4 rushers).

 

[Crimsoncrew]

Why the heck are guys talking about the blitz when they are only RUSHing 4? No one is "blitzing."

How many rush the passer in a 4-3 base D? 4

How many rush the passer in a 3-4 base D (depending on the scheme)? Usually 4, right?


One stat I would like to know, is Smith's completion percentage against 7 in coverage (4 rushers).

The first numbers reflect what Smith has done against a four-man rush. The second numbers reflect what he has done against five or more. The latter figures represent his numbers against the blitz.

 

[SY8goat]

The first numbers reflect what Smith has done against a four-man rush. The second numbers reflect what he has done against five or more. The latter figures represent his numbers against the blitz.

I know what the numbers reflect.

What I was trying to say: Is why are you dwelling on the blitzs when the stats say Smith, the receivers, and the O-line are having more issues with the 4 man rush. No blitzs...

 

[Z-Comeback]

Why the heck are guys talking about the blitz when they are only RUSHing 4? No one is "blitzing."

How many rush the passer in a 4-3 base D? 4

How many rush the passer in a 3-4 base D (depending on the scheme)? Usually 4, right?


One stat I would like to know, is Smith's completion percentage against 7 in coverage (4 rushers).

I posted results of plays in the first half against the Bengals where they only rushed 4 in this thread...it is lost in between the bickering

 

[Crimsoncrew]

Why the heck are guys talking about the blitz when they are only RUSHing 4? No one is "blitzing."

How many rush the passer in a 4-3 base D? 4

How many rush the passer in a 3-4 base D (depending on the scheme)? Usually 4, right?


One stat I would like to know, is Smith's completion percentage against 7 in coverage (4 rushers).

I know what the numbers reflect.

What I was trying to say: Is why are you dwelling on the blitzs when the stats say Smith, the receivers, and the O-line are having more issues with the 4 man rush. No blitzs...

MW and I had a disagreement about how well Smith handled the blitz against the Cowboys. He's pouty because these numbers tend to support my argument.

 

[clyde_carbon]

They're part of the formula, and the main reason Smith's QBR has been below average the past two weeks. I don't know the exact formula. I'm sure you can learn all about it with a google search or two.

Does it also factor in YPA on blitzes? Underneath routes are usually open when a team is blitzing, so unless Smith is actually making the other teams pay for blitzing, and not just throwing 3 or 5 yard passes, then this is all irrelevant.

 

[Crimsoncrew]

Does it also factor in YPA on blitzes? Underneath routes are usually open when a team is blitzing, so unless Smith is actually making the other teams pay for blitzing, and not just throwing 3 or 5 yard passes, then this is all irrelevant.

I don't know. I haven't had the time to really look into the QBR. Its main thrust is the likelihood of a win. I can't imagine your win likelihood changes much if you're facing 4th and 15 after a sack versus 4th and 5 after a short completion. On first or second down, taking the three or five yards isn't a terrible play. I'm fairly certain it does not take into account things like missing an open man.

The overarching point that I was getting at last week was that this is Smith. Getting anything against a blitz is an unexpected surprise. Getting several first down throws and a TD is all kinds of gravy. When the Cowboys realized they could get as much pressure with four guys as they could with five or six - primarily because they were dominating the one-on-one matchups up front - they were able to shut our offense down.

 

[clyde_carbon]

I don't know. I haven't had the time to really look into the QBR. Its main thrust is the likelihood of a win. I can't imagine your win likelihood changes much if you're facing 4th and 15 after a sack versus 4th and 5 after a short completion. On first or second down, taking the three or five yards isn't a terrible play. I'm fairly certain it does not take into account things like missing an open man.

The overarching point that I was getting at last week was that this is Smith. Getting anything against a blitz is an unexpected surprise. Getting several first down throws and a TD is all kinds of gravy. When the Cowboys realized they could get as much pressure with four guys as they could with five or six - primarily because they were dominating the one-on-one matchups up front - they were able to shut our offense down.

So:

EPA: Expected points added (could be positive or negative) that occur as a result of an action play.

Pass EPA: Clutch-weighted expected points added by the QB through pass attempts.

Run EPA: Clutch-weighted expected points added by the QB through scrambles, designed rushes and fumbles/fumble returns on running plays.

Sack EPA: Clutch-weighted expected points added by the QB through sacks and fumbles/fumble returns on passing plays.

Penalty EPA: Clutch-weighted expected points added by the QB on penalties.

Total EPA: Total clutch-weighted expected points added by the QB.

Action Plays: Plays on which quarterback has a non-zero expected points contribution. Includes most plays that are not handoffs.

Clutch Wt Avg: Average clutch weight on QB action plays. Smaller than 1 means QB was in relatively fewer clutch situations; greater than 1 means he was in more.

Total QBR: Total Quarterback Rating, which values quarterback on all play types on a 0-to-100 scale.

Looking at these scales in a quick glance, I'm not sure that using the QBR is the best method to calculate how well Smith plays under the blitz.

Can we get a comparison with other QBs?