Brady hit once every 8.5 attempts since mid 07

[Box_O_Rocks]

responding to you and Box. Great questions! Here's what the data say:

I looked at the 12 teams who made the playoffs in the 2007/08 season. In total, the sack rate for the 12 principal starters (the Redskins had two different starters for most of the regular season and the playoffs) was virtually identical for the Playoffs as for the regular season.

Regular Season: 5.1%
Playoffs: 5.5%

When you look a little more deeply at the data, of the 12 starters, Brady's Playoff stats stand out like a sore thumb. His Playoff sack rate (eight sacks) of 6.8% was 95% higher than his regular season rate of 3.5%, while the overall Playoff Average (above) was just nine percent higher than the Regular Season. The only QB's to come close were Vince Young, whose playoff stats were 52% higher and Eli Manning whose playoff stats were 44% higher. Phillip Rivers and Jeff Garcia were sacked at rates that were roughly half of their regular season rates. Payme wasn't sacked at all in the postseason. Favre was only sacked once.

So, I conclude that the postseason looks about the same as the regular season in general when it comes to sacks. Some QB's are sacked more in the Playoffs, but the fact that Brady's increase vs. the regular season (and especially vs. his experience of a 2.2% rate in the first ten games of the season, where his playoff rate is nearly three times as high) is an outlier really can't be disputed based on the data.

NFL Stats: by Player Position

2007 Postseason

QB                  # sacks  # games played  per game avg  high game  fumbles
Eli Manning            9            4            2.25          3 (2)      2
Tom Brady              8            3            2.67          5          1
Ben Roethlisberger     6            1            6             6          1
David Garrard          5            2            2.5           5          1
Todd Collins           3            1            3             3          1
Matt Hasselbeck        3            2            1.5           2          2
Vince Young            3            1            3             3          0
Philip Rivers          2            3            0.66          1          1
Tony Romo              2            1            2             2          0
Brett Favre            1            2            0.5           1          0
Jeff Garcia            1            1            1             1          1
Peyton Manning         0            1            0             0          0

I've added Fumbles because they are a recorded measure of contact, they may have been strip sacks, but at least it's a consistent measurement from NFL.com stats.

What we see is one game makes a difference. Brady had 5 sacks and a fumble against NYG. Roethlisberger had 6 sacks and a fumble against Jax. Garrard had 4 sacks against Pit. Eli had 3 sacks twice, but his high was 3 sacks and 2 fumbles against NE.

What it comes down to is match-ups, NYG had the personnel to exploit NE's OL, but please note they also had the advantage of having played them at the end of the regular season and Tom Brady was not as mobile as he usually is with a foot injury.
-- Does Stephen Neal staying viable for the entire game make a difference? I think it does. (Does the 100% turnover of the interior OL reflect that, I'd say yes.)
-- Does a healthy Brady make a difference? Yes. And yes there's a catch-22 here, if he is getting hit more, he's going to be playing hurt and he'll be hit more.
-- Note also that NYG's DL sold themselves out to keep that pressure up. On NE's last scoring drive there was a Giant DL sitting on the ground unable to get up after nearly every play - I think the coaches failed to eat enough clock, but the key point here is the intensity of the games goes up later in the season and into the playoffs as teams compete for a playoff berth and to survive in a one and done playoff. Curran is not accounting for the intangibles of late season/playoff intensity.

Let's check one more source for data: Tom Brady: Career Stats

If you were to graph the number of sacks Brady took over his career, there would be a clear downward trend, this despite his steadily increasing passing yards/game.

Conclusion: Brady is better protected by the combination of his O-line, his rapport with his receivers, and by his own experience than Mr. Curran and like-minded statistic-mangling worry warts can be bothered to see. /Panic-patsfans mode

 

[patchick]

it's a good idea to protect tommy and let's kick some butt this season.

I disagree! :enranged:

 

[PatsFanSince74]

...
Conclusion: Brady is better protected by the combination of his O-line, his rapport with his receivers, and by his own experience than Mr. Curran and like-minded statistic-mangling worry warts can be bothered to see. /Panic-patsfans mode

great stuff and I appreciate the effort that went into pulling it together. if i had the time after spending so much time on this today, i'd try statistically to reconcile the two story lines that are emerging; in any event, there are probably too little data to make an analysis of standard deviations and correlations meaningful here.

you're rightly observing that some of this is situational; the article and data that I presented ("mangled" to use your oh so kind phrase) are looking at a trend line. despite your language, I'll acknowledge that what you have written is very insightful, because statistical analysis looks both at events and trends. together they tell the story.

for now, this "worry wort" (another so pleasant phrase that you chose to use) is going to stick to his guns and say that he believes that the trend described in the article and validated by the data he analyzed is worrisome while also acknowledging that there's a lot of noise around the distribution of the data when you look at individual games, as you observe. it's the problem of any analysis with relatively few data points.

for me, this means that i will have a bit more concern for Tom Brady's health and protection this year than I did before this morning. Not "panic" (another one of your digs), but concern. I think its reasonable and warranted.

(BTW, can't we just have conversations like this without resorting to ad hominem denigration with terms like "mangling" and "worry wort" and "panic?". personally, i have found this to be a very good back and forth by reasonable people--yourself, Deus and Patchick--who are looking at the same data through different lenses. Why can't we just leave it at that without the name calling?)

 

[Oswlek]

I think it is pretty apparent BB got the message about protecting Brady considering they turned over every single interior OL backup, added a swing tackle, traded for a blocking machine at tight end and has kept at least 4 viable runners on the team since the SB. This offense won't be nearly as pass dependent as 2007 was.

Box, you are always good for some smarts. Thanks.

 

[DaBruinz]

NFL Stats: by Player Position

2007 Postseason

QB                  # sacks  # games played  per game avg  high game  fumbles
Eli Manning            9            4            2.25          3 (2)      2
Tom Brady              8            3            2.67          5          1
Ben Roethlisberger     6            1            6             6          1
David Garrard          5            2            2.5           5          1
Todd Collins           3            1            3             3          1
Matt Hasselbeck        3            2            1.5           2          2
Vince Young            3            1            3             3          0
Philip Rivers          2            3            0.66          1          1
Tony Romo              2            1            2             2          0
Brett Favre            1            2            0.5           1          0
Jeff Garcia            1            1            1             1          1
Peyton Manning         0            1            0             0          0

I've added Fumbles because they are a recorded measure of contact, they may have been strip sacks, but at least it's a consistent measurement from NFL.com stats.

What we see is one game makes a difference. Brady had 5 sacks and a fumble against NYG. Roethlisberger had 6 sacks and a fumble against Jax. Garrard had 4 sacks against Pit. Eli had 3 sacks twice, but his high was 3 sacks and 2 fumbles against NE.

What it comes down to is match-ups, NYG had the personnel to exploit NE's OL, but please note they also had the advantage of having played them at the end of the regular season and Tom Brady was not as mobile as he usually is with a foot injury.
-- Does Stephen Neal staying viable for the entire game make a difference? I think it does. (Does the 100% turnover of the interior OL reflect that, I'd say yes.)
-- Does a healthy Brady make a difference? Yes. And yes there's a catch-22 here, if he is getting hit more, he's going to be playing hurt and he'll be hit more.
-- Note also that NYG's DL sold themselves out to keep that pressure up. On NE's last scoring drive there was a Giant DL sitting on the ground unable to get up after nearly every play - I think the coaches failed to eat enough clock, but the key point here is the intensity of the games goes up later in the season and into the playoffs as teams compete for a playoff berth and to survive in a one and done playoff. Curran is not accounting for the intangibles of late season/playoff intensity.

Let's check one more source for data: Tom Brady: Career Stats

If you were to graph the number of sacks Brady took over his career, there would be a clear downward trend, this despite his steadily increasing passing yards/game.

Conclusion: Brady is better protected by the combination of his O-line, his rapport with his receivers, and by his own experience than Mr. Curran and like-minded statistic-mangling worry warts can be bothered to see. /Panic-patsfans mode

:youtheman::youtheman::youtheman::youtheman:

 

[RayClay]

-- Note also that NYG's DL sold themselves out to keep that pressure up. On NE's last scoring drive there was a Giant DL sitting on the ground unable to get up after nearly every play - I think the coaches failed to eat enough clock, but the key point here is the intensity of the games goes up later in the season and into the playoffs as teams compete for a playoff berth and to survive in a one and done playoff.

Hmmm......

 

[maverick4]

http://www.patsfans.com/new-england...tom-brady-most-hit-qb-league-2006-2007-a.html

http://www.patsfans.com/new-england...un-used-more-than-half-snaps-2007-2008-a.html


It doesn't take a genius to know that when a defense has a good idea about what you'll be doing, that they have a better chance of hitting your QB (or, stacking the box against a sure run, like anytime Maroney is in the game)

 

[maverick4]

You can think that the data says that I am wrong, but that isn't reality. In fact, the reality is that the Data provided shows only ONE thing. That Brady got sacked at twice the rate in the play-offs than he did in the regular season. And that stems from ONE game where the O-line gave up 5 sacks. .

Your dismissal of PF74's argument is as silly as saying, 'the 2001 Rams statistically had an amazing offense in both regular season and playoffs, IF you take out their last game where they got stuffed by the Patriots in the Superbowl'. Obviously the 2001 Patriots picked apart their scheme/offense, knew what they were doing on many plays/formations, and took them out of their element whether it be the inside slants, or using Faulk in the passing game. It's also as silly as saying 'Peyton Manning's 01-04 Colts offenses didn't have any problems, if you take out their last playoff game losses to the Patriots'. Obviously, they did have problems because the Patriots figured out their offense, stopped it, and they stubbornly refused to change and ended up scratching their heads how they kept losing.

The 2007 Giants may have already had a potent defensive line, but it was made even better because they didn't have to worry about the run anytime we were in shot gun, which we ran over 50% of the time the past three seasons. Their pass rush was made better because we were so predictable, like the 01-04 explosive Rams/Colts teams we so often beat.

 

[Oswlek]

The 2007 Giants may have already had a potent defensive line, but it was made even better because they didn't have to worry about the run anytime we were in shot gun, which we ran over 50% of the time the past three seasons. Their pass rush was made better because we were so predictable, like the 01-04 explosive Rams/Colts teams we so often beat.

The Giants had the success they did not because NE was predictibly passing, but because

1) NE's OL was dominated on both pass and run plays
2) Faulk left taking a key outlet and blitz pickup guy away.
3) Neal left taking NE's most athletic off the field and replacing him with someone who needed help on every down
4) Brady had a broken ankle which completely messed up the spacing and timing of the OL. Instead of sliding up and forward a couple feet to let the DE run around him, Brady could only sit there and hope the WR came open in time.
5) BB made a mistake of activating only two TE's so when K Brady went down, NE could only use 1 TE sets the rest of the way.

So, you want a guy with a broken ankle dropping back on every play, when the best blocking RB and TE are injured? He might not have even made it through the game had NE done that.

 

[maverick4]

for now, this "worry wort" (another so pleasant phrase that you chose to use) is going to stick to his guns and say that he believes that the trend described in the article and validated by the data he analyzed is worrisome while also acknowledging that there's a lot of noise around the distribution of the data when you look at individual games, as you observe. it's the problem of any analysis with relatively few data points.

(BTW, can't we just have conversations like this without resorting to ad hominem denigration with terms like "mangling" and "worry wort" and "panic?". personally, i have found this to be a very good back and forth by reasonable people--yourself, Deus and Patchick--who are looking at the same data through different lenses. Why can't we just leave it at that without the name calling?)

It's their standard tactic, Box and DaBruinz, to resort to petty name-calling and ad hominem personal attacks, if you disagree with them. They're so entrenched in their view on this, and consistently have been, that anything that goes against it will be automatically vilified by them.

They even nit pick at Kirwin's discussion of the stats. It's clear to me that he was referencing very specific statistics about how often QB's were getting hit - the 8.5 number did not come out of nowhere. The fact that in the article, conversationally he says "25 to 30 times" does not make it a vague guess, but rather the way someone would talk about something in conversation normally.

 

[maverick4]

The Giants had the success they did not because NE was predictibly passing, but because

So, you want a guy with a broken ankle dropping back on every play, when the best blocking RB and TE are injured? He might not have even made it through the game had NE done that.

If his ankle was broken (obviously it wasn't, but I'll play along with the exaggeration), Belichick wouldn't have played him. Also, if he was hobbled and the Giants D was so keyed in on blitzing when we were passing (and it was obvious when we were passing), it was moronic to call so many constantly slow developing pass plays. You did a good job listing all the injuries or missing blockers we had, which is why it made even less sense to keep putting Brady in a position to get hit. All the excuse making for that game sounds eerily similar like befuddled 01 Rams fans, or 03-04 Colts fans about those playoff losses.


I find it ironic that you can't see how predictable tendencies can positively impact the way a defense plays, especially since on the other site you wrote a long thread about how Maroney "dances" because defenses stack the box when he is in the game, knowing that he is going to run.

 

[Easywolf32]

Seriously what has been done as to not repeat getting our QB's heads and knees smashed in just like in the Pats - Giants superbowla nd even last season and this pre season? Nothing? Is that right? We're going with the same Offensive Line? man......

 

[Oswlek]

If his ankle was broken (obviously it wasn't, but I'll play along with the exaggeration), Belichick wouldn't have played him. Also, if he was hobbled and the Giants D was so keyed in on blitzing when we were passing (and it was obvious when we were passing), it was moronic to call so many constantly slow developing pass plays. You did a good job listing all the injuries or missing blockers we had, which is why it made even less sense to keep putting Brady in a position to get hit. All the excuse making for that game sounds eerily similar like befuddled 01 Rams fans, or 03-04 Colts fans about those playoff losses.


I find it ironic that you can't see how predictable tendencies can positively impact the way a defense plays, especially since on the other site you wrote a long thread about how Maroney "dances" because defenses stack the box when he is in the game, knowing that he is going to run.

No exaggeration at all. He was playing on a serious injury, much more so that people realize. I don't recall the plays well enough to know whether they were "slow developing" and I'm never watching that again.

I fail to see why reasons are "excuses". I am not dismissing the Giants' performance, I am explaining why it happened. Do you think BB just throws his arms up in the air after bad things occur or do you think he analyzes them and figures out why they did?

Re: Maroney, there is no inconsistency or irony in the two statements. Did you even read the reply I gave you on that thread?

Mav, I will go further into this if time warrants, but I disagree with your sentiment here.

Predictibility hurt Maroney's explosiveness and his public image, but it did little to hamper the offense as a whole. I am not contesting anything was wrong with the offense, I was solely pointing out illuminating factors.

I agree that Maroney should be used more in the passing game,but Maroney also has to earn those minutes over Faulk - a tough task for anyone.

Lastly (for now), your point about the shotgun snaps is flawed because, other than when Maroney was on the field and in superheavy short-distance plays, teams played NE to pass on every down anyway. Using a traditional drop back set would actually give the defense more time to get to the QB. The shoptgun gives the QB more time to read the field and makes it much easier to make a quick decision. There is a reason almost every team in the league uses a shot-gun formaion on long distance plays - because it is better at countering the full pass defense.

Had NE not been historically good at passing and, therefore, teams not sold out to stop the pass every down anyway, I would agree with you.

Also, it seems odd to complain about the nebulous "hits" when the season NE went shotgun heavy, Brady was in the bottom of the league in sacks. I find it hard to believe he was hit an extraordinary amount while rarely getting sacked. Additionally, Brady was sacked significantly more in 2006 when NE used the dropback set quite a bit more and had less overall pass attempts. Those indicate the exact opposite of what you are saying.

Teams played NE to pass on every down excepting many of Maroney's and short yardage. In the SB, the Giants were able to do this and still manhandle NE's OL on running plays. With no run blocking and limited RBs to even run, NE was forced to pass. If you know you are passing, and you know the other team knows, you go with the shotgun. It allows for the best pre and post snap view and offers the best chance for getting the ball out quick.

Why do you think every team in the league goes shotgun on 3rd and 10+? Because it counteracts the defenses advantage of knowing the pass is coming.

 

[BradyFTW!]

Oswlek: you're going to realize very, very quickly that maverick is a lost cause. He abandons all logic and objectivity in any discussion that even tangentially references McDaniels (which this one does, obviously, since McDaniels was the OC)

 

[maverick4]

No exaggeration at all. He was playing on a serious injury, much more so that people realize. I don't recall the plays well enough to know whether they were "slow developing" and I'm never watching that again.

Why do you think every team in the league goes shotgun on 3rd and 10+? Because it counteracts the defenses advantage of knowing the pass is coming.

1. There is no proof of the seriousness of his ankle injury, and if you're going to keep insisting on vague information like that, then how can you then dispute or dismiss any statistics pertaining to this thread?


2. I'm not debating the use of shot gun. Clearly it was designed to give the QB some extra time from dropping back, in obvious passing situations, when the D is pass rushing anyways. However, it is NOT CLEAR that shot gun does not prevent more QB hits. The two points are mutually exclusive. I don't know how you can keep repeating the purpose of shot gun in order to categorically dismiss that QB's aren't hit (not sacked) more from using it.

 

[maverick4]

He abandons all logic and objectivity

The height of hilarity, since this thread was started by PF74, a professional statistician.

I don't have a personal vendetta against McDaniels. He's my favorite ex-coordinator right now, and doing exactly what we need him to be doing in Denver for us, by being himself. He's no longer on the team, so I could care less what he does.

However, the issues raised in this thread and in past threads, are legit, and no matter how much the stats-whores try to play earmuffs since they love the sexy offensive numbers, the problem is still there and needs to be acknowledged.

Belichick clearly disagrees with you as well, since he completely re-tooled his RB, TE, and O-line positions in the off seasons, indicating a clear shift away from the shotgun-happy offense of the past 3 years.

 

[maverick4]

NE was forced to pass. .

You're using circular reasoning. Defenses know that when we are in shot gun, we pass 5 times for every 1 time we run it. We used shot gun over 50% of all the snaps from 2006-2008, clearly many times besides in 3rd and long.

New England wasn't forced to pass over and over, we stubbornly passed because we had been doing that all season. We lost on the same downfall for why the 2001 Rams and 2003-2004 Colts lost stubbornly not deviating from their successful schemes. Saying we were forced to pass due to some injuries (even though Brady was also injured as a passer according to your claim) is like saying Mike Martz was forced to keep calling slants and in-cuts against the 2001 Patriots, when that clearly wasn't the case.

It's lazy to just say that the Giants 4-man line was absolutely unstoppable that day and there was no way we could have won, and then say you're never going to see the tape again to confirm it or not.

 

[Oswlek]

Belichick clearly disagrees with you as well, since he completely re-tooled his RB, TE, and O-line positions, indicating a clear shift away from the shotgun-happy offense of the past 3 years.

So you are saying that the shotgun, which you admit helps the QB counteract the defense's knowledge also gets the QB hit more? How is that even possible? How can it give the QB more viewing time for better reads and quicker decisions yet still cause the QB to get hit more?

As I said before, I can see some validity to the point that NE telegraphed the pass giving the defense and edge. The problem with this is that the defense was already in pass mode anyway, telegraph or no. So that is really a non-factor.

One thing I do (sort-of) agree with you on is that BB clearly wants the run to be more of a focus than it was in 2007. Overhauling the entire backup middle of the OL, keeping two more true halfbacks and trading for a blocking TE clearly indicate this. More runs equals less hits, most likely.

Where I disagree with you is on the shotgun. I will be very surprised if NE isn't still in the shotgun for at least 60% of their passing plays. Brady is a surgeoun when he can scan the field like that, why would we want him at anything less than his best?

 

[Oswlek]

You're using circular reasoning. Defenses know that when we are in shot gun, we pass 5 times for every 1 time we run it. We used shot gun over 50% of all the snaps from 2006-2008, clearly many times besides in 3rd and long.

New England wasn't forced to pass over and over, we stubbornly passed because we had been doing that all season. We lost on the same downfall for why the 2001 Rams and 2003-2004 Colts lost stubbornly not deviating from their successful schemes. Saying we were forced to pass due to some injuries (even though Brady was also injured as a passer according to your claim) is like saying Mike Martz was forced to keep calling slants and in-cuts against the 2001 Patriots, when that clearly wasn't the case.

It's lazy to just say that the Giants 4-man line was absolutely unstoppable that day and there was no way we could have won, and then say you're never going to see the tape again to confirm it or not.

NE was killed in the trenches. They didn't trust Maroney to be the complete back that Faulk was. They lost their TE and best run blocking OL also.

There is no circular logic on this one. Do you remember when NE had a 2nd and 2 after an 8 yarder by Maroney following the Hobbs pick? What happened? Two straight runs and a punt on 4th and 5.

NE couldn't run so they had to pass. I don't see why that is hard to see.

 

[maverick4]

So you are saying that the shotgun, which you admit helps the QB counteract the defense's knowledge also gets the QB hit more? How is that even possible? How can it give the QB more viewing time for better reads and quicker decisions yet still cause the QB to get hit more?

Oswlek, so you disagree that the two are mutually exclusive? What about an analogy about running. Say a 3-TE package is used by some teams in short yardage, when both sides know it's a run. However, a team constantly using a 3-TE when even in non-obvious rushing situations, makes it harder to have success running the ball, as well as puts extra punishment on the RB to succeed. Even if the RB is a hall of famer and can still make it work, he's breaking more tackles and taking more damage simply because the opponent is keying in on what he is going to do. In that situation, you're relying on an elite talent to constantly execute against an opponent who knows what is coming. That hall of fame runner would have more success in a formation where the defense had to respect both run/pass, not just put the runner in the best (and most obvious) running formation over and over.

1. Here is a key point we disagree on. I disagree that the defense played pass most of the time regardless of the situation. They had to play both run and pass when Brady was under center. They played pass when we were in shot gun, which was over 50% of the time, and we pass/run 5:1 in shot gun the past 3 years.

If the defense played pass all the way even when under center (which they didn't), then it was even more stupid of the Pats the past 3 years not to use traps, draws, or inside runs more to punish teams who were overplaying pass.

2. Can you agree that with a tough/brave QB willing to stand in the pocket (like Brady), that the QB is more likely to take a hit in order to make a play work?