Brady hit once every 8.5 attempts since mid 07

[TruthSeeker]

I don't understand how an article can just toss out stats like those without giving you league averages, or at least comparisons to other elite QBs, to compare with. Without any context, I have absolutely no idea how to interpret those numbers.

My thoughts exactly. You can't draw conclusions from numbers in a vacuum - not that this stops them.

 

[Deus Irae]

 

[Box_O_Rocks]

The challenge is exacerbated by the fact that the article limits the evaluation period to a limited universe of games and includes "hits," which I don't think are formally tracked, so we have to take the word of the observer. In addition, sack stats are usually presented only for the regular season; to get reliable full season numbers we'd have to factor in the Playoffs and I just don't have the time to fire up a spreadsheet today. But, on the assumption that the observer is reliable and looking at facts we can gather, here are a few things that we can discover.

Looking just at 2007, since that's the period from which the article derives its data, we can compare Brady's sacks per attempt for the full season to the sacks on other elite QB's; this doesn't include hits. Any analysis like this has to acknowledge, of course, that the playcalling and style of play of QB's differs (see Ben below). I get this from ProFootBallReference.com.

Sacks per Attempt % (Sacks/Attempts + Sacks)

Brady: 3.5%
Brees: 2.4%
Favre: 2.7%
Payme: 3.9%
Rivers: 4.7%
Big Ben: 10.4%

This suggests that, throughout the entire season, Brady was sacked roughly 46% more often than Brees, 30% more often than Favre, 11% less often than Peyton, 27% less often than Rivers and a lot less often than Roethlisberger.

But, the article talks about the latter part of the season and the playoffs. So, if we look at the first ten games, Brady was sacked just 2.9% of the time; for the last six games (Philly on as the article argues), Brady was sacked 4.4% of the time, or more than 50% more often than in the beginning of the season. When we look at the Playoffs, he was sacked 6.8% of the time, or 230% more often than in the first six games of the season and nearly twice as often as in the entire regular season. If we look at his sack percentage for the last six games plus the playoffs, it is 5.2%.

Since the article is accurate in its arguments about sacks, barring the presentation of alternative data, I'll concede that it is likely also accurate in its stats on "hits" and will accept that its conclusion is valid, barring the presentation of contrary data. From the 11th game on of the 2007 season, Brady was sacked and hit more often than before and more often than other elite QB's, other than Roethlisberger, were sacked and hit on average over the season. This is probably not a good thing.

So, in the playoffs, against the best defenses in the NFL, you are more likely to get sacked?

 

[PatsFanSince74]

Just wanted to note that QB hits actually are formally tracked (on ESPN.com). You have to look at the box scores for individual games.

darn, why does that have to be on the HSPN!

 

[PatsFanSince74]

So, in the playoffs, against the best defenses in the NFL, you are more likely to get sacked?

interesting question. i'm a little busy now, but i'll look at the data on other teams and get back to you on it. see if there's a difference in what happened to TB last time in comparison to others.

 

[Spiral]

What worries me is that a coach like Ryan is not going to be deterred by a slant going for a touchdown (or three), or a roughing the passer penalty (or three). If he can get someone a clear shot on Brady that puts him out for a while, he'll do it. I worry about Light and I worry a lot about Kazcur. Anyone who believes these guys are quicker or more athletic than they were in 2007, let me know.

 

[PatsFanSince74]

How can one refute stats (hits taken by a QB) that aren't even official? They do not definie what a HIT is. The ESPN article that Maverick used to erroneously lambaste McDaniels, ignored sacks and didn't define what a hit was considered either.

While Curran and Kirwin mention the 19 sacks in those 10 games, what they don't mention is that Neal wasn't playing for in 4 of those games and in 2 more, he played less than half the game. One of the games that he played less than half was the SB and the Pats got schooled because they didn't recognize the stunts or couldn't break free to pick up the stunting DE.

Any article that hangs its hat on stats it refuses to define is not an article worth the Cyberspace its inhabiting.

In all honesty, Brady gets hit too much after he's gotten rid of the ball and the Refs let it go because he just bounces back up ususally. There really needs to be a better point of emphasis on hitting the QB late because it will cost this league QBs.

Also, it should be noted that Curran and Kirwan erroneously attribute the increase in the blitzing of Brady to Jim Johnson. What they fail to remember is that Johnson actually attributed it to Romeo Crennel and the Browns.

interesting points on the stats. i'd appreciate your comments on my reply to Patchick.

 

[PatsFanSince74]

I can endorse this post as a valid rebuttal.

the narrow point the article was making, which I have seen no data refute (lots of opinions, but no data), is that Brady was sacked and hit increasingly often as the 2007/08 season went on, that the rates were higher after game ten and in the playoffs than they were before and that those latter rates were generally higher than what was experienced by most other elite QB's. as far as I can see, that conclusion is valid. (an interesting point was raised in another post as to whether QB's in general are sacked more often in the Playoffs than in the regular season and I'm going to try to track that down when I have some time.)

the article does not get into assessing blame or causes, it just states the observable facts. the extrapolated conclusion is that more sacks/hits = more risk of injury; that also seems inarguable and the injury to Brady in his last regular season game and what most observers agree was a close call from the Haynesworth hit just three (preseason) games later (in which he played less than half the time) suggests that this is an area for concern. that's really all that the article, which I don't particularly like but which I cannot refute, is arguing.

 

[patchick]

But, the article talks about the latter part of the season and the playoffs. So, if we look at the first ten games, Brady was sacked just 2.9% of the time; for the last six games (Philly on as the article argues), Brady was sacked 4.4% of the time, or more than 50% more often than in the beginning of the season. When we look at the Playoffs, he was sacked 6.8% of the time, or 230% more often than in the first six games of the season and nearly twice as often as in the entire regular season. If we look at his sack percentage for the last six games plus the playoffs, it is 5.2%.

But don't you assume that every QB will be sacked significantly more in the playoffs because they're playing, on average, significantly better defenses? So you have to compare Brady's playoff stats to other QBs' playoff stats.

 

[patchick]

But don't you assume that every QB will be sacked significantly more in the playoffs because they're playing, on average, significantly better defenses? So you have to compare Brady's playoff stats to other QBs' playoff stats.

EDIT: I see Box and I were thinking alike.

 

[PatsFanSince74]

But don't you assume that every QB will be sacked significantly more in the playoffs because they're playing, on average, significantly better defenses? So you have to compare Brady's playoff stats to other QBs' playoff stats.

that's an interesting and valid question, so i'm going to track it down during my lunch hour. i presented the article because i thought it was interesting and invited data to refute it. so far, there's been a lot of blustering but no data that refutes the narrow point of the article (see my response to Robo above). the disturbing thing for pats fans is that the data from the last few quarters that TB has played include the hit on his knee that was a real deal and the hit by Haynesworth just a few quarters later that scared the bejusus, rightly or wrongly, out of a lot of people here.

 

[jmt57]

I don't understand how an article can just toss out stats like those without giving you league averages, or at least comparisons to other elite QBs, to compare with. Without any context, I have absolutely no idea how to interpret those numbers.

You beat me to it; I completely agree - any numbers are meaningless without being able to compare them to the rest of the data and averages.

In regards to 19 sacks in ten games - about two per game - without looking up league stats, doesn't two sacks per game sound about average? Yet it's being implied to be an alarmingly high number by the columnists.

To me the number of hits per attempt is meaningless. For example a QB that is being hit one out of every four pass attempts on a team that runs the ball 80% of the time is getting hit less often than a QB who is hit one out of every ten pass attempts on a team that passes the ball 60% of the time.

A more meaningful number is total number of hits, showing how much wear and tear a player has taken, over the course of a year, recent years, and his career. Sacks don't measure that because that implies that the only time a QB is hit hard is on a sack, which is not true. However since hits is not an official stat, I'm a bit skeptical as to the accuracy of those numbers.

 

[patchick]

that's an interesting and valid question, so i'm going to track it down during my lunch hour. i presented the article because i thought it was interesting and invited data to refute it. so far, there's been a lot of blustering but no data that refutes the narrow point of the article (see my response to Robo above). the disturbing thing for pats fans is that the data from the last few quarters that TB has played include the hit on his knee that was a real deal and the hit by Haynesworth just a few quarters later that scared the bejusus, rightly or wrongly, out of a lot of people here.

Absolutely, I'm not discounting the possibility that there's something real to this. I just don't see the evidence so far to alarm me too much. Even looking at the 2007 playoff stats, that's ALL the Giants' overwhelming rush. Until the Superbowl, Brady's playoff sack rate was his standard 1 in 20 attempts.

 

[DarrylS]

You beat me to it; I completely agree - any numbers are meaningless without being able to compare them to the rest of the data and averages.

In regards to 19 sacks in ten games - about two per game - without looking up league stats, doesn't two sacks per game sound about average? Yet it's being implied to be an alarmingly high number by the columnists.

To me the number of hits per attempt is meaningless. For example a QB that is being hit one out of every four pass attempts on a team that runs the ball 80% of the time is getting hit less often than a QB who is hit one out of every ten pass attempts on a team that passes the ball 60% of the time.

A more meaningful number is total number of hits, showing how much wear and tear a player has taken, over the course of a year, recent years, and his career. Sacks don't measure that because that implies that the only time a QB is hit hard is on a sack, which is not true. However since hits is not an official stat, I'm a bit skeptical as to the accuracy of those numbers.

I agree to look at these stats in a vacuum are somewhat useless...

 

[DaBruinz]

Since the article is accurate in its arguments about sacks, barring the presentation of alternative data, I'll concede that it is likely also accurate in its stats on "hits" and will accept that its conclusion is valid, barring the presentation of contrary data. From the 11th game on of the 2007 season, Brady was sacked and hit more often than before and more often than other elite QB's, other than Roethlisberger, were sacked and hit on average over the season. This is probably not a good thing.

How can you assume that its accurate in its presentation of other data when Sacks are an official stat measured by the league and hits are not?

That's the equivalent of saying because Joe can hit a nail with a hammer then Joe must be able to build a house...

The article is poorly done which is surprising for Curran. It makes no definition on what a hit was. And, as pointed out, offers no comparison. You comparing sacks of other QBs is meaningless and offers no support to the idea that Brady gets hit more often because you can get hit without being sacked.

 

[DaBruinz]

Please see my post above. That's not quite accurate.

Sorry, but its dead on accurate... Just because it flies in the face of the article doesn't change reality.

 

[DaBruinz]

Just wanted to note that QB hits actually are formally tracked (on ESPN.com). You have to look at the box scores for individual games.

QB Hits are NOT an NFL acknowledged stat. They are an ESPN generated stat in which they do not define what is considered a hit. And because they do not define what is considered a hit, there i no basis for comparison. There is no standard. Just like there is no standard on tackles. Rich Gosselin pointed this out 2 years ago in an article he did showing how there were teams that gave out too many tackles. He pointed out that, in some cases, teams were averaging 2 tackles per play.. Which you can't do. You can only have either 1 tackle or 2 assissted per play.

 

[DaBruinz]

the narrow point the article was making, which I have seen no data refute (lots of opinions, but no data), is that Brady was sacked and hit increasingly often as the 2007/08 season went on, that the rates were higher after game ten and in the playoffs than they were before and that those latter rates were generally higher than what was experienced by most other elite QB's. as far as I can see, that conclusion is valid. (an interesting point was raised in another post as to whether QB's in general are sacked more often in the Playoffs than in the regular season and I'm going to try to track that down when I have some time.)

the article does not get into assessing blame or causes, it just states the observable facts. the extrapolated conclusion is that more sacks/hits = more risk of injury; that also seems inarguable and the injury to Brady in his last regular season game and what most observers agree was a close call from the Haynesworth hit just three (preseason) games later (in which he played less than half the time) suggests that this is an area for concern. that's really all that the article, which I don't particularly like but which I cannot refute, is arguing.

But the article doesn't base itself on just facts. There is only one fact it based itself on. Sacks. Hits taken is not a stat. Its undefined and not recognized by the NFL.

They then used this fictitious number in conjunction with stats to make it sound like no other QB gets hit like Brady does. So, there is no comparison to relate to.

Its basically a poorly written fluff piece that tells us that Brady could get hurt if he gets hit.. The problem is that ANY QB, hell ANY PLAYER, can get hurt if he gets hit. And there are players who take a lot more beating than Brady does.

Please don't fall into the trap that maverick fell into. That these guys know what they are talking about.

Curran was a great reporter when he only had to deal with the Pats... Since he's gone national, there have been plenty of questionable articles that he's written.. Including the Gloom & Doom article he wrote in January where he basically said Brady wasn't going to play this season.

Kirwan is a hack who happened to be in the NFL at one point. Its wrong more often than he's right. So, why should you believe what's he's saying.

 

[PatsFanSince74]

You beat me to it; I completely agree - any numbers are meaningless without being able to compare them to the rest of the data and averages.

In regards to 19 sacks in ten games - about two per game - without looking up league stats, doesn't two sacks per game sound about average? Yet it's being implied to be an alarmingly high number by the columnists.

To me the number of hits per attempt is meaningless. For example a QB that is being hit one out of every four pass attempts on a team that runs the ball 80% of the time is getting hit less often than a QB who is hit one out of every ten pass attempts on a team that passes the ball 60% of the time.

A more meaningful number is total number of hits, showing how much wear and tear a player has taken, over the course of a year, recent years, and his career. Sacks don't measure that because that implies that the only time a QB is hit hard is on a sack, which is not true. However since hits is not an official stat, I'm a bit skeptical as to the accuracy of those numbers.

On the narrow points of the article, though, see my responses above to Patchick and others, #17 and #28

 

[PatsFanSince74]

But the article doesn't base itself on just facts. There is only one fact it based itself on. Sacks. Hits taken is not a stat. Its undefined and not recognized by the NFL.

They then used this fictitious number in conjunction with stats to make it sound like no other QB gets hit like Brady does. So, there is no comparison to relate to.

Its basically a poorly written fluff piece that tells us that Brady could get hurt if he gets hit.. The problem is that ANY QB, hell ANY PLAYER, can get hurt if he gets hit. And there are players who take a lot more beating than Brady does.

Please don't fall into the trap that maverick fell into. That these guys know what they are talking about.

Curran was a great reporter when he only had to deal with the Pats... Since he's gone national, there have been plenty of questionable articles that he's written.. Including the Gloom & Doom article he wrote in January where he basically said Brady wasn't going to play this season.

Kirwan is a hack who happened to be in the NFL at one point. Its wrong more often than he's right. So, why should you believe what's he's saying.

i think the data say that you are wrong on the narrow point the article is trying to make, as I have stated in a couple of responses above.