Brady's best teams

[glm]

Not saying this is the end all to be all, but it's intersting.

I basically took the expected wins of the teams (all explained here: Quarterback post-season records and Simpson’s Paradox » Pro-football-reference.com blog » Blog Archive, I was originally doing this to measure Brady during these seasons) and added them with their actual wins (did this for regular and postseason).

These are the results (The first number is the regular season expected and actual wins already added together, the next two are the postseason expected and actual wins):

07 Pats - 25.72 + 1.88 + 2 = 29.60.
04 Pats - 24.62 + 1.57 + 3 = 29.19.
03 Pats - 24.01 + 1.44 + 3 = 28.45.
06 Pats - 22.92 + 1.01 + 2 = 25.93.
11 Pats - 20.34 + 1.44 + 2 = 23.78.
01 Pats - 19.17 + 1.44 + 3 = 23.61.
10 Pats - 21.90 + 0.48 + 0 = 22.38.
05 Pats - 17.07 + 1.40 + 1 = 21.47.
09 Pats - 18.79 + 0.00 + 0 = 18.79.
02 Pats - 14.08 + 0.00 + 0 = 14.08.

Interesting to note, 2006 (10.92), 2004 (10.62), and 2003 (10.01) were all expected to more wins than 2007 (9.72).

Another thing, 2010 and 2011 were expected to be losing seasons (only 7.90 and 7.34, respectively). They were both lower than 2009 (8.79).

That's why 2007, 2010, and 2011 graded out as Brady's best seasons, without question (6.28, 6.10, 5.66, respectively) And if you're wondering about his worst seasons: 01 Brady (2.83), 09 Brady (1.21), and 06 Brady (1.08).

Thought it was interesting.

 

[supafly]

Interesting numbers.

I would think instantly that the 2003/2004 and 2007 teams would have been the best before opening the thread. They all had a decent defense and some kind of arguable balance in the running game.

Off the top of my head, 2002 and 2005 would be up there in 'worst' team seasons.

I am pretty surprised in regards to 2006, as the defense was still pretty damn good (surprising defensive numbers in some areas). I would not have guessed that would have been considered one of his 'worse' team yrs.

2011 seems to be right where I'd have expected, somewhere in the middle.

 

[PatsSox363804]

Amazing that in 06 we were a couple of bad PI calls away from winning the Superbowl.

 

[PatsFanSince74]

The author of the referenced piece adds a lot of caveats, to his credit. But, I did try once to duplicate a part of PFR's analysis and couldn't do so. I sent the web page a long email demonstrating my methodology and no one even bothered to answer or acknowledge it. So, beyond a repository of data, I don't rely on these guys for much.

If the OP wants to explain in simple language what the various stats are supposed to represent, I'd be happy to listen though.

 

[Fball]

Amazing that in 06 we were a couple of bad PI calls away from winning the Superbowl.

People always assume that getting past the Colts in that game would have been instant Lombardi, but that was not exactly a bad Chicago team. They had a great defense, and Grossman was a streaky QB (he happened to have a bad game against the Colts, but he was just as capable of putting up 300+ yards and a few TD's...)

 

[Ring 6]

Not saying this is the end all to be all, but it's intersting.

I basically took the expected wins of the teams (all explained here: Quarterback post-season records and Simpson’s Paradox » Pro-football-reference.com blog » Blog Archive, I was originally doing this to measure Brady during these seasons) and added them with their actual wins (did this for regular and postseason).

These are the results (The first number is the regular season expected and actual wins already added together, the next two are the postseason expected and actual wins):

07 Pats - 25.72 + 1.88 + 2 = 29.60.
04 Pats - 24.62 + 1.57 + 3 = 29.19.
03 Pats - 24.01 + 1.44 + 3 = 28.45.
06 Pats - 22.92 + 1.01 + 2 = 25.93.
11 Pats - 20.34 + 1.44 + 2 = 23.78.
01 Pats - 19.17 + 1.44 + 3 = 23.61.
10 Pats - 21.90 + 0.48 + 0 = 22.38.
05 Pats - 17.07 + 1.40 + 1 = 21.47.
09 Pats - 18.79 + 0.00 + 0 = 18.79.
02 Pats - 14.08 + 0.00 + 0 = 14.08.

Interesting to note, 2006 (10.92), 2004 (10.62), and 2003 (10.01) were all expected to more wins than 2007 (9.72).

Another thing, 2010 and 2011 were expected to be losing seasons (only 7.90 and 7.34, respectively). They were both lower than 2009 (8.79).

That's why 2007, 2010, and 2011 graded out as Brady's best seasons, without question (6.28, 6.10, 5.66, respectively) And if you're wondering about his worst seasons: 01 Brady (2.83), 09 Brady (1.21), and 06 Brady (1.08).

Thought it was interesting.

Your numbers are all wrong.
Expected wins in 2007 was 13.8 2010 was 12.3 2011 was 11.6 etc.
Not sure what you are looking at.

In any event I dont know why you would take any value from expected wins to begin with.
Expected wins is a calculation of statistics to predct a winner, which is silly when you have determined one.

 

[Ring 6]

Interesting numbers.

I would think instantly that the 2003/2004 and 2007 teams would have been the best before opening the thread. They all had a decent defense and some kind of arguable balance in the running game.

Off the top of my head, 2002 and 2005 would be up there in 'worst' team seasons.

I am pretty surprised in regards to 2006, as the defense was still pretty damn good (surprising defensive numbers in some areas). I would not have guessed that would have been considered one of his 'worse' team yrs.

2011 seems to be right where I'd have expected, somewhere in the middle.

His numbers are all wrong.

 

[Ring 6]

correct expected win #s
2011 11.6
2010 12.3
2009 11.6
2008 10.6
2007 13.8
2006 12.2
2005 9.1
2004 12.4
2003 11.4
2002 8.9
2001 10.8

But again, they are pretty meaningless.

 

[jays52]

That 04 team was a juggernaut. They would have crushed the 07 team.

 

[Palm Beach Pats Fan]

correct expected win #s
2011 11.6
2010 12.3
2009 11.6
2008 10.6
2007 13.8
2006 12.2
2005 9.1
2004 12.4
2003 11.4
2002 8.9
2001 10.8

But again, they are pretty meaningless.

I would think that subtracting expected wins from actual wins may give you a possible measure of performance in the clutch and ability to excel in situational football (perhaps superimposed with good luck/bad luck).

In other words, the stats show a certain level of core capabilities, and how well did you apply these abilities effectively to win the games?

 

[RobertWeathers]

Interesting numbers.

I would think instantly that the 2003/2004 and 2007 teams would have been the best before opening the thread. They all had a decent defense and some kind of arguable balance in the running game.

Off the top of my head, 2002 and 2005 would be up there in 'worst' team seasons.

I am pretty surprised in regards to 2006, as the defense was still pretty damn good (surprising defensive numbers in some areas). I would not have guessed that would have been considered one of his 'worse' team yrs.

2011 seems to be right where I'd have expected, somewhere in the middle.

2006 was very balanced and in 2004 they ran more than they passed.

 

[Ring 6]

I would think that subtracting expected wins from actual wins may give you a possible measure of performance in the clutch and ability to excel in situational football (perhaps superimposed with good luck/bad luck).

In other words, the stats show a certain level of core capabilities, and how well did you apply these abilities effectively to win the games?

I think its meaningless.
I don't understand how using statistics to predict who is most likely to win has any value when you know who won, especially when statistics are as much caused by the outcome as the outcome is caused by statistics (ie rushing yards and win %)
The only real value seems to be in judging how good their model of prediction is.

 

[glm]

Your numbers are all wrong.
Expected wins in 2007 was 13.8 2010 was 12.3 2011 was 11.6 etc.
Not sure what you are looking at.

If you're gonna say my numbers are all wrong, it'd be nice to show where my math went wrong. And to show your math too.

In any event I dont know why you would take any value from expected wins to begin with.
Expected wins is a calculation of statistics to predct a winner, which is silly when you have determined one.

As I said in the original post, it was to see how Brady was playing. The expected wins are looking at how the players around him performed, and the actual wins are that plus how Tom actually performed.

 

[Ring 6]

If you're gonna say my numbers are all wrong, it'd be nice to show where my math went wrong. And to show your math too.

I looked it up on their site. Its right on the team page for each season, which I listed in another post.

As I said in the original post, it was to see how Brady was playing. The expected wins are looking at how the players around him performed, and the actual wins are that plus how Tom actually performed.

Expected wins are an analysis of the statistics to determine who was more likely to win based upon statistics other than score.

 

[glm]

I looked it up on their site. Its right on the team page for each season, which I listed in another post.

And that's the problem. If you'd go to the link I provided you'd see I'm measuring something different.

Expected wins are an analysis of the statistics to determine who was more likely to win based upon statistics other than score.

K.

 

[Ring 6]

And that's the problem. If you'd go to the link I provided you'd see I'm measuring something different.






K.

Where does that link show the Patriots expected win numbers?
That article is an admittedly flawed attempt to judge a QB based on rushing yards differential.

 

[glm]

Where does that link show the Patriots expected win numbers?
That article is an admittedly flawed attempt to judge a QB based on rushing yards differential.

I calculated the Pats numbers. And you forgot the whole defense part of the article.

 

[Ring 6]

I calculated the Pats numbers. And you forgot the whole defense part of the article.

How did you calculate the numbers? The site has expected wins right on the season team page, and those numbers conflict with yours.

 

[ivanvamp]

That 04 team was a juggernaut. They would have crushed the 07 team.

I cannot think of a team in NFL history that would have (except in some weird fluke instance) "crushed" the '07 Patriots. That '07 team had a very, very good defense (not elite, but terrific) and the greatest offense the NFL has ever seen. They lost only one game all year and that was by 3 points in the last 30 seconds that required one of the most miraculous plays (and egregious non-calls) in league history to accomplish.

There were other teams in history that certainly could have beaten the '07 Patriots, but "crushed" them? Again, anything can happen in one game, but it would have taken something really remarkable for anyone to have crushed that squad.

 

[glm]

How did you calculate the numbers? The site has expected wins right on the season team page, and those numbers conflict with yours.

The expected wins I'm doing isn't taking into account brady.