Patriots Fumble Rates Compared to Other Teams

VectorPrime said:

I'm well aware that was the year we switched from a focus on power running to the flying circus. But Maroney got his touches.

Click to expand...

That's only part one. Here's a clue about part 2 (which is comprised of 2 sub-parts)

12
11
13
7
4
12
6
4
3
6
2
10
6

Deus Irae said:

That's only part one. Here's a clue about part 2 (which is comprised of 2 sub-parts)

12
11
13
7
4
12
6
4
3
6
2
10
6

Click to expand...

I give up.

You don't have to be coy, if you have info to put my mind at ease then just post it. I'm on the Pats side here.

VectorPrime said:

I give up.

You don't have to be coy, if you have info to put my mind at ease then just post it. I'm on the Pats side here.

Click to expand...

I'm not being coy. I'm actually laying things out. Those are Brady's fumble numbers. Answering you right now means you have to find your own numbers to verify, but here you go:

First 3 years - learning the game and the position, getting some of his highest sack totals
Next 2 years - Brady's fumble numbers drop significantly, and the sack numbers drop
2006 - Significant talent decline on the offense putting Brady in more dangerous positions

2007 - Huge talent upgrade, combined with the full implementation of the more shotgun based offense
2013 - Highest sack total since 2001


You've got a team that went from a mostly power run game, to one that is much more balanced, while pulling the QB away from center, which gives the QB a better chance to see the rush. 2011 is the year we have an outlier where high sack totals didn't lead to significantly more fumbles, and 2006 is the outlier where lower sack totals still saw high fumble numbers.

Deus Irae said:

I'm not being coy. I'm actually laying things out. Those are Brady's fumble numbers. Answering you right now means you have to find your own numbers to verify, but here you go:

First 3 years - learning the game and the position, getting some of his highest sack totals
Next 2 years - Brady's fumble numbers drop significantly, and the sack numbers drop
2006 - Significant talent decline on the offense putting Brady in more dangerous positions

2007 - Huge talent upgrade, combined with the full implementation of the more shotgun based offense
2013 - Highest sack total since 2001


You've got a team that went from a mostly power run game, to one that is much more balanced, while pulling the QB away from center, which gives the QB a better chance to see the rush. 2011 is the year we have an outlier where high sack totals didn't lead to significantly more fumbles, and 2006 is the outlier where lower sack totals still saw high fumble numbers.

Click to expand...

Excellent point, I;m sure Brady fumbles the ball a lot fewer times than most other QBs. But does that amount of fewer fumbles really account for the discrepancy between the Pats and other teams? I haven't done the math.

VectorPrime said:

Excellent point, I;m sure Brady fumbles the ball a lot fewer times than most other QBs. But does that amount of fewer fumbles really account for the discrepancy between the Pats and other teams? I haven't done the math.

Click to expand...

I had been trying to lay everything out. You told me not to be coy. Now you can do your own math.

Just add about a half dozen fumbles per year, for the most part.

By the way -- how do the numbers look for home vs. away fumbles? If I read what's being insinuated correctly, that could make a huge difference ...

Based on the assumption that fumbles per play follow a normal distribution, you’d expect to see, according to random fluctuation, the results that the Patriots have gotten over this period, once in 16,233.77 instances”.

This is from the original statistical analysis and I saw immediately that it indicates the entire article is flawed. There is no reason that fumbles would follow a normal distribution for many of the reasons above. First one team has had a 15 year coach with a near maniacal obsession with turnovers. Players who fumble are benched for multiple games despite their overall talent level. Dues highlights the reasons for lower QB fumbles and there is no reason to think they would have anything to do with the ball itself. Law firm was on the team from 08 to 11 an never fumbled, and had not ever fumbled in High School or College, individually he was a statistical anomaly. Once you understand it is not a normal distribution the entire thing falls apart.

Oh, and how is it possible the Chiefs went an entire year in this pass happy era and did not have a single touchdown by a wide receiver.

Fumbles are not a roulette wheel where the expected distribution of fumbles should be equal over enough tries unless the wheel is rigged.

Humans themselves are the biggest variable.

My short answer is that he's a ****** with graphing software. There are endless things you could graph that are "impossible."

It's impossible for someone to make a Hail Mary pass because it failed the last 50 times. You could kick 20 onside kicks and recover zero, then get two in a row. The Pats won like 14 coin tosses this year. Outliers are not impossibilities.

The only impossibility here is that there could be any conclusion drawn about 1 or 2psi being an overriding factor from a thousand other variables based off the data he has.

A graph on lost fumbles makes you 'just not want to believe the Patriots are liars and cheaters? Hey, excellent passive aggressive POV pushing. It's better than the cursory 'Who knows the truth? But studies show...it is the truth".
Anyone who is a "fan" of the Patriots wouldn't be surprised they have less fumbles over a long sample. Anyone who is a "fan" of the Patriots, considering the recent few days/today's BB PC, is already looking at this "Scandal" for what it really is.
Non fans, ones with mental issues, probably will be attempting to find ways to keep it going at BB/TB/Patriots.
I mean I don't want to believe that is true of you, but troll studies show....

Bobsyouruncle said:

Fumbles are not a roulette wheel where the expected distribution of fumbles should be equal over enough tries unless the wheel is rigged.

Humans themselves are the biggest variable.

My short answer is that he's a ****** with graphing software. There are endless things you could graph that are "impossible."

It's impossible for someone to make a Hail Mary pass because it failed the last 50 times. You could kick 20 onside kicks and recover zero, then get two in a row. The Pats won like 14 coin tosses this year. Outliers are not impossibilities.

Click to expand...

An outlier in 7 of 8 years?

VectorPrime said:

An outlier in 7 of 8 years?

Click to expand...

I don't have time right now, but I'll look at it closer later.

Off hand NE being 3rd in the league in fumbles per play does not strike me as amazing. I also don't know what the 7 of 8 years you refer to are. Is that some data where he picked certain teams and said look at this subset I created? Look how far NE is from my cherry picked data? ATL and NO have fewer fumbles per play. Once somebody finds they are using voodoo I'll send the witch doctor out to investigate NE's use of of the necronomican.

Bobsyouruncle said:

I don't have time right now, but I'll look at it closer later.

Off hand NE being 3rd in the league in fumbles per play does not strike me as amazing. I also don't know what the 7 of 8 years you refer to are. Is that some data where he picked certain teams and said look at this subset I created? Look how far NE. Is from my cherry picked data. ATL and NO have fewer fumbles per play. Once somebody finds they are using voodoo I'll send the witch doctor out to investigate NE's use of of the necronomican.

Click to expand...

7/8 are the years since 2006 that the Patriots were substantially under the league average in fumbles.

07
08
09
10
11
12
14

VectorPrime said:

7/8 are the years since 2006 that the Patriots were substantially under the league average in fumbles.

07
08
09
10
11
12
14

Click to expand...

That ignores '04, which was also in that same vein.

guys..problems with this

it says in there that the stats are based on the assumption that any given player is just as likely to fumble as another player.. thats obviously not true


second, the report accounts only for fumbles lost, not fumbles recovered by patriots. when you click on the link on the website, they go further and when you account for all fumbles, not only are we not an outlier, but we are 5th in the league. they try to justify it that well dome teams fumble less and when you take those out of the equation we are an outlier.

it also doesnt take into account each team's coaches differential treatment of fumbling

further, anyone who ever has player football, whether at the high school level, college, pros, knows there is one person who determines what is done to the football. and thats the quarterback. and i can assure you, unless its an option quarterback and even thats sketchy, they dont pick ball based on fumbling, but on passing the ball.

you can play many games with statistics. but if you read into this report, its apparent that its a waste of time

BTW, looking at the chart, the fumble percentages for the 'other' teams has also dropped since 2007. Prior to 2007, the only year in which the percentage was under 2% was 2005. Since 2007, that's happened 4 of 8 seasons, with another year being barely above that 2% threshhold.

Cool story, bro.

Now let's try some more useless graphs with statistical impossibilities….such as our rate of winning as compared to other teams in the last 15 years?

billdog3484 said:

guys..problems with this
second, the report accounts only for fumbles lost, not fumbles recovered by patriots. when you click on the link on the website, they go further and when you account for all fumbles, not only are we not an outlier, but we are 5th in the league. they try to justify it that well dome teams fumble less and when you take those out of the equation we are an outlier. time

Click to expand...

This is exactly what I wanted. Where do they go into all fumbles and not just those lost?

VectorPrime said:

7/8 are the years since 2006 that the Patriots were substantially under the league average in fumbles.

07
08
09
10
11
12
14

Click to expand...

Below average isn't an outlier. If they were indeed outliers 7 of 8 years it would be interesting. I'm not sure it would tell us anything, but it would be more interesting than being below average 7 of 8 years. It appears they weren't even the lowest in the league in fumbles most of those years.

PatsWickedPissah said:

It's fumbles lost. If you recover your fumbles the stat changes.
Graph the % running plays vs passing over the same timescale and you'll notice a decrease. Most fumbles are by RBs.

Click to expand...

As Football Outsiders has noted, fumble recovery is incredibly random.

Maybe this has to do with the fact that you will left in the doghouse if you fumble ONCE.

Que the conspiracy theories