Pats "nearly fumble proof" after 2006 Rule change proposed by (1 Viewer)

[NE_REVIVAL]

Link

from the link:

Initially, looking at weather data, I noticed the Patriots performed extremely well in the rain, much more so than they were projected. I followed that up by looking at the fumble data, which showed regardless of weather or site, the Patriots prevention of fumbles was nearly impossible. Ironically, both studies saw the same exact starting point: 2007 was the first season where things really changed for the Patriots. Something started in 2007 which is still on-going today.

Hey Buff, help me out here, what was the title of that Pats hating October thread you started? I can't seem to locate it; some ground breaking story about radios in helmets. It was the one where Rudnicki stepped in and admonished me for implying you were a Pats hating troll (let me be clear, I am NOT implying you are a troll). I think Rudnicki merged it with another thread, but I can't quite remember the title you gave it, do you remember?

TIA

something to the effect of pats' cheating ways not limited to videotaping. not sure that's verbatim. that was October and it's still a newsworthy headline/topic. whoa.

Ah yes, Thanks.

http://fivethirtyeight.com/datalab/your-guide-to-deflate-gateballghazi-related-statistical-analyses/

Here were some of the responses to Sharp’s posts:

• At Deadspin, statistics professors Gregory J. Matthews and (friend of FiveThirtyEight) Michael Lopez wrote a great, FireJoeMorgan-style, line-by-line takedown of Sharp’s most popular post. They refuted the 1-in-16,234 number (by Sharp’s own methodology it should be more like 1-in-297) and pointed out a massive data error in Sharp’s analysis of individual players (he mixed together some data that included special teams plays and some that excluded them). Matthews and Lopez also broke down team fumble rates by position, after which New England’s running backs and receivers don’t really look like major fumble-preventing outliers at all.
• SoSH Football Central’s Daryl Sng broke down Sharp’s aforementioned data errors in even greater detail. After excluding kick and punt returns (which make no sense to include because teams don’t have any access to “K balls”) and correcting for Sharp’s original mishmash of regular-season and playoff data, the players in Sharp’s sample fumbled only about 23 percent more as non-Patriots, not 88 percent as was originally stated.
• Political scholar Bill Herman also zeroed in on Sharp’s analysis of individual players’ fumble rates with the Patriots and other teams, identifying its aforementioned methodological errors. In addition, he looked at the six players featured in Michael Salfino’s Wall Street Journal article based on Sharp’s work, finding that their difference in fumbling was statistically significant. Of those six players (Danny Amendola, BenJarvus Green-Ellis, Danny Woodhead, Wes Welker, Brandon LaFell and LeGarrette Blount), four were common to Sng’s dataset, but both analyses found a 23 percent increase in fumbling while playing for teams other than New England.
• Like Matthews and Lopez, data analyst Tom Hayden repudiated Sharp’s assumption that his “plays per fumble” metric was normally distributed across NFL teams (a necessary condition for the 1-in-16,234 claim).
• The harshest counterargument belonged to data scientist Drew Fustin. Fustin challenged Sharp’s choice to exclude dome teams (Sharp’s own post says outdoor teams barely fumble more often than those based in domes), instead looking at fumble rates across all teams in outdoor games only — whereupon the Patriots don’t even rank first in the NFL at fumble avoidance over the 2010-2014 period. He also questions whether Sharp’s decision to use that 2010-14 period was a case of cherry-picking the timeframe that would make the Patriots look most like an extreme outlier.

And much like the radio in the helmet nonsense, I think we are done here.

 

[Buffaloes]

Political scholar Bill Herman also zeroed in on Sharp’s analysis of individual players’ fumble rates with the Patriots and other teams, identifying its aforementioned methodological errors. In addition, he looked at the six players featured in Michael Salfino’s Wall Street Journal article based on Sharp’s work, finding that their difference in fumbling was statistically significant. Of those six players (Danny Amendola, BenJarvus Green-Ellis, Danny Woodhead, Wes Welker, Brandon LaFell and LeGarrette Blount), four were common to Sng’s dataset, but both analyses found a 23 percent increase in fumbling while playing for teams other than New England.

some good discussion RE: the article. the bolded is also pretty interesting and something we touched on a page earlier.

 

[12punch]

I've actually been amazed by my fellow pats fans through all this at how quickly and thoroughly they've debunked each bit of nonsense as it comes up.

gotta hand it to those guys -- no other fanbase in the league would be crushing like this

 

[seahawk 17]

Anyone catch what Belichick said about the footballs they practice with? That might help you hold on when using a good football in a game.

I heard, but what do you expect him to say?

 

[12punch]

GUISE I FOUND A STAT THE PATRIOTS HAVE BEEN GOOD AT THE LAST 5 YEARS!!!!!!!4234242

 

[HellToupee]

Link

from the link:

Initially, looking at weather data, I noticed the Patriots performed extremely well in the rain, much more so than they were projected. I followed that up by looking at the fumble data, which showed regardless of weather or site, the Patriots prevention of fumbles was nearly impossible. Ironically, both studies saw the same exact starting point: 2007 was the first season where things really changed for the Patriots. Something started in 2007 which is still on-going today.

Stevan Ridley

 

[moleculo]

Link

from the link:

Initially, looking at weather data, I noticed the Patriots performed extremely well in the rain, much more so than they were projected. I followed that up by looking at the fumble data, which showed regardless of weather or site, the Patriots prevention of fumbles was nearly impossible. Ironically, both studies saw the same exact starting point: 2007 was the first season where things really changed for the Patriots. Something started in 2007 which is still on-going today.

Hey Buff, help me out here, what was the title of that Pats hating October thread you started? I can't seem to locate it; some ground breaking story about radios in helmets. It was the one where Rudnicki stepped in and admonished me for implying you were a Pats hating troll (let me be clear, I am NOT implying you are a troll). I think Rudnicki merged it with another thread, but I can't quite remember the title you gave it, do you remember?

TIA

something to the effect of pats' cheating ways not limited to videotaping. not sure that's verbatim. that was October and it's still a newsworthy headline/topic. whoa.

Ah yes, Thanks.

http://fivethirtyeight.com/datalab/your-guide-to-deflate-gateballghazi-related-statistical-analyses/

Here were some of the responses to Sharp’s posts:

• At Deadspin, statistics professors Gregory J. Matthews and (friend of FiveThirtyEight) Michael Lopez wrote a great, FireJoeMorgan-style, line-by-line takedown of Sharp’s most popular post. They refuted the 1-in-16,234 number (by Sharp’s own methodology it should be more like 1-in-297) and pointed out a massive data error in Sharp’s analysis of individual players (he mixed together some data that included special teams plays and some that excluded them). Matthews and Lopez also broke down team fumble rates by position, after which New England’s running backs and receivers don’t really look like major fumble-preventing outliers at all.
• SoSH Football Central’s Daryl Sng broke down Sharp’s aforementioned data errors in even greater detail. After excluding kick and punt returns (which make no sense to include because teams don’t have any access to “K balls”) and correcting for Sharp’s original mishmash of regular-season and playoff data, the players in Sharp’s sample fumbled only about 23 percent more as non-Patriots, not 88 percent as was originally stated.
• Political scholar Bill Herman also zeroed in on Sharp’s analysis of individual players’ fumble rates with the Patriots and other teams, identifying its aforementioned methodological errors. In addition, he looked at the six players featured in Michael Salfino’s Wall Street Journal article based on Sharp’s work, finding that their difference in fumbling was statistically significant. Of those six players (Danny Amendola, BenJarvus Green-Ellis, Danny Woodhead, Wes Welker, Brandon LaFell and LeGarrette Blount), four were common to Sng’s dataset, but both analyses found a 23 percent increase in fumbling while playing for teams other than New England.
• Like Matthews and Lopez, data analyst Tom Hayden repudiated Sharp’s assumption that his “plays per fumble” metric was normally distributed across NFL teams (a necessary condition for the 1-in-16,234 claim).
• The harshest counterargument belonged to data scientist Drew Fustin. Fustin challenged Sharp’s choice to exclude dome teams (Sharp’s own post says outdoor teams barely fumble more often than those based in domes), instead looking at fumble rates across all teams in outdoor games only — whereupon the Patriots don’t even rank first in the NFL at fumble avoidance over the 2010-2014 period. He also questions whether Sharp’s decision to use that 2010-14 period was a case of cherry-picking the timeframe that would make the Patriots look most like an extreme outlier.

And much like the radio in the helmet nonsense, I think we are done here.

just because the initial study was poorly performed (which I was critical of when it first came out), doesn't mean it's wrong.

The Wyatt Earp defense is particularly interesting. In essence, it says that an odd occurrence isn't that odd because someone has to be the outlier. That's all fine and dandy, but it doesn't mean that it isn't in fact an outlier.

The 538 guy could have covered the same ground by simply stating that correlation <> causation.

 

[Eminence]

Oh, I get it. You're suggesting the Patriots fumble less because they're using deflated balls.

 

[12punch]

dude, the myth has been busted already

you trolls need to move on to the next bogus 'study' or wikileak, or wtfever

 

[Doctor Detroit]

dude, the myth has been busted already

you trolls need to move on to the next bogus 'study' or wikileak, or wtfever

Or do like Pats fans and start threads about Ray or Jerry Lewis killing dudes.

 

[KellysHeroes]

Most of those fumbles are Ridley's

 

[JIslander]

Most of those fumbles are Ridley's

So if a Ridley "correction" is made, the data looks more damning for Pats

 

[Dinsy Ejotuz]

There are some good criticisms of this piece (presentation, including kick-returners, etc), but the Wyatt Earp effect doesn't really apply at all IMO.

This guy had a hypothesis, checked ONE variable and found it significant with ~99% probability (-2.71 sds per the revised data) for the team in question in his hypothesis.

Even if you don't assume a normal distribution (maybe, maybe not) his data would probably be significant. Regardless, he didn't just jam a bunch of stuff into a hopper and pull the single outlier.

 

[TheAristocrat]

This is also ridiculous but I suppose haters have to hate somewhere/something.

(Not a Pats fan)

 

[parrot]

Most of those fumbles are Ridley's

So if a Ridley "correction" is made, the data looks more damning for Pats

A Ridley "correction" was made, by Bill Belichick, yet people still wonder why his players don't fumble.

 

[Old Smiley]

There are some good criticisms of this piece (presentation, including kick-returners, etc), but the Wyatt Earp effect doesn't really apply at all IMO.

This guy had a hypothesis, checked ONE variable and found it significant with ~99% probability (-2.71 sds per the revised data) for the team in question in his hypothesis.

Even if you don't assume a normal distribution (maybe, maybe not) his data would probably be significant. Regardless, he didn't just jam a bunch of stuff into a hopper and pull the single outlier.

That's the whole point of the Wyatt Earp effect. Starting from an outlier and working back is BAD statistics, which is what this study is. Hearing it from a lot of numbers people now.

 

[Dinsy Ejotuz]

There are some good criticisms of this piece (presentation, including kick-returners, etc), but the Wyatt Earp effect doesn't really apply at all IMO.

This guy had a hypothesis, checked ONE variable and found it significant with ~99% probability (-2.71 sds per the revised data) for the team in question in his hypothesis.

Even if you don't assume a normal distribution (maybe, maybe not) his data would probably be significant. Regardless, he didn't just jam a bunch of stuff into a hopper and pull the single outlier.

That's the whole point of the Wyatt Earp effect. Starting from an outlier and working back is BAD statistics, which is what this study is. Hearing it from a lot of numbers people now.

"Starting with an outlier" usually involves looking at many (hundreds, thousands) variables/data points, grabbing the ones furthest from the norm, and then trying to come up with an explanation. It doesn't work (usually) because when you look at, say, 10,000 outcomes 10 of them are going to be 1000:1 shots just on random chance.

That's (apparently) not what this guy did at all.

ETA: I'm not defending his overall finding -- just saying that if you look at ONE thing and it's a huge outlier it's different than "starting with an outlier".

 

[ROCKET]

Good thing BB figured out a way for the ball to bounce back to a Patriot after coughing it up 3 times vs the Ravens. I'm sure if someone got their helmet on the ball that 2 lbs. less of psi would prevent a fumble every time. This whole thing is officially ridiculous.

 

[parrot]

There are some good criticisms of this piece (presentation, including kick-returners, etc), but the Wyatt Earp effect doesn't really apply at all IMO.

This guy had a hypothesis, checked ONE variable and found it significant with ~99% probability (-2.71 sds per the revised data) for the team in question in his hypothesis.

Even if you don't assume a normal distribution (maybe, maybe not) his data would probably be significant. Regardless, he didn't just jam a bunch of stuff into a hopper and pull the single outlier.

That's the whole point of the Wyatt Earp effect. Starting from an outlier and working back is BAD statistics, which is what this study is. Hearing it from a lot of numbers people now.

"Starting with an outlier" usually involves looking at many (hundreds, thousands) variables/data points, grabbing the ones furthest from the norm, and then trying to come up with an explanation. It doesn't work (usually) because when you look at, say, 10,000 outcomes 10 of them are going to be 1000:1 shots just on random chance.

That's (apparently) not what this guy did at all.

ETA: I'm not defending his overall finding -- just saying that if you look at ONE thing and it's a huge outlier it's different than "starting with an outlier".

Or in this case, 32, then eliminating a quarter of them based on specious criteria.

 

[Old Smiley]

Good thing BB figured out a way for the ball to bounce back to a Patriot after coughing it up 3 times vs the Ravens. I'm sure if someone got their helmet on the ball that 2 lbs. less of psi would prevent a fumble every time. This whole thing is officially ridiculous.

That's been my thinking. In fact, I would think that a less inflated ball would be more prone to "squirt" out of there.

 

[ConstruxBoy]

I've realized that Tom Brady is Jimmie Johnson and Bill Belichick is Chad Knaus. Everyone in NASCAR is doing some cheating, but Jimmie and Chad are just better at cheating than any other team. So they win more often. And it's hard for the the other teams to complain about it because they are cheating as well, just not as much or as effectively as the 48 team.

The Patriots are basically the same thing. That's why you don't see many teams jumping on them for Spygate of Deflategate. They do those things too, just not as often or as well as the Patriots. Some people say that's OK. I personally think it sucks. I also think it lessens the accomplishments of the Patriots and the 48 team a little bit. However, Championship banners fly forever and people will forget about the cheating over time. So it's probably a smart, valued move. I just don't like it.

 

[12punch]

I've realized that Tom Brady is Jimmie Johnson and Bill Belichick is Chad Knaus. Everyone in NASCAR is doing some cheating, but Jimmie and Chad are just better at cheating than any other team. So they win more often. And it's hard for the the other teams to complain about it because they are cheating as well, just not as much or as effectively as the 48 team.

The Patriots are basically the same thing. That's why you don't see many teams jumping on them for Spygate of Deflategate. They do those things too, just not as often or as well as the Patriots. Some people say that's OK. I personally think it sucks. I also think it lessens the accomplishments of the Patriots and the 48 team a little bit. However, Championship banners fly forever and people will forget about the cheating over time. So it's probably a smart, valued move. I just don't like it.

yeah, but your opinion doesn't count for 2 craps, so I'm sure everybody with banners will be hanging them with or without your approval.

the world doesn't actually turn around you, believe it or not.

in the meantime, if you happen to have any worthwhile football info take advantage of this forum to post it.

 

[ConstruxBoy]

I've realized that Tom Brady is Jimmie Johnson and Bill Belichick is Chad Knaus. Everyone in NASCAR is doing some cheating, but Jimmie and Chad are just better at cheating than any other team. So they win more often. And it's hard for the the other teams to complain about it because they are cheating as well, just not as much or as effectively as the 48 team.

The Patriots are basically the same thing. That's why you don't see many teams jumping on them for Spygate of Deflategate. They do those things too, just not as often or as well as the Patriots. Some people say that's OK. I personally think it sucks. I also think it lessens the accomplishments of the Patriots and the 48 team a little bit. However, Championship banners fly forever and people will forget about the cheating over time. So it's probably a smart, valued move. I just don't like it.

yeah, but your opinion doesn't count for 2 craps, so I'm sure everybody with banners will be hanging them with or without your approval.

the world doesn't actually turn around you, believe it or not.

in the meantime, if you happen to have any worthwhile football info take advantage of this forum to post it.

 

[GoBirds]

I've realized that Tom Brady is Jimmie Johnson and Bill Belichick is Chad Knaus. Everyone in NASCAR is doing some cheating, but Jimmie and Chad are just better at cheating than any other team. So they win more often. And it's hard for the the other teams to complain about it because they are cheating as well, just not as much or as effectively as the 48 team.

The Patriots are basically the same thing. That's why you don't see many teams jumping on them for Spygate of Deflategate. They do those things too, just not as often or as well as the Patriots. Some people say that's OK. I personally think it sucks. I also think it lessens the accomplishments of the Patriots and the 48 team a little bit. However, Championship banners fly forever and people will forget about the cheating over time. So it's probably a smart, valued move. I just don't like it.

yeah, but your opinion doesn't count for 2 craps, so I'm sure everybody with banners will be hanging them with or without your approval.

the world doesn't actually turn around you, believe it or not.

in the meantime, if you happen to have any worthwhile football info take advantage of this forum to post it.

You have to understand he's had to have been on a 10 day coke bender with the amount of posting he's done in the cheating threads and is probably losing patience.

Your analogy is probably pretty accurate but some want to believe it's 100% BB is wicked smaat. It does hang the banners though....even if they got caught for something else they won't have to give the trophies back so maybe it's not that bad an idea.

 

[moleculo]

I've realized that Tom Brady is Jimmie Johnson and Bill Belichick is Chad Knaus. Everyone in NASCAR is doing some cheating, but Jimmie and Chad are just better at cheating than any other team. So they win more often. And it's hard for the the other teams to complain about it because they are cheating as well, just not as much or as effectively as the 48 team.

The Patriots are basically the same thing. That's why you don't see many teams jumping on them for Spygate of Deflategate. They do those things too, just not as often or as well as the Patriots. Some people say that's OK. I personally think it sucks. I also think it lessens the accomplishments of the Patriots and the 48 team a little bit. However, Championship banners fly forever and people will forget about the cheating over time. So it's probably a smart, valued move. I just don't like it.

NASCAR, #####ES!

#micdrop

 

[moleculo]

A much better statistical analysis, by someone who knows what the hell he is doing. Nearly all of my objections from the original analysis has been addressed. Draw your own conclusions:

http://www.advancedfootballanalytics.com/index.php/home/analysis/team-analysis/227-a-look-at-ne-s-fumbles

 

[ookook]

I have seen no analysis that did not have the Patriots fumbling similar to the league prior to 2006 and more after. Some close the gap for the latter, but none eliminate it. Begging the details of the analysis (include Qbs or not? Special teams or not?) is one thing, but re-analysis has not reversed the direction of the trend.

 

[The Great Snook]

I think this may be a case of "statistics lie". I think a more plausible answer which anyone who is a NE fan or has owned Ridley in an FFL league can tell you is that Belechick doesn't play people who put the ball on the ground. If you can't hold on, you don't play.

 

[Morton Muffley]

I have seen no analysis that did not have the Patriots fumbling similar to the league prior to 2006 and more after. Some close the gap for the latter, but none eliminate it. Begging the details of the analysis (include Qbs or not? Special teams or not?) is one thing, but re-analysis has not reversed the direction of the trend.

Ok, then check out Drew Fustin's analysis. He is a data scientist who shows that the exclusion of NO and Atlanta for all the wrong reasons makes a huge diff. Basically instead of removing dome teams, Fustin removes games played indoors. The result is that Not only is NE NOT an outlier, they don't even lead the league in fewest fumbles over this period! In other words, NO's and Atlanta are even better at not fumbling when PLAYING OUTDOORS!!! So again, looks like sharp excluded teams that are just as good as NE at not fumbling simply because they (NO and ATL) are good at preventing fumbles whether they are playing indoors or outdoors.

 

[SteelerMurf]

Wouldn't it be interesting if they had fumblitis the first game when balls will be guarded from before to after game?

 

[ookook]

I wasn't relying on Sharp's analysis. I read Fustin's as well.

Re-introducing dome teams still has Patriots as: (a) increasing dramatically after 2006; (b) being second in fumbles (behind ATL and in front of NO, with all three lookig like "outliers", whatever that means); and © the odds of the discrepencies occurring by chance is very small, both within the team over the years and across teams since 2006.

It doesn't prove anything, but is consistent with the observation that the reduction happened when teams were allowed to manipulate balls on their own and that they had a large number of underinflated balls at halftime. All analyses I have seen show exactly what we'd expect if a team had an advantage, but extremely unusual and unlikely to be observed by chance alone, and that advantage started after the 2006 change in rules that Brady lobbied for. The timing of the shift is suspicious.

Brian Burke did regression analysis to model total number of offensive (special teams is excluded) fumbles based on each team's completions, incompletions, sacks, and run attempts. He also took into account indoor/outdoor in his model rather that excluding any teams. This allowed him to display number of fumbles beyond what would be predicted. For the 7 years before the change in rules, the Patriots fumbled on offense 2 times more than would have been predicted. For the 8 years following, they fumbled about 5.5 times less than would be expected. Atlanta fumbled on offense less that would be predicted 4 of the 7 years before the change in rules and 5 of 8 after. New England only once in the 7 years before, 7 of 8 after. The particular trend is much more profound for New England.

As Burke points out: "It's not like we're scanning the stats of all 32 teams to find unusual patterns over any period and then making accusations of improper behavior. In this case, the accusations already exist based on non-statistical evidence and we're scanning NE's stats in a certain period to see if they're consistent or inconsistent with the accusations."

http://www.advancedfootballanalytics.com/index.php/home/analysis/team-analysis/227-a-look-at-ne-s-fumbles

 

[Dinsy Ejotuz]

As Burke points out: "It's not like we're scanning the stats of all 32 teams to find unusual patterns over any period and then making accusations of improper behavior. In this case, the accusations already exist based on non-statistical evidence and we're scanning NE's stats in a certain period to see if they're consistent or inconsistent with the accusations."

http://www.advancedfootballanalytics.com/index.php/home/analysis/team-analysis/227-a-look-at-ne-s-fumbles

This is what I was saying up thread. A lot of people who should know better (because if *I* know...) were invoking the whole "Wyatt Earp" thing when it absolutely didn't apply.

 

[Green and Gold]

Wouldn't it be interesting if they had fumblitis the first game when balls will be guarded from before to after game?

No fumbles, but too small of a sample size in any case.