DC Tom Posted January 27, 2015 Posted January 27, 2015 papazoid just sent me this link.... http://drewfustin.com/2015/01/27/patriots-fumble-comments/ Now, maybe Tom and other smart guys understand this article, I really do not understand some of the statiscal equations.... but i found it interesting none the less. Seems like Sharp is getting less and less credibility...that cant be good for business I don't necessarily agree with his analysis...but I do his conclusion. Mostly, that's an excellent example of how to lie with statistics. He disagrees with the original methodology, so he picks a different methodology and comes to a different conclusion. I'm not saying he IS lying...just that he shows how dependent a statistical conclusion can be on a statistical methodology, which is what makes it so damned easy to lie with statistics. Good thing to remember the next time you read...anything in the news, really.
plenzmd1 Posted January 27, 2015 Posted January 27, 2015 How can you not consider that? "Let's lump all the games together to see if they cheated in some!" What sort of a moron doesn't see that? And I'm the idiot. Sure. Haha..guess my sarcasm did not come through BTW, if you get a chance, can you look at the link i posted above? Guy is a Physicist at Chicago looks like...would love your opinion on if his theories, equations, etc pass the sniff test...all of it way to far above my cognitive abilities. Thanks in advance I don't necessarily agree with his analysis...but I do his conclusion. Mostly, that's an excellent example of how to lie with statistics. He disagrees with the original methodology, so he picks a different methodology and comes to a different conclusion. I'm not saying he IS lying...just that he shows how dependent a statistical conclusion can be on a statistical methodology, which is what makes it so damned easy to lie with statistics. Good thing to remember the next time you read...anything in the news, really. Thank you
TheBillsWillRiseAgain Posted January 27, 2015 Posted January 27, 2015 This has NEVER been about QB fumbles specifically -- it's been about fumbles in general. I think it's safe to presume that no one is arguing that softer balls prevent strip sacks (it would be a foolish argument). In theory, such balls are intended to help RBs and other ball carriers. The fact of the matter, however, is that QBs virtually always lead their teams in fumbles because of blindside sacks, taking too long to get rid of the ball, and wild scrambles -- not because of qb sneaks, which is what most of Brady's runs are. Consequently, such fumbles have to be factored into the total. But neither Brady nor Manning is a fumbler. They're careful with the ball. The times they do fumble represent the inevitable outcomes of an occupational hazard. Their teams don't fumble much collectively as a consequence. Also, you seem to not be getting my point about statistical insignificance. 0.019 is greater than 0.015. But it's an insignificant difference. I suggest looking at the bad team fumble and sack rates. Look at the Jets, for instance. Look at Washington. Alternatively, look at Seattle. They fumble a lot more, and it has nothing to do with the ball difference. It's because their QBs are careless with the ball, hold onto it too long, scramble wildly a lot, and/or have bad pocket sense. Comparing Brady and Manning on the fumbling front is like comparing Godzilla vs. Mothra. They're outliers because they're really, really good at avoiding sacks and taking care of the ball. I think we can agree that we're not going to agree on this.
DC Tom Posted January 27, 2015 Posted January 27, 2015 Haha..guess my sarcasm did not come through BTW, if you get a chance, can you look at the link i posted above? Guy is a Physicist at Chicago looks like...would love your opinion on if his theories, equations, etc pass the sniff test...all of it way to far above my cognitive abilities. Thanks in advance Thank you I should have added that his analysis is still wrong, because he still didn't address the central thesis that the rule change giving the away team control of their own balls led to the decrease in fumbles. I haven't heard anyone make any sort of reasonable argument either way of any benefits in domes vs. outdoors - and if they did, I'd criticize them for thinking that's as meaningful a difference as though a 70 degree and sunny with no wind day in an outdoor field is all that much more hostile an environment than a dome. Really, the only statistical analysis that anyone can do that isn't going to be bull **** is to correlate fumbles against at least the following variables: home vs. away, temperature on the field, precipitation on the field, wind on the field (i.e. I don't care what it's doing in the parking lot, I want to know what it's doing inside the stadium - which could be wildly different, particularly in a dome), and condition of the field. And then adjust for the competition they faced - e.g. if a team somehow plays ten games against the top eight ball-stripping teams in the league, that skews the correlation. And eliminate plays that are obvious outliers (e.g. Megatron's "fumble" in the end zone - that's him intentionally dropping the ball, and obviously has nothing to do with anything else. Or any time a guy's knocked unconscious and drops the ball.) And exclude special teams plays, since they use a different set of balls provided by the league. That is probably the minimum you need to do to find anything resembling a meaningful trend that isn't bull ****. Basically, it's a multivariate problem that you can't solve simply by cherry-picking a couple of variables and plotting them in Excel. And no, I am not going to do it.
NoSaint Posted January 27, 2015 Posted January 27, 2015 or maybe the data now goes back to 2007, the EXACT year when teams could start playing with their own footballs...prior to that, with the genius BB at the helm, the Pats were right there in the middle of pack. I know,, a minor coincidink... http://www.sharpfootballanalysis.com/blog/2015/the-new-england-patriots-mysteriously-became-fumble-proof-in-2007 they could before that too - just not on the road. right? home team did everything.
plenzmd1 Posted January 27, 2015 Posted January 27, 2015 And no, I am not going to do it. sissy they could before that too - just not on the road. right? home team did everything. Saint, i see you catching up in the thread now... BTW, I am sending you and Kirby PMs
NoSaint Posted January 27, 2015 Posted January 27, 2015 (edited) papazoid just sent me this link.... http://drewfustin.com/2015/01/27/patriots-fumble-comments/ Now, maybe Tom and other smart guys understand this article, I really do not understand some of the statiscal equations.... but i found it interesting none the less. Seems like Sharp is getting less and less credibility...that cant be good for business its the basic questions from the get-go. The "study" was way oversimplified. It was interesting but given the info you couldnt make any rational conclusion. the point may prove true down the line, but as this blog points out, there are a lot of basic questions it doesnt answer, and shortcuts that probably shouldnt have been taken. our playoff streak is statistically crazier than the fumbles, I bet Edited January 27, 2015 by NoSaint
Dibs Posted January 27, 2015 Posted January 27, 2015 Posted this in other threads....can see it best placed here.... """"" I think that the "Patriots don't get sacked much" devil's advocate argument for the reason that they have an extremely low fumble percentage is fair.....so I thought I would crunch some numbers and see what happens.... Below are Tom Brady's and Matt Cassel's numbers through their careers. I have decided to use the combined sack and rush numbers as those were the plays where the QB theoretically has to hold onto the ball. Source: NFL.com player stats Cassel(http://www.nfl.com/p...562/careerstats) NFL.com player stats Brady(http://www.nfl.com/p...211/careerstats) Matt Cassel fumble rates with Patriots(2007+2008): 124 plays 8 fumbles (1 in 15.5) 4 fumbles lost (1 in 31) Matt Cassel fumble rates pre-2007 with Patriots and post Patriots(2009+) 305 plays 40 fumbles (1 in 7.6) 15 fumbles lost (1 in 20.3) Tom Brady fumble rates 2007+ 413 plays 36 fumbles (1 in 11.5) 15 fumbles lost (1 in 27.5) Tom Brady fumble rate pre-2007 421 plays 59 fumbles (1 in 7.1) 25 fumbles lost (1 in 16.9) It looks like there is a clear reduction in the fumble rate for the Patriots QBs after the 2006 season. """""
BRH Posted January 27, 2015 Posted January 27, 2015 It looks like there is a clear reduction in the fumble rate for the Patriots QBs after the 2006 season. """"" And we can attempt to explain this away via a multivariate analysis, or we can use Occam's Razor.
Alphadawg7 Posted January 27, 2015 Posted January 27, 2015 Posted this in other threads....can see it best placed here.... """"" I think that the "Patriots don't get sacked much" devil's advocate argument for the reason that they have an extremely low fumble percentage is fair.....so I thought I would crunch some numbers and see what happens.... Below are Tom Brady's and Matt Cassel's numbers through their careers. I have decided to use the combined sack and rush numbers as those were the plays where the QB theoretically has to hold onto the ball. Source: NFL.com player stats Cassel(http://www.nfl.com/p...562/careerstats) NFL.com player stats Brady(http://www.nfl.com/p...211/careerstats) Matt Cassel fumble rates with Patriots(2007+2008): 124 plays 8 fumbles (1 in 15.5) 4 fumbles lost (1 in 31) Matt Cassel fumble rates pre-2007 with Patriots and post Patriots(2009+) 305 plays 40 fumbles (1 in 7.6) 15 fumbles lost (1 in 20.3) Tom Brady fumble rates 2007+ 413 plays 36 fumbles (1 in 11.5) 15 fumbles lost (1 in 27.5) Tom Brady fumble rate pre-2007 421 plays 59 fumbles (1 in 7.1) 25 fumbles lost (1 in 16.9) It looks like there is a clear reduction in the fumble rate for the Patriots QBs after the 2006 season. """"" You mean in 2007 when they added Randy Moss and Welker and broke all the offensive records...but I guess having immensely more talent had nothing to do with it, must be the balls
CardinalScotts Posted January 27, 2015 Posted January 27, 2015 its not the qb fumbling it's the other players on NE with the easier to grip ball....qb strip sack or blind side rush not a lot you can do
Dibs Posted January 27, 2015 Posted January 27, 2015 You mean in 2007 when they added Randy Moss and Welker and broke all the offensive records...but I guess having immensely more talent had nothing to do with it, must be the balls The plays I looked at were QB runs and sacks. The WRs never touched the balls in the plays that I listed. I showed the number of times per play where the QB fumbled(on plays where they were likely hit).
Alphadawg7 Posted January 27, 2015 Posted January 27, 2015 Tom Brady: 2007 sacked 21 times Matt Cassell: 2008 sacked 47 times, tied for most in league behind same line Tom Brady: 2009 sacked 16 times. Its pretty damn clear that Brady is a large reason why they don't turn the ball over as much in both INTs and Fumbles. He clearly, without a shadow of a doubt takes substantially less hits because of his skill. He is considered one of the best operators inside the pocket of all time for a reason. And the most annoying thing here is this so called dramatic turn is being attributed to the ball and ignoring its the same year they added Randy Moss and Welker and they went off like no other team in history. If you think that was because of the ball and not because of 2 of the greatest players of all time in Moss and Brady, not to mention one of the most productive WRs of this generation in Welker then I don't know what to tell you. Stats already proved that the difference in the Pats and Manning led teams is negligible at best. The truth of the matter is that pre 2007, the NE offense was made up of marginal talent at best outside of Brady. 2007 and on that team for the most part has at a high level of talent on the offense, and at times, Elite levels of talent. BB became stricter and has no patience for guys who turn the ball over either. But if I go by what you guys say then the conclusion is that Peyton Manning cheats too because his team fumbles at a very similar rate. That bastard.
BillnutinHouston Posted January 27, 2015 Posted January 27, 2015 (edited) You mean in 2007 when they added Randy Moss and Welker and broke all the offensive records...but I guess having immensely more talent had nothing to do with it, must be the balls Yes, now you're getting it...sort of. Good mention of Welker - he was one of the six who fumbled twice as often when not in a Pats** uniform as he did when in one. Edited January 27, 2015 by BillnutinHouston
papazoid Posted January 27, 2015 Posted January 27, 2015 (edited) papazoid just sent me this link.... http://drewfustin.com/2015/01/27/patriots-fumble-comments/ Now, maybe Tom and other smart guys understand this article, I really do not understand some of the statiscal equations.... but i found it interesting none the less. Seems like Sharp is getting less and less credibility...that cant be good for business i've been hacked...(laffin)...but i'm serious....i didn't send you anything...it wasn't me... Edited January 27, 2015 by papazoid
Dibs Posted January 27, 2015 Posted January 27, 2015 ...... Its pretty damn clear that Brady is a large reason why they don't turn the ball over as much in both INTs and Fumbles. He clearly, without a shadow of a doubt takes substantially less hits because of his skill. He is considered one of the best operators inside the pocket of all time for a reason. ..... Quite likely true......but what do you think explains the fact that he fumbled substantially less times per hit after 2006.....or that Matt Cassel's numbers followed suit and actually fumbled less times per hit than Brady did?
Alphadawg7 Posted January 27, 2015 Posted January 27, 2015 (edited) Yes, now you're getting it...sort of. Good mention of Welker - he was one of the six who fumbled twice as often when not in a Pats** uniform as he did when in one. Ahhh, so you are one of those guys who makes things up are you...weird, his career stats of him playing 4 years not with the Pats...2 in Miami and 2 in Denver....suggest otherwise. http://espn.go.com/nfl/player/stats/_/id/5941/wes-welker Lets see, as a WR he has ZERO...again ZERO fumbles outside of NE. Not one fumble as a WR in Denver or Miami in 4 years. He had SIX with NE. Not sure where you went to school, but at my school Zero is not more than 6 and its certainly not "twice as often". So how can he fumble twice as often away from NE when he's never fumbled a single time away from NE in his entire career? lol On a side note: He has 11 fumbles as a returner outside NE and another 5 as a returner with NE. However, returners don't return balls that his QB uses, they returns the ball of the other team and its a K ball. So they have no relevance in this discussion as he would not have been returning a "deflated" ball as it was the opposing teams ball and their K ball on top of that. So actually, Welker is significant evidence that the Pats are better at teaching ball control, because Welker cut his fumbles down big time in the return game once he got to NE. Edited January 27, 2015 by Alphadawg7
plenzmd1 Posted January 27, 2015 Posted January 27, 2015 i've been hacked...(laffin)...but i'm serious....i didn't send you anything...it wasn't me... I know, but it supported your argument better than Pioli left....so i thought i would help ya out!
Alphadawg7 Posted January 27, 2015 Posted January 27, 2015 Quite likely true......but what do you think explains the fact that he fumbled substantially less times per hit after 2006.....or that Matt Cassel's numbers followed suit and actually fumbled less times per hit than Brady did? Hits on the QB don't just equate to fumbles...everyone is different, everyone has different way of holding the ball, everyone has different size hands, etc. Its not black and white. Furthermore, the offense Brady ran from 2007 was one of the best offenses in history and since then has been among the best in the league most years. Its immensely more talented than what he had 2000-2006. Better teams protect the ball, its also why they are usually better teams. And it didn't massively drop in one year, it gradually kept going down...which coincides with the offense getting better and having more talent. And again, the stats are very close to the same fumble rates of Peyton Mannings team. So to single out the Pats makes no sense in that regard.
Dibs Posted January 27, 2015 Posted January 27, 2015 Hits on the QB don't just equate to fumbles...everyone is different, everyone has different way of holding the ball, everyone has different size hands, etc. Its not black and white. Furthermore, the offense Brady ran from 2007 was one of the best offenses in history and since then has been among the best in the league most years. Its immensely more talented than what he had 2000-2006. Better teams protect the ball, its also why they are usually better teams. And it didn't massively drop in one year, it gradually kept going down...which coincides with the offense getting better and having more talent. And again, the stats are very close to the same fumble rates of Peyton Mannings team. So to single out the Pats makes no sense in that regard. That's all well and good.........but what do you think explains the fact that he fumbled substantially less times per hit after 2006.....or that Matt Cassel's numbers followed suit and actually fumbled less times per hit than Brady did? I'll do the pre-2007/post-2006 stats for Manning later if you like. Maybe they too will show a substantially lowering in fumble rate.
Recommended Posts