Replies

And that fumble thing has been totally debunked multiple times. It’s just bad math and stupid thinking. [Discostu, post #30]

He’s already been shown the fumble statistics in the article are a mixture of cherry picking and junk. (Lepton, post #40)

Are these fumble stats "bad math?" "Cherry picking?" Junk stat science? If it's been debunked, by whom? When? Where? Did they use the same variable comparison suggested by two FREEPERs on this thread -- Edzo4 & WinMod70?

Take a look @ what they said:

I wonder if bradys improvement in 2007 had anything to do with having randy moss as a receiver? Instead of the non hall of fame receivers he had before. (Edzo4, post #28)

There are many variables in the game of football, the most significant are talent and preparation. You portend that the only change in the history of Patriots football is the ball. Really? (WinMod70, post #19)

In the above posts, both Edzo4 & WinMod70 suggested that the hyper completion stats Brady had in 07 vs. 06 was due to wide receiver personnel (Edzo mentions Randy Moss, who played full season wise 07 to 09 for Pats); & WinMod mentions how "talent" is a key consideration.

Indeed, personnel is a vitally important variable -- one that can't be understated.

I wanted to see if a personnel study of the Patriot backfield would indeed "debunk" the whole thing. Instead of generic stats flying around, let's just boil this down to specific players and crunch the numbers! (Let's not only apply personnel justifications to Brady's sudden completion % explosion in 2007 -- but to the fumble-free per touch ratio as well.)

There's two ways to do that:
1. You review how Patriot RB did up thru 2006 and then compare it to 2007 & beyond when the NFL changed the footballs supplies' rules.
2. You review how Patriot RB did anytime between 2007 -- 2014 -- as long as you have some Patriot years from that same period.
In both cases above, the player's performance is being measured against himself -- how he did either in a Patriot vs. non-Patriot uniform (#2), or, how he did in the same Patriot uniform on either side of that 2006-2007 divide.

Also, this study then aggregated ALL of those RB stats. In case, somebody might claim that the results could be skewed by including fumble figures from the first two seasons of a player's career, a time when they might be more inclined to fumble, within the second chart below, I segmented those out. (It wound up dropping a couple of "primary" comparison backs into the "marginal" area -- and wound up dropping Danny Woodhead altogether out of the "marginal" category.

Upon reviewing Patriot rosters...which RB qualified for such a case study?

RB who didn't make the "cut" due to not making either of the above criteria:
a. Bolden, Vereen & Ridley didn't play before 07 & haven't played for other teams...certainly, somebody can add them to the future eval list if they sign with other teams.
b. Corey Ridley retired after 06 season.

Who made this case study cut?

Nine Patriot RBs fit the initial criteria; yet I had to place 4 of them in the "marginal" category as having a questionable enough carries for "measurable weight" purposes...their stats might be helpful in reviewing on a "collective" basis, but not so much on a personal comparison.
1. Heath Evans (only 45 carries for NE; yes, no fumbles, in 07-08) vs. 3 fumbles in 119 carries elsewhere
2. Fred Taylor (106 carries, only 1 fumble for 09-10 NE; and yes a better non-fumble ratio than elsewhere ... 26 fumbles in 2428 attempts ... but not significant difference and hard to compare vs. only 106 carries)
3. Lamont Jordan (80 carries, only 1 fumble for 08 NE; not only not enough carries, but this is a slight flip of Taylor the other way in that he didn't fumble much elsewhere -- 8 fumbles, 817 carries...but again not enough significant difference)
4. Danny Woodhead (more of a RB receiver; only 24 carries for '13 chargers...only 35 carries for NE)

That left 5 RBs who have a significant number of carries (over 200 on each side) to serve as the best players to measure:
1. Laurence Maroney (add 07-09 NE, 407 carries vs. 06 NE, '10 Denver, 211 carries)
2. BenJarvus Green-Ellis (08-11 NE, 510 carries vs. 12-13 Cincy, 498 carries)
3. LeGarrette Blount (13-14 NE, 213 carries vs. 10-12 TB & 14 Pit -- played for 2 teams in 14, 491 carries)
4. Kevin Faulk (99-06 NE, 632 carries vs. 07-11 NE, 232 carries)
5. Sammy Morris (07-10 NE, 334 carries vs. 00-06 Buff & Miami + '11 Dallas, 402 carries overall)

So how did this "Case Study" turn out?

(See next post!)