Just skimming...your evaluations are highly dependent on a few picked RBs. ...but at least you eliminated the original articles’ conflation of Brandon Tate Patriots WR fumbles with those as a Bengals kickoff returner (roughly half the total result, as I recall).
BenJarvis Green Ellis didn’t fumble once in college either. Why did he start fumbling with the Bengals? Maybe because they don’t place such an emphasis? (my Bungals are legendary). Cincinnati placed 16th on average for those years.
Tampa Bay placed 19th (zero Blount fumbles in 2012, and 1 with Steelers). In 2011, Blount had but 3 of his teams 30 fumbles. With 30 fumbles, he was the least of their fumbling troubles...and that was his career high, which he also reproduced in 2013 with the Pats. Atlanta, for example totalled 34 fumbles during that 3 year span.
Sammy Morris’ fumbling is almost entirely based on the effects of one year - 4 of Morris 12 career fumbles were in 2006, with the other 8 evenly divided between the Pats and the other non-pats seasons.
Kevin Faulk only fumbled 4 times as a RB after his first two years...three of which were in 2003.
The rest of the RBs are about the same or improved after they leave the Pats.
On the other hand, the patriots will bench even their lead rusher for fumbling since they’ve had the example of BJGE to show that fumbles can be avoided. After one ‘freebee’, Ridley got benched sometimes for multiple games every time he fumbled....or almost fumbled.
The Patriots have only been tops in fumbles twice from 2007-2014. Atlanta has a similar run of low fumbles being in the top 5 all but two years beginning in 2006 - but placing 11th instead of 4th in 2008, and 9th instead of the Patriots 24th in 2013. Indianapolis, from 2004 to 2010 does likewise. Those are the only two I tracked...but it’s not like the Patriots are first every single year.
On the other hand, New England went to a short passing game with picking up Welker in 2007, and continuing with Edelman and Amendola - where Tom Brady would rather throw an incompletion or low pass than lead his receiver into a big hit, which each keeps his receivers healthier and makes them less likely to fumble.
These are verbal demonstrations of how thin your analysis is. Follow the links I posted to you earlier for several different analyses of the mangling of statistics that Sharps employed.
I’m done. Have fun.
Warren Sharp took a broader approach with individual players -- including "touches" (receptions & runs):
New England Patriots Fumble More Often When Playing for Other Teams
His analytics zero in on 19 players who had both NE time & non-NE time...five of which had 300+ touches for the Pats (Welker, Maroney, Green-Ellis, Morris and Woodhead)
Of course, people have rightly id'd one potential considerable flaw in his #s: Those fumble #s reflect special teams' fumbles, which don't belong since teams don't use team-supplied footballs.
(I say potential because I think it looks like it could be a "wash"...for example, yeah, Brandon Tate's special teams fumbles for the Bengals shouldn't be included in his figures; yet guys like Wes Welker's fumble #s also goes down, etc.)
So in looking @ Sharp's charts at the above link, I figured I needed to know which of those 19 players (+ Faulk -- he doesn't include any pre-NE 2007 stats for comparison analysis like I did) -- were involved in special teams returns...and then how many of those fumbles were wrongly included in his analysis.
And, frankly, I was a little surprised to see backs like Faulk having been a special teams returner early in his career -- so even my analysis needs a very slight corrective. Faulk had 3 fumbles as a special teams returner -- but one of them occurred after 2007 so probably won't alter the ratios all that much.
I was surprised to see that many of his fumbles came on pass catches.
Well, anyway, all that info is available online (for people who know how to find it)...so this weekend...took some time to capture it all into one doc:
Mainly, pinpointed which of those 20 players had special teams returns -- and if they committed fumbles on specific seasons -- were they special teams' related (how many?)?
Secondly, Sharp didn't include playoff stats...which, given that the whole thing blew up during playoffs, I think is relevant & broadens the "touches" for measurement comparisons. So I began adding those #s in as well.
All that's left to do now is to crunch the #s and then run the new fumble-ratio #s as an improvement on Sharp's take.
If I get time later to do that tonight, may have that up first thing in #.
(I'm actually surprised nobody else that I could find ran a "corrective" on Sharp's #s to give us the best comparison...but, again, I think -- after tabulated the special teams' based fumbles...it'll probably be a wash or close to it)
Bottom line I've found -- is you do need about all 20 players -- their collective stats totaled -- to see if a huge distinction was looming as a shadow from the past.
Later in the week, it'd also be interesting to do a breakout:
What's the ratio comparison of WR, TE, RB fumbling after a catch vs. a RB rush fumble...and to see if all those #s are also similar. (There were a few backs I didn't want to take the time to delve into their steep history -- like Fred Taylor, for example -- so I don't have all the raw data on all 20 of those players).
Tate has so few Non NE "touches" (only 35) -- that for him to have 11 fumbles with only 35 touches weighs down that side heavily.
In checking ALL 11 of those fumbles were either as a kick or punt returner.
The rest are probably "a wash"...but Sharp blew it to include him without realizing what that was doing to his analytics.