for display only
Big Blue Interactive The Corner Forum  
Back to the Corner

Archived Thread

A Small Reason To Be Optimistic - Fumble Luck

Bockman : 2/24/2015 12:26 pm
While some football fans and talking heads are slow to come around to challenging long-held cliches about football, one area that is gaining traction is how we look at fumbles.

It has been proven quite definitively over the past decade or so that while CAUSING fumbles is a skill, RECOVERING them is largely a random event. What this means in the long term is that teams who had bad fumble "luck" one season have a tendency to regress back to the mean the following season. More fumbles recovered can obviously go a long way towards an extra win or two in close games.

This article is about 10 days old at this point, but not sure if people knew this fact:
"The 49ers and Giants were the two unluckiest teams this year when it comes to fumbles. Here’s how to read the New York line: The Giants fumbled 21 times, and lost 14 of them; given the 46% fumble loss rate, that means New York lost 4.4 more fumbles than expected. The Giants opponents fumbled 24 times, but lost just 9 of them; this means that Giants opponents lost two fewer fumbles than expected. As a result, the Giants recovered 6.4 fewer fumbles than expected. There were 45 total fumbles in Giants games in 2014, and New York recovered just 16 of them, for a 35.6% rate."

What a kick in the nuts. Not only did we not recover our own fumbles, but we also had a hard time recovering the opponents' fumbles as well. The article further writes:
"New York and Dallas each fumbled 21 times, and each incredibly lost 14 of those fumbles. Giants fans can blame Eli Manning for just four of those fumbles, as Larry Donnell (4), Quintin Demps (2), Rashad Jennings, Preston Parker, Daniel Fells, and yes, even Odell Beckham, combined for the other ten... Yes, that’s safety Quintin Demps, who fumbled once on an interception return and once on a kickoff return.

Can our luck with fumbles turn around next year? The thing with probabilities is that no matter the odds, its not a guarantee until it hits 100%. However, with stats like the above, there's some reason to be hopeful.


2014 Fumble Recovery Data - ( New Window )
You can say luck  
robbieballs2003 : 2/24/2015 12:31 pm : link
but I remember Spags' first year here he went nuts that loose balls weren't picked up and people weren't constantly running to the ball. I know it sounds elementary but that does help a tremendous amount.
robbie I see what you're saying, but its two different issues  
Bockman : 2/24/2015 12:39 pm : link
running towards the ball doesn't guarantee that you'll recover it either...

Bottom line, it's impossible to predict how an oblong inflated ball will bounce with 5-10 275lb beasts all going after it
I remember one of those Eli fumbles...  
BamaBlue : 2/24/2015 12:52 pm : link
all Jennings had to do was fall on the ball, but instead he tries to pick it up and it's recovered for a TD.
This subject always makes me think of Bradshaw's fumble in SB42  
WideRight : 2/24/2015 12:53 pm : link
He was a rookie. I am certain Coughlin coached him until blue in the face about ball security in the most important game of his life.

Then of course he coughs up the ball on a handoff from Eli, and Pierre Woods falls right on it. That is the essence of fumble luck, but what it fails to account for is the fact that Bradshaw recovered the ball, and it tuns out the Giants won the game.

So fumble luck is just another stat for quants; Giants were losing the fumble war because they sucked, not the other way around. And they will likely get better, not because of "regression to the mean" but because those responsible for their sucktitude will not be on the field anymore.
Witness the Seattle  
RetroJint : 2/24/2015 12:53 pm : link
game. It's not just luck. Seattle's O-Line does the best job in the league of sealing off the play from trailing pursuit. The Giants had three chances to recover fumbles in the drive that sealed the win. The 'Hawks O-Line also is adept at pushing Beast Mode from behind, using maneuvers that are technically illegal. So it's not just luck.

Ah, another Spagnuolo comment, I see. All will be well. From a different perspective consider that in basketball, talent gets the deep rebounds. Know what I mean by that? It's just not desire and rebounding-machine drills. Similarly talent recovers fumbles. Fewell was stuck with many poor players. If most of these return in 15, the Giants defense will blow again. Poor players are poor players because a. They miss assignments b. Lose their individual match-ups & c. Commit penalties because they can't legally perform the task. You can add force & recover fumbles . Not that Fewell's defenses did not get their fair share of turnovers-they always did . It's just if LT uses the Claw to rip the ball from behind and guys like Mark Collins are after the ball, I like the Giants chances.
Robbie  
Semipro Lineman : 2/24/2015 12:58 pm : link
implying that fumble luck can be cured solely by a change in coaching philosophy is wrong and you know it.
With all due respect guys, did you even read the article?  
Bockman : 2/24/2015 1:01 pm : link
This really isn't even a debate - recovering fumbles is random. Period. Already proven by many people through many articles at multiple sites over the past decade.

Guess who was best last year? Miami. Real powerhouse there.

Mighty Seattle? They were only 14th.

Jags and Texans are in the top 5.
Semipro and Bock  
Reale01 : 2/24/2015 1:26 pm : link
I agree with Robbie to an extent. Hustling to the ball on every play will increase your odds of recovering fumbles. Offensive lineman who follow the RB downfield and hustle downfield when a pass is caught will be closer to the ball if it is fumbled. Defensive players who run to the ball even when someone else is making the tackle will be closer to the ball when it is fumbled.

Of course, there is a random element, and an element of luck. It would be completely random if all teams showed the same amount of hustle and determination to be near the ball - but they don't.

The Giants were not a hustling, get-to-the-ball team on either offense or defense last year IMO. The fumble recovery stats seem to support that observation.

You will never convince me - or Robbie - that being close to the ball does not increase your chances. When I coached I always told my players to do your job and then run to the ball. Don't let a fumble surprise you - expect one on every play - you will be there just a little quicker and that can make all the difference. In 10 years, we ALWAYS recovered more fumbles than our opponents and one year it was 80%. Ten consecutive years of excellent luck?

Here's the rub about this  
Semipro Lineman : 2/24/2015 1:31 pm : link
you are implying without evidence that the current coaching staff is not teaching players to run to the ball. So do you believe that the coaching staff of a top defense like the 49er's are also not teaching their guys to run to the ball? I mean they also sucked last year at recovering fumbles so maybe they need "Spags" as well
Also real  
Semipro Lineman : 2/24/2015 1:35 pm : link
what level where you coaching?

To be honest, there is a point in time when running to the ball isn't instinctive for players but as they progress through the various levels upward, it becomes more natural because they have heard it before so often. Same thing for basketball where I have coached a bit. Younger players will occasionally watch a ball go out of bounds.
.  
arcarsenal : 2/24/2015 1:42 pm : link
Spags will DEMAND that they recover the fumbles.
Reale  
Bockman : 2/24/2015 1:46 pm : link
I'm sure you were an awesome coach and no doubt got your kids ready to play - perhaps better than your opponents.

But to piggyback off Semipro's comment, one anecdote doesn't disprove all the research that shows what I said in my OP - that FORCING fumbles is skill and RECOVERY is luck/random/whatever.

MIA, JAX and HOU were all better at recovering fumbles this year than SEA and NE. So I'm sorry but I respectfully disagree that recovering fumbles is something that is coachable at the NFL level.
the giants  
area junc : 2/24/2015 1:55 pm : link
had two of the best "fumble recoverers" I've ever seen in Rich Seubert and Ahmad Bradshaw. i still dont know how bradshaw recovered that fumble in SB42. may have saved the game. ripped it away from ty warren like a pitbull
its not just  
area junc : 2/24/2015 1:56 pm : link
random luck

the giants D had its head up its ass on almost every fumble last year. u dont pay attention, u dont recover
Bockman  
WideRight : 2/24/2015 2:04 pm : link
Stats can fool even those with above average intelligence, so use both sides of your brain in debate.

First, and most obvious, the recovery rate is 46%. Not 50%. So its near-random, but not random. So if its not random, there must be reasons, albeit small.

So now listen to the coaches, and see if what they are saying can be quantitated. It can't. So it fits into that albeit small but measurable difference, nicely, and the world is OK.
RE: its not just  
Semipro Lineman : 2/24/2015 2:21 pm : link
In comment 12150836 area junc said:
Quote:
random luck

the giants D had its head up its ass on almost every fumble last year. u dont pay attention, u dont recover


So once again, did the Seattle defense have its head up its ass because according to that chart, their defense performed worse than the Giants at recovering opponent's fumbles? The same thing with Denver and a few other teams.
Sure, each individual play  
Enoch : 2/24/2015 3:00 pm : link
can and often does include impressive feats of (or failures in) athleticism and/or awareness. The problem is that these factors are of statistically insignificant significance when compared to (1) being in the right place when the ball comes loose and (2) the bounce of an oblong ball not being particularly predictable.

Coaches do emphasize this to varying degrees and with varying techniques, but no NFL coaching staff has been able to maintain a consistently better-than-average rate of fumble recovery over a statistically significant period.
RE: Bockman  
Bockman : 2/24/2015 3:06 pm : link
In comment 12150861 WideRight said:
Quote:
Stats can fool even those with above average intelligence, so use both sides of your brain in debate.

First, and most obvious, the recovery rate is 46%. Not 50%. So its near-random, but not random. So if its not random, there must be reasons, albeit small.

So now listen to the coaches, and see if what they are saying can be quantitated. It can't. So it fits into that albeit small but measurable difference, nicely, and the world is OK.


Again, did you even read the article? The very first paragraph -

There are few statistics more random in all of sports than fumble recoveries. When a football is on the ground, it’s not the case that better teams are more likely to fall on the ball than bad teams: in the NFL, recovering fumbles is nearly all luck and little skill. This is a fact widely accepted by all statisticians, and I also ran a study which confirmed such intuition just last year.
RE: Sure, each individual play  
Bockman : 2/24/2015 3:07 pm : link
In comment 12151001 Enoch said:
Quote:
can and often does include impressive feats of (or failures in) athleticism and/or awareness. The problem is that these factors are of statistically insignificant significance when compared to (1) being in the right place when the ball comes loose and (2) the bounce of an oblong ball not being particularly predictable.

Coaches do emphasize this to varying degrees and with varying techniques, but no NFL coaching staff has been able to maintain a consistently better-than-average rate of fumble recovery over a statistically significant period.


Perfectly said, thank you.
One small factor that could affect recovery rate  
Semipro Lineman : 2/24/2015 3:16 pm : link
is simply the number of fumbles that go out of bounds.
Coaches can't consistently be better than average because...  
WideRight : 2/24/2015 4:07 pm : link
they don't recover fumbles. Players do. Teams are not consistent because the players are not consistent. Every fumble situation is different so no pattern will emerge.

Watch Bradshaw's fumble recovery in SB42 and describe the relative amount of skill and luck involved.

And I read the article. No different than the others: lots of luck and little skill. If you think that means no skill, then you have the reading comprehension problem.

Players have to exploit every possible advantage to win; that small non-random component is key. Approximately 8% (46/50).
Not sure about the eight percent  
Semipro Lineman : 2/24/2015 6:29 pm : link
figure but if you're sure that it can be related to coaching then explain how bad teams with supposedly bad coaches are generating positive fumble recovery numbers.
I tend to believe that recovering a fumble is similar to  
Marty in Albany : 2/24/2015 7:53 pm : link
rebounding in basketball. Some players are better at it than others.

The direction of the bounce may be random, but let's just say there is a football sitting three feet from my grasp and three feet from JPP's grasp, does anyone really think I'll get the ball fifty percent of the time? Does anyone think I'll get the ball even once, and live to tell about it? ;-)
I love the  
Semipro Lineman : 2/25/2015 4:37 pm : link
counter-arguments in this thread
semi pro and Bockman  
Reale01 : 2/25/2015 7:00 pm : link
Sorry I guess I seemed a little full of myself. I do think pursuit and alertness plays a role in the recovery rate. Players in the NFL should be better at running to the ball than they are at the youth and high school levels where I coached. 80% is not reasonable, but you can be better than average at it by being around the ball and anticipating the event rather than reacting to it. Your team WILL be better at getting fumbles if they are better at being around the ball and anticipating.

However, LUCK will always play a part and sometimes it is a large part. You can have 5 players around the ball and have it bounce directly into the hands of the one player from the other team who happens to be there.

The Giants did not seem to be alert and did not seem to swarm to the ball with the same aggression as some teams. Is it the players or the coaching? I don't know. There were several cases last year - Jennings failure to pick the ball up in the end zone being the most blatant.

In the end the point that we have an opportunity to improve is well taken. That would not be a likely possibility if we had recovered 70% last year. To say it is random. If you are around the ball and have average luck you will do better than a team who has fewer players around the ball with average luck.

Back to the Corner