NOTICE: This league is using the BLEEDING EDGE game engine. For more information, click here.

The new user interface is in preview!

Want to check it out? Click here! (If you don't like it, you can still switch back)

League Forums

Main - Bug Box

Avoid fumble vs. fumble percentage regression analysis

By warrior462
9/24/2024 9:37 am
I have had a hunch for a long time that the avoid fumble attribute is essentially meaningless, and does not actually do what it is intended to do. I decided to slog through some data to test that out this morning. While the scope of the analysis is pretty limited since the data is so tedious to gather, the results provide strong evidence towards confirmation of my hunch.

I looked at the 20 non-QB players with the most fumbles at the current point in the German NFL season. I recorded their avoid fumble attribute, and from their number of fumbles and total touches (carries + receptions + kick returns + punt returns) I calculated their fumble percentage. Ideally there would be a strong negative correlation between avoid fumble and fumble percentage.

What I found initially was an extremely weak POSITIVE correlation (meaning, from the data, you would expect a player with higher avoid fumble to fumble the ball MORE often). After throwing out the two most absolutely egregious outliers (94 and 100 avoid fumble with ridiculous 14.3% and 11.8% fumble percentages, respectively), there was an extremely weak negative correlation with R^2=0.167.

I do not think that avoid fumble is the only attribute that has a very loose (if any) connection to actual performance. I would really like to see an effort to make the results of each game more reflective of the quantified talents of the players involved. I'm sick of coming into a game where I've got Julio Jones in his prime being covered by Joe Biden in his current state (I think he's still alive), and Julio comes out of the day catching 2 of 7 for 11 yards and a drop.

Re: Avoid fumble vs. fumble percentage regression analysis

By Cjfred68
9/25/2024 12:50 pm
Playing devils advocate here but a ball carrier with 100 avoid fumble gets tackled by a defensive player with 100 strip ball so what's the formula used to calculate a possible fumble?

Does player fatigue and player hand injury factor into the equation?

Re: Avoid fumble vs. fumble percentage regression analysis

By Pernbronze
9/25/2024 3:55 pm
I've noticed the same. I used to regularly lead the league in fumbles despite having only 100 avoid fumble players so now I just ignore it and my fumbling has dropped.

Re: Avoid fumble vs. fumble percentage regression analysis

By Mcbolt55
9/25/2024 5:11 pm
Cjfred68 wrote:
Playing devils advocate here but a ball carrier with 100 avoid fumble gets tackled by a defensive player with 100 strip ball so what's the formula used to calculate a possible fumble?

Does player fatigue and player hand injury factor into the equation?



It’s got to be one of JDB’s great secrets, like how the best rated qb in the league can’t stop throwing picks, but his backup leads the same roster to a championship. The math doesn’t make sense…

Re: Avoid fumble vs. fumble percentage regression analysis

By Cjfred68
9/25/2024 7:52 pm
I'm gonna sound like a broken record to some but I'm convinced each player has a hidden clutch attribute.

Most players are neutral or even in their clutch attribute while a few have a positive "Brady" rating while others have a negative "Leif" rating.

So some 90+ overall QBs perform bad and lose close games while some lower rated players seem to always out play their attributes.

It's just a theory of mine based on playing 100s of seasons and I have only observational evidence.

In the XFL league we have this QB who is the highest rated QB in the league and he loses often and throws a ton of picks while bouncing around the league to several teams/owners




Re: Avoid fumble vs. fumble percentage regression analysis

By warrior462
9/26/2024 5:25 am
Cjfred68 wrote:
I'm gonna sound like a broken record to some but I'm convinced each player has a hidden clutch attribute.

Most players are neutral or even in their clutch attribute while a few have a positive "Brady" rating while others have a negative "Leif" rating.

So some 90+ overall QBs perform bad and lose close games while some lower rated players seem to always out play their attributes.

It's just a theory of mine based on playing 100s of seasons and I have only observational evidence.


I agree completely. I have seen tons of anecdotal evidence of exactly this, and it is certainly not just QB's. I've had what appeared to be garbage RB's and DE's who put up incredible numbers, while very talented guys in the same system couldn't do a thing. It's very frustrating when what you see doesn't match up with what you get, but that seems to be the norm around here.

Re: Avoid fumble vs. fumble percentage regression analysis

By Mcbolt55
9/26/2024 6:28 am
I guess that sorta simulates reality then too? Plenty of talented guys never make an impact, while fringe players randomly explode into all stars in the right situations…

Re: Avoid fumble vs. fumble percentage regression analysis

By ibblacklavender02
10/13/2024 5:46 am
Man I wonder the same about speed....My players always get caught by Joe Biden...I understand fatigue but man sometimes it's right after a timeout

Re: Avoid fumble vs. fumble percentage regression analysis

By Cjfred68
10/13/2024 7:52 am
Speed is king but ball carry becomes a factor after the player gets the ball and directly effects YAK

Re: Avoid fumble vs. fumble percentage regression analysis

By warrior462
10/13/2024 9:09 am
Cjfred68 wrote:
Speed is king but ball carry becomes a factor after the player gets the ball and directly effects YAK


Everybody knows a football weighs about 250 pounds, so it makes perfect sense that a 90 speed receiver carrying a ball will always get run down by a 60 speed LB who started 10 yards behind him, because the LB doesn't have a super heavy ball to carry.