FOF Mythbusters: Are there more/too many fumbles in 6.3/6.3a?

[Ben E Lou]

Several people have asked me about this in various places, so I thought I'd post some numbers in one central location. The metric I'm using is fumbles per 100 touches. Here's a cut and paste from the code pulling the data, so there can be no question that there's a "bug" in the SQL. ;)

100*sum(fumbles)/(sum(rushattempts)+sum(timessacked)+sum(catches)+sum(kickreturns)+sum(puntreturns)) AS fmbperhundredtouches

So I'm capturing every rushing attempts, sacks, catches, kick returns, and punt returns. I'm pretty sure that leaves nothing out. I'll post the recent seasons of several leagues.

[Ben E Lou]

WOOF has played three weeks on 6.3, one week on 6.3a.

Fumbles per touch in most recent seasons:

2016: 2.47
2017: 2.51
2018: 2.50
2019: 2.35


2020 (6.3/6.3a): 2.41

[Ben E Lou]

My understanding is that PFL played a couple of weeks this season on 6.2, then most of the year on 6.3, and final handful of games on 6.3a. (Someone from there feel free to chime in and correct me if that's not entirely accurate.)

2027: 2.47
2028: 3.49
2029: 2.52
2030 (current): 2.07

[Ben E Lou]

FOWL is a bit of a different league. It's worth noting that there's a significant nerfing of QB talent there by regenerating draft classes over and over until no QBs rated over 6.5 adjusted are in the class. That's important to note because of the huge sense rush correlation to good QBs. Looking at the data, the year-to-year variation seems to be most wide there, and I suspect it's because more teams are forced to go with low sense rush guys for a year or so, then abandon them. FOWL has done one full season with 6.3, and another full season with 6.3a. Anyway...

2037: 2.05
2038: 3.07
2039: 2.25
2040: 1.89
2041: 2.16 (version 6.3)
2042: 2.80 (version 6.3a)

[Ben E Lou]

USFL has run only one week, on 6.3a, so beware of noise/sample size.

2015: 2.06
2016: 1.80
2017: 2.50
2018: 2.42 (one week of 6.3a)

[Ben E Lou]

So, it looks like only one league (FOWL) has seen any significant rise with 6.3/6.3a, and it's worth noting that:

1. The fumble rate is well within the range of normal for the league.
2. The talent in that league is very different from what FOF expects, so ymmv.

The other three leagues are showing, if anything, a slight decrease in the fumble rate. But as with the increase noted in FOWL, it's worth noting that the decrease is also well within the range of normal for the league.


CONCLUSION: Based on the data so far, there's no reason to believe that the fumble rate has changed significantly, if at all.

[Yoda]

What about the fumble rate inside each catagory?
ie- fumbles per punt return, fumbles per carry, fumbles per catch, ect-

[Ben E Lou]

The data isn't available to tease that out. In the main stats, FOF only tracks fumbles. I know of only two people outside of Ann Arbor who could get that info.