Are the Georgia Bulldogs Due for More Lost Fumbles in 2011?
Our old friend kleph has taken a look a turnover margin in the SEC, and his charts vividly confirm the maddening reality reported a month ago by Senator Blutarsky, who noted that, of "the ten teams which improved their turnover margins the most from 2009 to 2010," the only one "which didn’t win more games in 2010 was the team with the second greatest improvement (+26!) in all of D-1 football, our beloved Georgia Bulldogs."
How depressing a stat is that? It’s so depressing that even SEC-hating Big Ten bloggers feel enough sympathy for us to write: "Georgia went from –16 to +10 and still finished two games worse than they did last year. That would seriously harsh my buzz if I was a Georgia fan." Actually, Georgia fans don’t use "buzz" as a favorable term, for obvious reasons, but I appreciate the show of solidarity, nevertheless.
If you want to talk about harsh, though, click on the first link of the opening paragraph of this posting and scroll down to the chart marked "2010 Average Turnover Margin/Game." What you will see is that the Alabama Crimson Tide led the league, which only makes sense. After all, the Tide won ten games, capped off by a Capital One Bowl shellacking of the Michigan St. Spartans, finished in the AP top ten, and went 5-3 against ranked opponents. It figures that ‘Bama would’ve been good at a statistic generally accepted as consequential.
Slightly behind the Tide in second place, however, were the ‘Dawgs. Georgia was somewhat better than the Cotton Bowl champion LSU Tigers and the Gator Bowl champion Mississippi St. Bulldogs, and the Red and Black were significantly better than the Sugar Bowl champion (in due time) Arkansas Razorbacks, the national champion Auburn Tigers, and the SEC East champion South Carolina Gamecocks. All five of the Athenians’ conference losses were to teams with worse average turnover margins than the Bulldogs’.
The rest of kleph’s charts paint a more thorough picture, though. Georgia recovered a middling 45.45 per cent of opponents’ fumbles in 2010, which was only good enough for a sixth-place finish in the SEC. The five teams ranked ahead of the Classic City Canines in that category combined to go 4-0 against the ‘Dawgs last fall.
Such numbers matter because, as kleph notes, "once the ball hits the ground the law of averages takes over. While this means it is impossible to guess what a team will do any given year, it does suggest that unusual outliers will return to the mean. . . . [I]f you suffered terrible fumble luck or had good fumble fortune one season, don't plan on it happening again the next." Football Outsiders confirms the relative randomness of fumble recoveries.
That thought is troubling for this reason: Georgia lost only 40 per cent of its own fumbles in the 2010 season. Only two teams in the SEC were better at covering up their own dropped ovals last year, and only one of those was more than one percentage point better. The Bulldogs, in other words, are unlikely to lose fewer (or even as few) fumbles in 2011 than they did the year before.
The Red and Black coughed up the pigskin 20 times last season, exactly as many times as the eventual BCS champion did, but the ‘Dawgs failed to retain possession a league-low eight times when putting the ball on the carpet. Of course, three of those eight lost fumbles occurred on the South Carolina goal line when the Athenians were trailing the Palmetto State Poultry by eight points, on the Mississippi State goal line when the Bulldogs were trailing the Magnolia State Mongrels by seven points, and on the Colorado 30 yard line when Georgia was trailing the Centennial State Ungulates by two points.
Statistically, the ‘Dawgs are due to lose more fumbles this year than they did last year, yet nearly half of the ones they lost last year were excruciating game-changers that certainly shifted the momentum and arguably caused losses not just of possession, but of the contests themselves. If you take away those three giveaways, Georgia probably doesn’t win all three of those games, but the Bulldogs darned sure don’t lose all three of those games, either.
The prospect of a greater number of negative outcomes in a category that gave rise to much wailing and gnashing of teeth a year ago gives us all cause for pause, but we may take comfort (cold though it may be) in this fact: not all turnovers are equally devastating, and, while the Red and Black may be due for more of them overall, surely, if there is any justice to be found in the midst of randomness, they will not occur as often in such critical situations. We are willing to accept a statistical regression to the mean, as long as we are able to inch nearer to the norm in angst, as well. There, too, we are more than due.
Go ‘Dawgs!
12 comments
|
0 recs |
Do you like this story?
Comments
I meant to make this point on TSK, but...
While kleph presents some excellent and interesting statistical information, the nature of the statistic leads one to assume that, if you’re Alabama (for example), things can only get worse.
Thus I would add a second fact in which we may take comfort: while we may not be able to control who recovers a fumble, we can absolutely control who fumbles. While we will miss Washaun Ealey’s talents, we will certainly not miss his propensity for fumbles. If we don’t fumble at all in those 3 games, we need not rely on chance.
And hopefully Big John and Kwame “the animal” Geathers will cover enough square footage between the two of them to make the odds a bit better than even….
actually, no
i argue that for alabama, things can only get better. for all the variables a team can control in terms of the turnover margin, the crimson tide have show stability under saban. throw fewer picks, drop fewer balls and grab lots and lots of interceptions.
the one statistic that skill and coaching have no effect is turnover recovery, and alabama was woefully unlucky with that last year. it’s almost certain this will return more to normal next season.
but, back to the georgia numbers, i have to say these perplexed me immensely. for pretty much every other team you could make at least a general connection between their turnover margin and win loss record but georgia’s made no sense at all. it bugged me enough i went back and pulled together this chart showing the point differential for each game side by side with turnover margin:

i’m sorry but there is something is going on here. last season, georgia won every game it had a positive turnover margin with the sole exception of auburn. given the close scores in florida, arkansas and colorado games, turnovers can be credited as a direct cause for the outcome. but with south carolina, ole miss and UCF… i got nothin.
Remember the Rose Bowl: The Story of the Alabama Crimson Tide & the Grandaddy of Them All
Yeah, it's depressingly and distressingly weird.
If you looked at Georgia’s statistics from 2009 alongside the Bulldogs’ numbers from 2010, you’d assume immediately that the Red and Black had a better season in 2010, but it just didn’t translate onto the field.
One problem with the turnover issue is that, in 2009, while Georgia was horrible in turnover margin, there was really only one game (against Kentucky) that you could point to and say for certain, “These specific turnovers cost us this game.” In 2010, there were four games (against South Carolina, Mississippi State, Colorado, and Florida) where you could argue with a straight face, “If you remove this single specific turnover, you’d have gotten a different outcome.” Georgia still might have lost a couple of those games, anyway, but, if it had just averaged out, 6-7 would’ve been 8-5, and 2010 would have produced the same record as 2009 but would have featured improved play, and we’d be heading into this season feeling we were on the upswing.
Go 'Dawgs!
here's a graphic that should give you cause of optimisim
this is alabama’s interception stats for the past five years from my article on this topic earlier this week:

the reason for alabama’s insanely good turnover margin is in this graphic. the first two years of the nick saban experience were ok in terms of interceptions but then the 3-4 kicked in and it went co-co coca puffs. the whole point of saban’s defense is to put the ball into the waiting arms of the secondary.
i expect georgia will see a similar step forward in defensive play this season under grantham’s system.
and that pink line shows exactly how valuable greg mcelroy was to alabama’s offense. he simply didn’t throw picks. murray may not be that careful with the ball but he’s not far off the mark.
Remember the Rose Bowl: The Story of the Alabama Crimson Tide & the Grandaddy of Them All
Serves me right for posting at 7AM...
Definitely meant to say better for Alabama. Worse for Miss. State (hopefully on October 1st…).
I agree with you completely. I was mostly attempting to comfort depressed Kyle in that we can take comfort not only in hoping that the timing of the fumbles will not have as big of an impact, but also we can hope not fumbling at all will be a focus of the offseason, particularly with a new featured back coming in.
My last sentence was mostly facetious.
no worries
the key issue, though, is this is one element of the overall turnover margin. and you got that spot on.
Remember the Rose Bowl: The Story of the Alabama Crimson Tide & the Grandaddy of Them All
That's a good point, NOLADawg.
To be fair, kleph and Football Outsiders both made that point, and I did not quote them adequately for that proposition, but you’re right. As Football Outsiders notes, stripping the ball and holding onto the ball are skills; they are not random. Likewise, a quarterback’s propensity for putting the ball in the proper location and a defensive back’s propensity for reading the ball and jumping the route are not random, so interceptions typically are not flukes of the hey-look-what-I-found variety. The only random element is what happens when the ball hits the turf, when it is pretty much a roll of the dice (although attempts at fumble recoveries may be executed well or poorly). Washaun Ealey’s departure may well mean fewer fumbled balls by the Bulldogs, but it is statistically likely that those fewer Georgia fumbles will result in a higher percentage of loose footballs recovered by the opposition.
Go 'Dawgs!
one qualifier to your observations, kyle
Statistically, the ‘Dawgs are due to lose more fumbles this year than they did last year…
i don’t know if you can say this with authority. my only contention in these articles has been that extreme outliers are unlikely to be replicated. to posit more than that would be to delve further into the methodology of the premise that fumble recoveries are random than i’m truly comfortable with. and i most emphatically am not arguing for any type of “pendulum effect.”
that said, this topic sparked a lively discussion over at tomahawk nation on that very point. eventually, one commenter made this observation about recovery rates in light of their apparent randomness (an observation that was later approved by FO almumnus and SB Nation CFB stats maven bill barnwell):
Every team has the likelihood of the following recovery rates:
0% – 0.1%
10% – 1%
20% – 4.4%
30% – 11.7%
40% -20.5%
50% – 24.6%
60% – 20.5%
70% – 11.7%
80% – 4.4%
90% – 1%
100% – 0.1%
by that reasoning, georgia was well within the probability “sweet spot” last season even if it was on the negative side of the axis. thus it’s difficult to make any argument that there should be anything dramatically different in 2011 concerning this category than we saw in 2010.
Remember the Rose Bowl: The Story of the Alabama Crimson Tide & the Grandaddy of Them All
Actually, that stat does seriously harsh our buzz...
… since we still managed to defeat Tech last year.

by 






























