Our friend Kyle at Dawg Sports has taken issue with my characterization of games played in Atlanta as home games for the Bulldogs, labeling it "sheer nonsense." In the spirit of lively CFB discussion, I'll return the volley by asserting that his rationale is, in turn, sheer nonsense. His claims, and my responses:
Kyle? Kyle NO!
1) Kyle says the facts are a wash, since UGA is 3-3 in Atlanta in recent memory (in 3 Peach Bowls and 3 SEC Championship games).
From a purely statistical perspective, there's no way we can draw anything close to a factual conclusion from a sample set of six games. (Well, there's no such thing as statistical fact, anyway, but one can get pretty close when the p values are low enough.) The variability is just far too extreme, and as I noted above, even a sample set of 30 is generally too small to draw meaningful conclusions. As a result, this becomes a purely qualitative exercise rather than a statistical one.
FROM the qualitative standpoint:
2) Atlanta is home to Georgia Tech, UGA's biggest in-state rival. Kyle argues that "Presuming that the 'Dawgs enjoy home field advantage in the Yellow Jackets' back yard is as preposterous as assuming that U.C.L.A. enjoys home field advantage when playing U.S.C. on the road because, after all, the game is in Los Angeles."
That's not even close to being a fair comparison. In how many of the six games listed above did UGA actually play Georgia Tech? Zero. So is your assertion, then, that Georgia Tech fans shell out $100 (or whatever the actual ticket cost was) simply to go into the Georgia Dome en masse and cheer against UGA? THAT, I think, is preposterous. I wouldn't be surprised if a few did, and maybe more than that just to see some good football, but not even in the same galaxy as being enough to provide any sort of noticeable chorus against the Dawgs.
Further, if UCLA were to play, say, Washington State at the Coliseum, I absolutely, unconditionally WOULD label it a home game for UCLA, as I think any rational football observer would.
What was the distribution of Red vs Purple shirts in the stands for any of the LSU-UGA battles in Atlanta? Or the tilt with West Virginia? I seem to remember utter silence in the Georgia Dome from the games UGA lost, but it could be just me...and I know LSU and WVU fans are not the "golf-clap" sort.
3) "Anyone who presumes, on the basis of a theory unsupported by actual facts, to treat games played in Atlanta as "home" games for Georgia needs to be logically consistent and treat Sugar Bowls played in the Superdome as "home" games for Louisiana State. What's good for the goose is good for the gander."
Actually, I have no problem arguing that the Sugar Bowl in '01 in which we annihilated Illinois was indeed a home game for LSU. I didn't attend the game as I couldn't get a ticket - but a friend of mine from Illinois did and has been permanently scarred by the experience (heh, Geaux Tigers!). I'd even argue that we had a slight home field advantage in the BCS title game vs Oklahoma in 2003 (which I did in fact attend) though OU fans represented themselves VERY well - they'd been buying title game tickets all season long as OU rolled every opponent in the midst of all sorts of "Best team ever?" talk. New Orleans was absolutely flooded with purple that weekend; inside the Dome was a relatively safer refuge for OU fans, but they certainly did not equal the LSU contingent in number.
I ask you, Kyle: if your Dawgs had to play LSU in New Orleans, would you consider it a home game for LSU? I sure would. Not quite to the extent of Tiger Stadium, but pretty damn close.
4) "If anything, I believe playing in Atlanta represents a disadvantage for the 'Dawgs. As noted by And the Valley Shook, it's only a short drive from Athens, so it lacks the urgency of a business trip and offers too many local creature comforts for the team to be focused. I believe the Mountaineers' greater concentration on the game at hand during the first 16 minutes of the Sugar Bowl attests to how distracting it was for the Red and Black to be so close to home."
I believe this is the first I've ever heard of a notion like "Near-Home-Field Disadvantage." If Atlanta is distracting for the near-local team, then good God, what should New Orleans do to LSU's mentality when we're down there? And is it just a testament to the monumental coaching abilities of Pete Carroll that he was able to churn out such a brilliant stretch of football from 2002-2005 with Hollywood in USC's backyard, with Snoop Dogg, Spike Lee, and Will Ferrell roaming his team's sidelines, with Matt Leinart's birthday parties being hosted by Nick Lachey while Alyssa Milano was drooling all over him? I can't help but find it very difficult to believe that.
In any event, I'd be happy to rerun the actual numbers while keeping the Atlanta games classified as road/neutral for the Dawgs, but off the top of my head if I recall correctly, over this particular time span (1998-2005) it would actually hurt their performance measurement.
I think two broad conclusion that can be drawn are that a) it's pretty difficult to do just a cursory examination of home field advantage given the large sample sets required to draw any significant conclusions from them, and b) someone needs to actually do it!