Comments on: Thoughts on the Weekend

By: Mark

Mark — Thu, 19 Nov 2009 20:49:52 +0000

The SEC has boasted the most draft picks since 2000, but the margin is not that significant.

Since 2000 the average number of draft picks per team by conference is as follows (ex. SEC teams average 3.3 NFL draft picks per team):

SEC 3.3
Big 10 3.2
Pac 10 3.0
ACC 3.0

By: sandymex

sandymex — Wed, 11 Nov 2009 15:42:43 +0000

It’s funny to hear Pac 10 homers brag about how competitive their conference is when they adamantly oppose playoffs. They further try to rig the system so USC plays home games against the Big 10 rather than real bowls.

HP focuses on just a few “central mean” weighting methods among the many that are used to compare conferences. Sagarin agrees that the SEC is the strongest conference when he equally weighs the performances of teams.

The “central mean” method counts Georgia’s performance 6 times as heavily as it does Florida’s! The funny part is that this method has the SEC as #1 if Georgia had beaten Florida two weeks ago. Think how silly it is to penalize the SEC because it has two undefeated teams. The central mean method is so skewed that the performances of Florida, Alabama and LSU together count as much as the performance of Georgia.

Further evidence of this silliness: USC gets upset by Oregon and the Pac 10 gets a boost. Oregon gets upset by Stanford and the conference gets another boost. UCLA beats Washington and the conference gets a third boost.

But this is the silly world we live in as long as the Pac 10 cowards oppose playoffs.

By: gamedaytribe

gamedaytribe — Wed, 11 Nov 2009 04:08:47 +0000

You can selectively pick data to bolster your arguments, as others have pointed out. You won’t convince people of your views based on that. As I keep saying, it’s insufficient data at this point due to all of the variables we keep talking about.

By: Hannah

Hannah — Tue, 10 Nov 2009 17:35:58 +0000

5 of the SECs top 6 teams have wins against non-conference BCS schools – only one of those teams has a BCS non-conference loss. UGA to Okie State (while Bryant was still on the team, and I think scored two TDs in that game). Top half of SEC is approximately 5-1 in BCS non-conference play, and top half of PAC 10 is approximatetly 2-4. Someone double check those numbers, but the overall picture is generally correct. So, how is the PAC 10 the best conference this year at this point?

By: Hannah

Hannah — Tue, 10 Nov 2009 16:43:41 +0000

4 of the PAC 10s top 5 teams have non-conference losses (to BCS teams). Good for them for scheduling these games, but if you’re the best you should win more of these games!

By: Andy

Andy — Tue, 10 Nov 2009 16:40:24 +0000

to gamedaytribe – that’s something that most fans would agree with more or less and is why so many people want a playoff system with 12 or 16 teams. At least then, you would almost certainly not leave out a team who deserves a shot at #1. But, I think I’ve got an iota of common sense and a solid grounding in logic and mathematics, and all of that tells me that regardless of all the problems with polls and CFB (inlcuding for instance the benefit the PAC 10 receives by being matched against the Big 10 in the “Grand-daddy of them all”) the SEC in the last few years has been the best conference by many statistical measures. So why not acknowledge that while at the same time saying you’d like the system to be improved?

By: gamedaytribe

gamedaytribe — Tue, 10 Nov 2009 16:02:15 +0000

I should have elaborated more.

1. I’m not interested in “you suck, no you suck” type
arguments (principally because I’m not a teenager
any more). I believe people are trying to really
evaluate fairly based on what they know, however
flawed that might be.

2. I’m not actually even trying to make the case that
my team is better than your team or that my conf
is better than your conf either (because I don’t
have the right data for that).

3. My real point is to point out the fallacies of the
system and the arguments people are making to
prop up the rankings and the system.

Nobody has the right set of data to establish
who’s the better team, who’s the more deserving
team, who’s got the better record, who’s more
likely to win, or whatever criteria you want to
choose.

People pointing out BCS final records should
realize those teams got to play in the BCS final
due to a whole lot of flawed polling to begin
with. So it’s a vicious cycle – flawed data being
used to further prop up flawed rankings compounded
by flawed statistical reasoning.

i.e. not only nobody’s right, they don’t have the
data or the system right now to get it right.

The arguing is fun – and generates eyeballs, traffic
(i.e. money) to websites who survive on that type
of thing.

It’s all very frustrating to anybody with an iota
of common sense, much less a solid grounding in
logic and mathematics.

By: gamedaytribe

gamedaytribe — Tue, 10 Nov 2009 15:37:20 +0000

Perry,

Right. Re: polls. Yes. Well. That’s my point.
People really don’t know what they’re talking about.

By: Perry

Perry — Tue, 10 Nov 2009 15:30:16 +0000

@ Gamedaytribe -

I’m not saying Florida = USC, I’m saying Florida > USC. And the only reason I stacked them up against each other is because both USC and Florida are the #1 team in their conference based on the AP poll. Let me make this more clear:

Florida #1 > USC #11 (lost to WASHINGTON)
Alabama #3 > Oregon #14
LSU #9 > Arizona #18 (lost to WASHINGTON)
Auburn ~= Stanford #25 (lost to Wake Forest)
Tennessee > Oregon St.
Mississippi ~= California
South Carolina > Arizona St (lost to WASH…)

By: Hannah

Hannah — Tue, 10 Nov 2009 15:06:31 +0000

Last three years:

PAC 10 12-5 and SEC 19-7 in Bowl Games. SEC has 9 more bowl appearances and still has a better winning percentage.