Lock of the Year

Jane Lynch wins the Emmy for Best Supporting Actress in a Comedy for Glee.

No one else has a chance.  She's so good it's ridiculous.

"Finding a hairstyle that doesn't make you look like a lesbian."

Read More...

Summary only...

Early Bowl Projections = Not Smart.

So CNNsi has put out its Bowl Projections, based upon its own rankings of teams. It seems like they've attempted to use the BCS rules in filling out the slots, but I think they did it kind of wrong.

Here's their setup:

BCSNCG: #1 USC vs. #2 LSU
Rose: Michigan (Big 10 Champ) vs. Rutgers (At-Large)
Fiesta: Texas (Big XII Champ) vs. Wisconsin (At-Large)
Sugar: Florida (At-Large) vs. Oklahoma (At-Large)
Orange: Va. Tech (ACC Champ) vs. West Va. (Big East Champ).

Their only rule, it seems, in selecting At-Large teams is that they had to be in the Top 14 of their rankings. It also seems like they've abided by the BCS rule that no more than 2 teams from any particular conference may be selected to play in BCS games.

Here's where they go wrong: the order of selection.

I'll assume for the sake of argument that CNNsi's rankings are fine.

The way the BCS works is as follows:

STEP ONE

#1 and #2 in the BCS rankings are automatically slotted in the BCSNCG.

So far, CNNsi is fine, placing USC and LSU in the BCSNCG.

STEP TWO

Conference Champions not selected to play in the BCSNCG are assigned to the Bowl that conference is traditionally aligned with. Big 10 and Pac 10 to Rose. SEC to Sugar. Big XII to Fiesta. ACC to Orange.

CNNsi gets this almost exactly right, but it's hard to say perfectly. They slot Michigan in the Rose, Texas in the Fiesta, and Virginia Tech in the Orange. The problem is that they also appear to slot West Virginia in the Orange as Big East Champ. The Big East Champ is not automatically slotted in the Orange Bowl. Now, it may be that the Orange Bowl can select West Virginia to play, and they are eligible for the BCS because they are conference champs, but the fact is that West Virginia as Big East Champs is treated just like any At-Large selection (only the Mountaineers are required to get picked by some bowl).

STEP THREE

Here's where it gets complicated, and it seems to me that CNNsi screwed up. Let's break this down a little.

(A) ORDER OF NON-CONFERENCE-TIE-IN SELECTIONS.

The order of selecting teams rotates from year to year, and also depends on the selections of conference champions to play in the BCSNCG. The order in 2008 is:

1. Bowl that would have received #1 team because of conference tie-in.
2. Bowl that would have received #2 team because of conference tie-in.
3. Orange Bowl
4. Fiesta Bowl
5. Sugar Bowl

So, assuming USC and LSU are #1 and #2 respectively, the order of at-large, non-conference-tie-in choices would be:

1. Rose
2. Sugar
3. Orange
4. Fiesta
5. Sugar

(B) TEAMS AVAILABLE TO BE SELECTED.

CNNsi states that the team must be in the top 14 to be selected by a BCS bowl. This somewhat aligns with BCS rules, but it doesn't go all the way. There are teams that MUST be selected, and teams that CAN be selected, and there are further limitations on that.

Assuming the conference champions CNNsi has chosen, and assuming that their Top 14 teams mirror the BCS rankings, here's how things would end up:

MUST GET PICKED
West Virginia - by virtue of winning the Big East
Florida - by virtue of finishing #3 in the BCS but not winning their conference title.

CAN GET PICKED
Louisville
Oklahoma
Wisconsin
Ohio State
Rutgers
Arkansas
Penn State

CANNOT GET PICKED
Everyone else

Now, there's a secondary BCS rule that comes into play right here. No more than 2 teams from any particular conference can be selected to play in BCS games in a single year. This cuts down the possible selections significantly.

Now, the options for at-large picks are:

Florida (mandatory)
West Virginia (mandatory)
Oklahoma
Louisville OR Rutgers (not both)
Wisconsin OR Ohio State OR Penn State (no more than one)

Arkansas is now ineligible.

So now we know the 5 options for the non-conference-tie-in bowls. Let's see if we can project how the selection would happen.

PICK ONE - ROSE BOWL
The Rose Bowl would have the first choice of all these teams, to replace #1 USC and face Big Ten Champ Michigan. The only limitation as to which team it can pick is this: it cannot pick Florida unless the Sugar Bowl consents to letting them (the BCS has a rule saying the first choice bowl can't choose another team from the conference of the #2 team unless the bowl that would have received that #2 team consents). Let's assume that the Sugar Bowl won't consent to letting the #3 team slip out of their hands (and a safe assumption because the Sugar Bowl ends up with the last pick, so they'll want at least one big draw). Let's also assume that the Rose Bowl won't pick another Big 10 team to play Michigan.

So they can choose West Virginia, Oklahoma, Rutgers or Louisville.

By CNNsi's rankings, WVU is 4th, Louisville is 5th, Oklahoma is 9th and Rutgers is 12th (Michigan is 6th). For some reason, CNNsi picks Rutgers to play in this game.

Objectively, I think you can make an argumentthat Oklahoma would be the biggest draw (largest fan base). The best team would be WVU (assuming CNNsi's rankings). For CNNsi to choose Rutgers to play in this game (as the third best team in its conference and selected ahead of the two teams that finished higher), seems to be a great stretch. Were I selecting, under the assumptions CNNsi has made, I think Oklahoma would probably end up the choice, but just barely over West Virginia. Rutgers would not be in the equation.

Better Prediction: Michigan vs. Oklahoma

PICK TWO: SUGAR BOWL (First Choice)
No doubt that the Sugar Bowl would choose #3 Florida. CNNsi is correct on this choice.

PICK THREE: ORANGE BOWL
The Orange Bowl has no limitation on choices as to who should play against Virginia Tech. Further, since no other ACC teams are listed, they don't have conference matchup issues like the Rose Bowl. The following teams can be picked:
West Virginia
Louisville OR Rutgers
Wisconsin OR Ohio State OR Penn State

WVU is still ranked #4, Louisville is #5, Wisconsin is #10, Ohio State is #11, Rutgers is #12, and Penn State is #14.

Probably, WVU is the choice here. They're ranked ahead of Louisville, and the Cardinals had played there just last year. There may be a temptation to grab one of the Big 10 teams and their legion of fans, but the gap in quality of teams is probably too great to pass up one of the Big East teams.

Therefore, CNNsi is probably correct that WVU is the pick, but it should be clear that this is a "choice" and not a mandatory conference-tie-in.

PICK FOUR: FIESTA BOWL
Texas is tied in as Big XII champ. The Fiesta can then choose any of the following teams to play the Longhorns:
Louisville (#5) OR Rutgers (#12)
Wisconsin (#10) OR Ohio State (#11) OR Penn State (#14)

Sexiness vs. fanbase. Any of the Big Ten teams would travel better, but Louisville is objectively the best team left. It'd probably be left to how big the gap is between Louisville and the 3 Big Ten teams (i.e., if Louisville is 11-1, and Wisconsin's the best of them at 9-3, Louisville is probably the choice, but if Louisville is 11-1 and Wisconsin is 10-2, it's probably a closer call). Of all of the BCS bowls, the Fiesta is probably the least concerned with ticket sales, believe it or not. University of Phoenix Stadium is the smallest BCS locale by a good margin (only 63,000 unless expanded as it is for BCS title games). And with Texas already rolling in, it's probably unlikely that they'd have trouble selling out the place even with Louisville.

One could argue either way, so we will.
Option A: #8 Texas vs. #5 Louisville
Option B: #8 Texas vs. #10 Wisconsin

PICK FIVE: SUGAR BOWL
The final pick goes to the Sugar Bowl to pair someone against Florida. The choice for the Sugar depends largely on the choice of the Fiesta.

If the Fiesta chooses Louisville, the Sugar can choose between Wisconsin, Ohio State and Penn State.
If the Fiesta chooses Wisconsin, the Sugar can choose between Louisville and Rutgers.

It's probably a safe assumption that the Sugar will take whichever team the Fiesta didn't.

Option A: #3 Florida vs. #5 Louisville
Option B: #3 Florida vs. #10 Wisconsin.

So there you have it, a full analysis of the BCS selection. Basically, CNNsi screws up when it chooses a #12, third-in-their-conference Rutgers team to be the first team taken off the board by the Rose Bowl. Under the scenario CNNsi laid out, it's highly unlikely that the BCS bowls would consider Rutgers at all. The only advantage Rutgers has is the New York TV market. Rutgers does not have a sizable traveling fanbase. Rutgers is not geographically proximate to any of the bowls. And Rutgers wouldn't have the "they deserve it" argument behind them, having finished behind several other teams. There's just no justification for placing them in that bowl, aside from favoritism towards the local team closest to CNNsi's location. And that's poor punditry.

Read More...

Summary only...

OK, just one small taste of douchebaggery

So I was hunting through Michael's archives at Braves and Birds trying to find a comment thread where I was praising Valencia's lineup (lucky, not prescient... anyway, it's here). On my way spelunking around there, I spotted this post on a gem of an article by our favorite SI college football writer, Stewart Mandel.

The crux: 16 teams had a real chance to win the national title, in Mandel's opinion, based upon a number of reasons. One major reason appears to be mediocrity in the previous season:

While the past two seasons have played out largely as predicted, there have been three seasons since 2000 in which a team that finished with five losses the year before rose up to win a national title the following season: 2000, Oklahoma (7-5 the year before); 2002, Ohio State (7-5); and 2003, LSU (8-5). Auburn's undefeated 2004 team also went 8-5 the preceding season. All four teams started the year outside the Top 10. This year the Buckeyes, Texas, Notre Dame and USC sit atop the preseason coaches' poll, but the No. 1 team come January could just as easily be a lower-rated squad such as No. 11 Miami (9-3 in '05), No. 13 Louisville (9-3), No. 15 Michigan (7-5) or No. 17 Iowa (7-5).

That "five losses" bit should stick out like a sore thumb. Here's why. Of the 16 teams Stewart Mandel listed, half of them - EIGHT (!!!) - ended up with at least five losses in 2006. I'm no good at math, but I'd bet someone can put together a study as to what's more likely: bouncing from 5 losses to a national title, or going from 5 losses to mediocrity. Beer is on me if it's the former.

Some other fun facts:

• Mandel listed 4 ACC teams with a chance at winning the national title. None won the ACC. OK, nobody on earth picked Wake to win the ACC. But the 4 teams Mandel did choose... each and every one of them lost at least 5 games by the end of the season.
  
• Mandel only listed 2 SEC teams. Not listed is the eventual national champion. Listed is South Carolina, which finished 5th in the SEC East. Neither team listed made the SEC title game and neither played in a BCS bowl.
• Mandel picked a random non-BCS-automatic-qualifying conference team, Utah. The Utes lost 5 games and finished 3rd in that non-BCS-automatic-qualifying conference.
• 6 of the 16 teams had lost a game (putting them in dire position for qualifying for the BCS title game) by Saturday night, September 9.
  

What was the purpose of this list? I don't know. I don't understand preseason predictions at all. I think they're useless in terms of accuracy. What I do think this column was useful in doing: driving readers angry enough to increase readership and get fulminating emails for mailbags. But when it's August and we're all starving for some football, there'll be some big previews with words arranged in sentences posing as analysis. Ignore it. Seriously. It's pointless.

Read More...

Summary only...