The definitive guide to the battle for 2nd place in the BCS Legend: SOS = Strength of Schedule QW = Quality Win in the BCS (top 10 in the BCS polls) ........10th = -.10, 9th = -.20, ....1st = -1.00 CG = Championship Game NYT = New York Times Current BCS USC 6.89 LSU 9.04 After Arkansas Win (USC does not play this week) (could change w/Fla, GA, Bama, La. Tech, & AZ games affecting SOS for LSU) USC 6.95 LSU 7.78 Current Projected SOS: USC 1.44 LSU 1.04 This is projected with the following remaining games: Arizona losing to Arizona State.........Nov. 28 @ 2:00 (FOX SW) Toledo losing to Bowling Green.........Nov. 29 @ 11:00 (ESPN) Georgia Tech losing to Georgia.........Nov. 29 @ 12:00 (ABC?) Rice losing to La. Tech...................Nov. 29 @ 2:00 Florida losing to Florida State...........Nov. 29 @ 2:30 (CBS) Hawaii losing to Alabama.................Nov. 29 @ 6:45 (ESPN) Hawaii losing to Boise State.............Dec. 6 @ 10:30am Syracuse losing to Notre Dame.........Dec. 6 @ 12:00 (NBC?) Biggest games affecting SOS (in order of importance): Hawaii beats Alabama.................USC 1.32 LSU 1.28 Florida beats Florida State...........USC 1.48 LSU 0.84 Arizona beats Arizona State.........USC 1.48 LSU 0.88 Rice beats La. Tech...................USC 1.44 LSU 1.24 Toledo beats Bowling Green.........USC 1.36 LSU 1.04 Who will LSU play in the SEC CG if LSU, GA, Tenn & Fla ALL win this week? The SEC tie-breaker now comes down to the new rule that was adopted just a few weeks ago, in which the worst team in the BCS polls is dropped out. If the remaining two teams are within 5 spots of each other in the BCS, then head-to-head outcome is used. I project a Florida win over Florida State will propel Florida into 8th place in the BCS, edging out Tennesee who would finish in 9th. What would also help Florida and hurt Tennessee would be a Miami loss to Pittsburgh. I project Georgia to finish ahead of Texas in 6th spot with a slim chance at finishing 5th ahead of Ohio State as well. So if Georgia finishes 6th and Florida is the 2nd best SEC team at 8th, then head- to-head is used. Florida beat Georgia 16-13 earlier this year. LSU would face Florida. What is the importance of this? Many things. The most important is the fact that Georgia would stay 6th place in the BCS, yielding a -.50 quality win (QW) factor to our BCS score. In other words, our overrall BCS score will be diminished by .50. If we play Georgia, they would be projected to finish 10th after an LSU loss in the SEC CG. This would reduce our credit to just .10. Secondly, it's a revenge game for the loss earlier this year. Facing Georgia has the opposite effect. It's a revenge game for Georgia against LSU. Lastly, it's just difficult to beat the same team twice in one year. Here are my projected final BCS standings if LSU plays Georgia in SEC CG: ..............Computer....Media.....Losses........QW........SOS........Total LSU............2.17.........3 00..........1..........-.10........1.04........7.11 USC............2.67.........2.00..........1............00........1.44........7.11 Now, for the "what ifs": LSU....(with Miami losing to Pittsburgh).......-.20......................7.01 LSU plays Florida in the SEC CG: (LSU beating what would be NYT #2 Ranked Team Florida would vault LSU to 2nd in NYT yielding a change in computer poll results) (The .20 gain in SOS by Florida beating Florida State is lost right back due to LSU playing 9-3 Florida instead of 10-2 Georgia) LSU..............2.00.......3.00..........1............-.70........1.04........6.34 USC.............2.83........2.00..........1..............00........1.44........7.27 The 2.17 and 2.67 computer scores are the most conservative numbers I came up with as fas as LSU is concerned. In other words, this is the worst outcome I can predict for LSU. It is possible that those numbers could be as good as 2.00 for LSU and 3.00 for USC. I use the numbers above for safety sake. It is obvious for multiple reasons that a Miami loss to Pittsburgh is big for LSU. Georgia would finish 9th instead of 10th in the BCS if LSU beats Georgia in the SEC CG. The extra -.10 QW could be the deciding factor. Also, as indicated earlier, it helps Florida's chance of finishing ahead of Tennessee, which is ultimately what we need in order to create a little breathing room between LSU and USC and allow for some fluctuation with SOS. Under this scenario, Georgia would finish 5th and Florida would actually still finish 9th in the BCS even after losing to LSU (remember Miami FL and Florida State would be knocked out the top 10). This yields a total of -.70 QW rating adding up the Georgia and Florida victories. This would give LSU a 0.93 edge (6.34 vs. 7.27) using the projected SOS. There is no way for USC to make up this margin in SOS even if all the other 7 games listed above go USC's way. In case you are wondering what happens in case of tie for 2nd in the BCS, you'll be happy to know its head-to-head competition, then win against highest ranked BCS team. Since LSU and USC didn't play each other, LSU wins tie-breaker with win over Georgia (who can finish no worse than #10), and USC beat #15 Washington State.
A few comments: I don't see how Florida passes Tennessee unless Miami also loses. I think there is a chance that a lackluster Tenn. win might cause them to fall back behind Ohio State in the AP, thus causing them .5 BCS points. Agree with Ga. jumping Texas, but the only way I see Georgia jumping both Texas and Ohio State is if (a) Bowling Green loses (not likely) and (b) Ohio State falls to 8 in the ESPN/USA Today poll behind Tennessee, because 10 of OSU's 11 opponents are done for the season. If Florida wins AND they edge Tennessee to get to the SEC Title Game, you can stop the analysis right there because an LSU win in Atlanta over the Gators seals the deal (LSU retains Georgia's QW points and gets a strength of schedule boost with a second game against Florida). Also, your projected LSU strength of schedule after an LSU-Ga rematch of 1.04 is actually pretty conservative, for it assumes things like losses by Florida and Alabama this weekend. If both the Gators and Tide lose this weekend, I'll eat my shoe! Not going to happen. Agree that the 2.17 final computer ranking is a "worst case" for a 12-1 LSU. We, therefore, both agree that, barring an LSU loss, Tiger fans can reasonably plan to ring in the New Year on the corner of Bourbon and St. Peter's streets (I tend to eschew those touristy-type places myself, but you gotta go to Cat's on New Years if you're in town!)
A correction: 1.04 SOS already takes into account an Alabama win so an Alabama loss would be very bad (see my projections above) Comments: I project Tennesee to pass Florida mainly on the strength of computer polls. I have Florida passing Tennessee in every computer poll except 2. One of those 2 can be thrown out for Florida. Just as an omen to this coming week's demise for Tennessee, you need look no further than the Jeff Sagarin computer poll which projects the final results. He has Tennessee going from 11th to 18th in his poll even after beating Kentucky. It's gonna happen. Florida will pass Tennessee. I do agree that Miami losing should help since it clears the way for Florida to finish no worse than one spot behind Tennessee in 1 media poll and 2 spots in the other (1.50 difference). Also, Tennessee SOS goes from 1.32 to 1.68 this week (projected). Florida's goes from .36 to .16. Even if you call it a draw in the computer polls (for simplicity sake), then you have the following: Tennessee 7.5 (media) + 1.68 (SOS) - 0.30 (QW) + 2 (losses) = 10.88 Florida 9.0 (media) + 0.16 (SOS) - 1.30 (QW) + 3 (losses) = 10.86 It's close, no doubt, but the that's with the conservative notion of Tennessee and Florida having identical computer poll results. I actually project Florida with a 6.0 and Tennesee with a 7.5 in the computer polls. Of course if Florida goes out an dominates Florida State and Tennessee barely beats Kentucky (which could happen since Kentucky has played every SEC game pretty tough), then that may propel Florida ahead of both Tennessee and Ohio State into 7th place leaving no doubt Florida finishing ahead of Tennessee in the BCS.
I love your projections, however, where did you get your SOS projections and projected computer ranks. To my thinking, it seems that the computer rankings are a 1:1 ratio based on wins and strength of schedule. Where do your projections have the SOS ranking as far as 1-50 for example.
I am not an expert on SOS projections. I use Rich Tellshow's website for this information (search for BCS under Yahoo and you will see a reference to Rich Tellshow's BCS website). He projects future SOS by figuring the higher of the two teams in the BCS as the winner of each game. It's a simple approach, but as good as any. The computer ranks I figured myself. And believe me, I have looked at this 400 different ways before coming out with such bold predictions. In fact, these are far from bold. I actually believe LSU will end up 2nd in all 7 computer polls. The only problem is I couldn't commit to the Peter Wolfe poll, since there is no rhyme or reason to his rankings (for instance, Texas went down 6 spots in his poll after winning this week!). All I know is that LSU will be either 2nd or 3rd in that poll. I guarantee you LSU ends up 2nd in every computer poll (5 polls) except NYT and Wolfe. NYT is still up for grabs. Florida would end up 2nd in the NYT poll after beating Florida State. If LSU plays and beats Florida, LSU would end up 2nd in NYT. If LSU plays and beats Georgia (who is tied for 7th in NYT), LSU could end up anywhere from 2nd to 5th, it would be that close. Even if Florida wins and LSU ends up playing Georgia in the SEC CG, then at least Florida goes ahead of USC in the NYT poll, causing USC problems. Computer polls are not as simple as you think. SOS is vital, but it's there is much more to it than that.
For those wondering where Rich Tellshow's website is, it is at: http://www.geocities.com/rtell/ I'm sorry, but I'm missing how Bowling Green and Toledo have any bearing on LSU. According to the schedules I have for those two teams, they don't play LSU or USC or any of their opponents. I must be missing something. However, not on the list was North Texas and New Mexico State, and that did have a small positive impact for LSU. We gained one on our Opp-Opp standing.
once my head stopped spinning, I realized that what we need to do is WIN out and let the rest take care of itself.
Well, that's not a bad philosophy to take, but this thread is for those of us who enjoy looking at the possibilities in detail.
And this is primarily based on an LSU perspective of how things should turn out. The Insiders/Dawgpost reported a story today (11-26-03) regarding Mark Richt who stated that Georgia is technically still in the hunt for a Sugar Bowl bid.
