Let’s deal with this subject before it gets done to death through over-saturation by the end of the week and with the knowledge that it has a good chance to be moot1 by this time next week:
Does undefeated Ohio State have a better claim to the BCS Championship game than the one-loss SEC Champion (for the sake of this discussion let’s use Auburn)?
In my opinion, no they do not.
There’s some faulty logic in use to support the case for the Buckeyes and it’s that logic I most want to address. The pro-OSU argument holds that the Bucks have won all their games and isn’t that what counts most?
Yes, should OSU beat MSU next week they will have won all their games in their league and so be best in their league.
NCAA Football is not the AL Central where you can look at the all-important-loss-column to determine who is winning. Put another way: if the winner of the NY-Penn League went undefeated, you wouldn’t say they’re a better team than the Detroit Tigers because wins are what count most would you? No, you would say, who have they played?
Unlike MLB, NFL, NBA, NHL, etc., CFB has five or six other co-equal leagues playing the same sport and the quality of the sport varies from league to league. As result the champion of one league might be worse than middle of the pack teams in another league. For example, drop Minnesota (4-4 in conference) into C-USA and they go undefeated. Would the OSU fans back an undefeated C-USA Minny’s claim? Doubtful.
It happens that the NCAA acknowledges this league disparity and that is why some leagues’ champions automatically qualify for BCS bowls. But even within AQ leagues, the same principles apply: LSU was 5-3 in the SEC and would probably win any other league; SEC 3rd place finisher Alabama surely would.2
In other words, no: going undefeated is not all that matters. What really matters is are you the best team nationally, now.
And since it cannot be settled on the field, other metrics get used. Evaluations on strength of conference and strength of schedule are valid for trying to connect these dots.
But my personal favorite evaluation criterion is the eye-ball test and that’s where OSU has a problem. My eyes just saw OSU give up SIX HUNDRED YARDS in their most important game in years to a four loss team. Then I watched Auburn hang 34 points on the undefeated top scoring defense in the country.
Thus, Auburn looks better than OSU to me.
It’s not like the soft OSU road performance was a fluke either. Here’s OSU’s road games: Cal (1-11), Northwestern (5-7), Purdue (1-11), Illinois (4-8), and now Michigan (7-5). Every road game not against a 1-11 team was a struggle. And Cal netted over 500 yards.
My point here is not that OSU is a bad team. They are a good team and two years of no losses is a remarkable accomplishment. My point is that record –any team’s record– is only one of several metrics to be used in identifying the two best teams who should play in the Championship. It’s not the only metric.
If we’re being honest with this argument, then NIU should be in the BCS Championship game should they beat BGSU next week and OSU lose to Sparty. That won’t happen because, and only because, NIU’s strength of schedule is weak. If that argument can be applied to NIU, it can be applied to any team.
But OSU is ahead of Auburn in computer polls.
BCS uses six computer rankings, tosses the high and low out, and the average of the remaining four comprises the ‘computer portion’ of the rank. I admit to finding it odd that the computers see OSU on top of Auburn… but that’s where the weighting of an early season loss becomes arbitrary. It’s where a Penn State win at #15 Wisconsin last week causes a bizarre ripple effect through the computer because OSU crushed PSU a month ago.
But don’t have too much faith in the computers because you can make them say anything depending on the algorithm. Take Sagarin. He publishes four ratings and BCS uses one of them. In those rankings:
OSU is 7th [“RATING”], 4th [DIMIN_CURVE], 9th [PREDICTOR], 3rd [PURE_ELO];
Auburn is 11th, 6th, 15th, 6th.
One might say those numbers prove OSU is better than Auburn. One might also say how are OSU 9th and Auburn 15th in the “PREDICTOR” ranking and what to make of Arizona State at #3 in that particular ranking?
Sagarin himself says:
"The PURE_ELO will be used by the BCS. However it is less accurate in its predictions for upcoming games than is the PREDICTOR."
In other words, the computer ranks can be used to support whatever your bias is so I take them with a grain of salt.
Here’s how I rank em.
Without a whole lot of thought and relying on the criterion of who would beat whom on the field based on what I have seen, here’s how I stack em:
FSU, Auburn, Alabama, OSU, Mizzou, OkSt, Sparty, SCar, ASU, UCF.
And if you don’t plodding through a muddy polling apparatus,3 I’d be interested in seeing how you rank them. (This polling software is a bit clunky, so I think you need to rank all the teams until there is a critical mass. Ie., if you down vote FSU, it goes to #20, not #2…) (Hey, it’s something.)
- OSU will have their hands full with MSU; I’ve seen at least one opening line with Mizzou favored over Auburn. [back]
- If we want data to support our league disparity assertions, here is Sagarin’s conference ratings page.
- SEC-West (Auburn, Bama),
- PAC-12S (ASU, UCLA),
- PAC-12N (Stanford, Oregon),
- SEC-East (Mizzou, SC, UGA),
- B1G-Legends (Sparty, Iowa),
- Big 12 (OKST, Baylor),
- ACC-Coastal (Duke,)
- B1G-Leaders (OSU),
- ACC-Atlantic (FSU).
- I promise I looked high and low for a plug-in or script that let me create a poll like this one at the Freep. Couldn’t find one. [back]