"I'm looking forward to the inevitable upcoming post on the BCS, in light of the injustice Texas endured this weekend."
I'm not sure if Adam predicted the future here, or I'm simply giving the readers what they want. (Pay attention to this reader(s), ask and ye might just receive.)
For those of you who are less informed, but want to know what Adam is talking about here, Oklahoma (OU) gets to go to the Big XII championship game rather than Texas (UT), despite Texas beating OU on a neutral field. This is because the tie was broken by OU being very slightly ahead of UT in this weeks BCS rankings. (What are the odds that anyone reading this didn't know this, yet actually cares about the subject? Slim.)
Now, I've never really heard Adam be much of a UT fan, but he was raised (if not born) in Texas, so I assume that he'll be disappointed with me when I say that I think the end result (OU over UT) was right, even if the course to get there was a bit screwy.
First, it is important to remember that it's not like the BCS tie breaker is the first and only tie breaking method used by the Big 12 conference. In the event of a 2 way tie, the head to head game is the tie breaker, which makes sense. But, this was not a 2 way tie. It's been quickly forgotten that Texas Tech (TT) also has a single loss this season. TT beat UT, UT beat OK and OK beat TT. Other than that, each team is undefeated. So, unscrambling things is a bit trickier. In the event of a 3 (or more) way tie, the Big 12 tiebreaker procedure calls for a comparison of:
1. Teams head to head record (each 1-1)
2. Teams record w/in their division (undefeated).
3. Records against next highest teams in division starting w/ #4. (undefeated)
4. Reocrds against all common opponents. (undefeated)
5. BCS ranking. (OU wins)
6. Winning percentage (tie).
7. Winner chosen by draw.
So, if we had chosen to skip the BCS mess, the winner would have been selected by picking names out of a hat, or ping-pong balls, or a three sided coin. I don't think that would be better. Its a tricky thing, deciding three way ties. I'd be curious for suggestions of what specific measurable metric should be used instead of the BCS. (Poll 100 people?, time of possession?, turnover margin?, graduation rate?, winner-take-all game of Battleship?) Remember, that people are always loathe to use margin of victory, as it encourages running up the score.
But, since I already said I agree with the selection of OU (if not the method, though I don't particularly have a better one) I will now support that claim. First off, this blog does a decent job of comparing UT, OU and TTs games against each other. Basically, OU came out way ahead on net points and net yards between the two games. Analyzing the games would show that perhaps the score was "run up" a bit in the OU/TT game, you can't fault a team for wanting to build up a large cushion in the first half against one of the most explosive offenses in the nation.
OU likewise wins a schedule comparison, in my opinion. This is where TT gets dropped from the conversation. They beat Eastern Washington, UMass, Nevada and SMU. That's 2 I-AA opponents, an ok WAC team and a bad C-USA team. About 50 or 60 teams in the nation could expect to go undefeated in those games. UT beat FAU, UTEP, Arkansas and Rice. While none of those teams are out right horrible, none of them are all that good either. OU, played Chattanooga, Cincinnati, Washington and TCU. Two of those teams (Chattanooga, and UW) are just terrible, but Cincinnati (10-2, BCS #13) and TCU (10-2, BCS #11) are pretty good. A schedule of 2 good teams and 2 terrible teams is tougher than 4 so-so teams. A #25 team would expect to beat all 4 of UT's non-conference opponents, whereas they would likely be glad to finish 3-1 against OU's opponents.
(The schedules also show that TT had to go to OT to beat Nebraska, UT had a single close game vs OSU, and OU has no win of less than 14 points.)
So, there you go. Really, the big looser in all of this is Oklahoma State, which is probably a top 20 team, maybe a top 15 team, yet couldn't do better than 4th place in their division of their conference. Ironically, their chances of going to a BCS game would have been better if they had moved to the MWC. Going undefeated against TCU, BYU and Utah is easier than against OU, UT and TT.
Addendum: I meant to say this initially, but forgot, what with Firefox crashing a few times during my post. How can I support OU when UT beat them, head to head? (Though, TT beat UT, and OU beat TT, but OU lost to UT, even though UT lost to TT . . . . ) It's all about sampling size. One game is a really crummy way to figure out if one team is "better" than another. 12 games is still pretty crummy, but that is something we can actually work with. This is the underlying difficulty with college football. 119 teams, and they only play 12-13 games each. When faced with this 3-way tie connundrum, I believe it is best to use the entire season as an attempt to gauge which team really is the "best". It's a poor system, but it's all we've got to work with.
12 comments:
Boomer!
In case you were wondering about the history of tiebreakers in high school sports (which I SO think you are), my sophomore year, East's women's soccer team was tied with Bountiful and Woodscross for the top spot in the region going into state. Each team had beat one of the other teams (just like the UT, TT, OU teams) but East ended up going into state in first place. How?
Coach Kernodle won the coin toss.
And that year, we got 2nd place, losing to Roy. Bountiful and WX lost in the first or 2nd round.
Bet you're glad I shared that, right?
A sound analysis. I can't argue with anything you've said or even the end result. It's just the whole BCS system is unsatisfying, as it seems to be every year.
Poor TT gets no love - but I have no opinion on whether it should be UT or OU in the title game. There are strong arguments for both.
If OU loses to Mizzou in the Big12 title game (not likely, but possible - which secures an automatic BCS berth for Mizzou) do you think that Texas (who didn't even win their conference) should get the chance to play for the National Championship?
So, I disagree that OU is the better pick based on the comparison of the three games. While I understand the arguments for comparing the results of the games the way you're approaching it, by making the score of the game relevant, you're introducing a much larger impact of relative factors into the result.
Consider, BYU dominated UW in Seattle. Yet they only won by scoring 1.037037... times as many points. This was a combination of crooked officiating and really really bad luck. BYU then proceeded to dominate UCLA scoring infinity times more points. Infinity is still at last check much greater than 1.037037. So using that you can predict how badly UW must have beaten UCLA.
(You can do it arithmetically if you prefer. The results will be more reasonable; however, they still won't make sense.)
No, the only fair way to settle this is to put all three teams on the same field. Start out by playing a half at a time and if that doesn't work (it probably won't) go to overtimes. The first team to beat both of the other teams wins.
I'm not sure if everyone knows this, but there are two people named Ben who are known to comment on my blog. One is from Oklahoma, the other from Idaho. The first comment on this post is from the OK Ben. ID Ben is comment #5.
Ben: I agree that bringing scores into comparisons is asking for trouble, but what other metric do we have to use? I would be glad to hear arguments for UT deserving to go to the Big12 championship game, and I'm certain that there are some interesting arguments, though I don't know what they are other than the head to head match up. And using the head to head game just drags us into the spiral of A > B > C > A.
Another interesting tidbit that I just saw on espn.com is that Texas would have advanced to the Big 12 title game if the Big 12 used the same tiebreaking strategies as the ACC, C-USA, the MAC or the SEC.
Things like this are what make the BCS look retarded. There is too much emotion and opinion that goes into these polls. All champions should be decided on the field.
I told you, make them settle it on the field all three teams in the same place at the same time. Missouri could have waited a week.
Consider what you'd do if the three teams played exactly the same schedule. And the pollsters had all three tied. Further, suppose the scores of the respective games all played at the winners home field were:
21 - 20 A beats B
21 - 20 B beats C
34 - 20 C beats A
Game 3 was an overtime win. As dominating an overtime win as possible in fact. Who would you send?
Same scenario but now
7 - 6 A beats B in the middle of a snow storm.
93 - 0 B beats C on a clear calm day but 33 C players were suspended for undisclosed violations of team rules.
93 - 0 C beats A in a worse snow storm at A's home field with 20 C players still out for undisclosed rule violations.
Interesting though the mathematics may be, this is a case of the metric you choose deciding what is most important. If the metric is score, then teams who play in favorable conditions have an advantage. If the metric is quality of schedule, then teams who play teams who get lucky have an advantage. (Here BYU gets hammered by Jake Locker breaking his hand. Other conference schools get a huge
bonus from Oregon State upsetting USC.)
The argument for Texas if you want to make it is this. OU loses to Texas on a neutral field. Texas is now ahead. Texas Tech barely beats Texas with home field advantage. Texas wins this game if one factors in home field advantage. (Another reason I don't like that kind of calculus). Now OU beats Texas Tech in Oklahoma. The only team who won a non-home game here was Texas. And they almost won the other game.
Now, the argument for Texas Tech. The only game they lost was to another 1 loss team on the road. Sure they lost big, but if they'd played not to lose big instead of to win would you like them more?
--Ben (ID -- though now WA)
Tysqui, the ACC tiebreaker rule in question reads:
The tied team with the highest ranking in the Bowl Championship Series Standings following the conclusion of regular season games shall be the divisional representative in the ACC Championship Game, unless the second of the tied teams is ranked within five-or-fewer places of the highest ranked tied team. In this case, the head-to-head results of the top two ranked tied teams shall determine the representative in the ACC Championship Game.
This is not better. The proof is left as an exercise for the reader.
Tyler, I'm curious as to what your "decide it on the field" method would be. While Ben's method would be very satisfying, I think we would all agree that it would be next to impossible to implement. Where would they play? A neutral site? Which teams play first? (Which ever team plays last faces a partially worn out opponent.) Would there be a planned week off before the championship game just in case of a three way tie, or would you schedule it the week before? How do you schedule a venue the week before? Do you now have to schedule the championship venue for two weeks? For a system that has spent decades trying to figure out how to crown a national champion, creating a non-scheduled extra game with different parameters would be impossible. (That doesn't mean it's a worse idea than the BCS).
I'm not sure that there exists a metric such that it wouldn't be problematic which could account for all possible situations.
I'm curious as to who the ACC tie breaker would pick out of this situation. OU is #2, UT is #3 and TT is #7. They're all within 5 spots of each other.
Finally, Ben, I would like to see an overtime game where a team scores a touchdown, gets the 2 point conversion, and then returns a turnover for a touchdown. About the only less likely scenario for an overtime is something involving a safety.
So, in the one case I've seen it done, they met at a neutral field on three days notice. The team that was rested did win the second half. They then went to the OTs and I don't remember who won. It would obviously be more logistically difficult in the college world.
As for the ACC, the BCS tiebreaker begins by eliminating the third place team. Then the head to head winner of the top two gets ahead. Of course if OU had one more loss, then Tech would get the advantage
As for 'on the field' I was referring to the BCS in general and not actually the system used in the Big 12 (yes, deviating from the subject of your original post). If there was a playoff system in place, Texas would not be nearly as riled up for being left out of the Big 12 Championship game because they would still have a chance to play for the National Championship (and would possibly actually benefit from extra rest as compared to Oklahoma).
As for a 3-way tie, high school football has often implemented a tie-break game. Two teams play one-half and (decided by a coin toss) and the winner stays. First team to beat both of the others in one half of football wins. While I think it's a better option than the system in place, I don't think that it's really feasible.
I would simply prefer a playoff in general that is not forced to pit conference champions against conference champions (because some conferences suck).
Post a Comment