When, Who, and Where

Submitted by bronxblue on December 3rd, 2008 at 2:17 PM

As Sgt. Wolverine noted (as have others), a playoff only crowns a playoff champion, not a true National Champion. That's why, instead of trying to shoehorn some logic into the BCS or crying out for a 6, 8, or 12, or 16-team playoff, college football might as well just go back to its earlier pre-1998 system and just accept the "Mythical" before the NC and give the power back to the human polls.

When

Now hear me out. I look at sports at the elite level the same way I view entering an elite university, the gaggle of beautiful people trying to enter a trendy club, or obtaining a great job. At this level, when everyone/everything has virtually the same resume and skills, people are left trying to distill out the slimmest of stratification to justify why one team is better than another, and by default people believe that head-to-head somehow "proves" who is better. Yet, as we've seen numerous times, there is a myriad of variables that can greatly affect even the outcomes of these head-to-head match ups, let alone a full playoff. First off, when you play an opponent can have at least as much of an effect as who you play. We all remember 2006, when OSU and UM were two of the best three teams in America, played each other in a solid game, and then had to wait 50+ days to take on Florida and USC. Rust, overconfidence, over-scheming, etc. can all set in over that time, and what you get are two blowouts that I severely doubt would have happened if the games were played a week after the Big 10 finale. Conversely, look at Texas this year - they played 4 ranked teams on consecutive weekends, and then finally lost on a last-second touchdown grab. Teams get hot, teams get tired, players get hurt, etc., and when the talent is approximately equal, all a playoff shows is who was able to get hot, avoid injuries, and find a favorable draw for 3-4 weeks. So not only does a playoff only crown a playoff champion, but it now also crowns a playoff champion for the past month.

I still remember the kerfuffle that Colorado made a few years ago when they got red-hot at the end of the season, destroyed Nebraska (the fighting Eric Crouches) and beat Texas for the Big-12 title, and complained they were shut out of a chance at the NC even though they were playing the best football in the country at the time (of course, they then were destroyed by Mr. JJ Harrington and Oregon, yet more proof that "when" is a big deal). Of course, this same team had already lost to Fresno State and rather convincingly to Texas earlier in the season. This team wasn't the best team in America that season, but match them against Miami that year, one week after beating Texas, and who knows how that game plays out.

Who

And the benefit of a favorable draw also plays into the inherit weakness of a playoff in CFB because it all falls on who you play. Look at the participants/scores in last year's BCS games:

Rose - USC/Ill (49-17)
Sugar - Georgia/Hawaii (41-10)
Fiesta - WVU/OU (48-28)
Orange - Kansas/VT (24-21)
NC - LSU/OSU (38-24)

I see a slew of blowouts plus one game (KU/VT) that neither team really seemed to want to win. So does that mean USC, Georgia, WVU, KU, and LSU were the best 5 teams last year, and LSU was the clear champion? Of course not. In fact, I would hazard to guess that if you flipped LSU with Georgia, that NC game turns out differently. Or flip WVU and KU, and maybe OU doesn't get obliterated and the world is deprived of the Bill Stewart Face. And yet, because these lineups were defined by a combination of tie-ins and expected ratings, people somehow believe that they provide an "objective" metric for how good each team was. That just doesn't make sense to me. These match ups never aligned the best teams against each other no more than the traditional bowl tie-ins created the best match ups for determining a true NC. All you have are a bunch of somewhat-arbitrary match ups that may place a superior team in the crosshares of some insanely-hot team, or a team that just lost its best player for some reason (remember Cinci back when Kenyon Martin played - that team was a legitimate NC contender in basketball until Martin injured himself right before the tourney).

Well, the same thing would happen with a playoff. How do you decide who is "better" when the BCS and the polls have yet to show an ounce of consistency in determining who the best 2 or 3 teams are a given week? Sure, SOS, conference strength, head-to-head match ups, etc. provide a component, but ultimately you are just left with more distractions and after-the-fact justifications, not a true barometer of what teams truly are the "best".

For example, take this season. Provided Alabama loses this weekend to Florida, why should Florida be a #1 seed and Oklahoma a #2, or vice-versa? How about Texas, Utah, Alabama, and PSU battling over the #3-6 spots? Depending on which slot each of these teams fall into, their chances of winning or losing vary wildly. Presuming Boise St. and Ball St. round out this 8-team playoff, Texas would play a weak Ball St. team while Alabama and PSU would hammer away on each other for the opportunity to then play Oklahoma or Florida in the next round. One week earlier, Alabama is the #1 team in America, and now you have them fighting for their life against another team that was one missed field goal away from a perfect season. That's the second inconsistency of a playoff - you may only be as good as the team you match up with that weekend, yet everyone somehow presumes that all the other teams that lost either contemporaneously or earlier are somehow worse. So not only do you only crown a playoff champion for the past month, but you now only crown a playoff champion who played that particular configuration of opponents.

And yes, a round-robin works in theory, but there is no way you can expect college football teams to play into February, even with a shortened regular season. The games are too nuanced and too physical - the reason it works in March Madness is because players and coaches generally can gameplan and recover for another opponent in a few days, while in football the players would be walking corpses after late December.

Where

Finally, the third factor that people tend to forget about a playoff series is that certain teams enjoy the benefit of home field advantage even when it was not necessarily intended. Since I presume the only way a playoff will work is with the consent of the bowl alliance, most of these playoff games would need to be played on "neutral" fields sponsored by the various bowls. True, you should reward teams with "better" records with home field, but you'll inevitably have the problem we've seen over the past few years with certain warm-weather schools (looking at you, LSU and USC) playing meaningful games virtually at home because of these relationships. Heck, USC virtually never has to leave Southern California during bowl season, and I still believe that LSU playing in the Superdome last year helped them immensely against OSU. Sure, the traditional bowl games were not anything better (USC still has the Rose Bowl, LSU the Superdome, Florida/FSU/Miami the Orange Bowl, etc.), but at least there you could argue bowl tie-ins were at work and every team kind of understood the worst-case scenario at the beginning of the season. Admittedly this is the weakest of my arguments, and that is why I left it until the end. That said, now you are left crowing not just a playoff champion who played a particular configuration of opponents over a shortened time frame, but also a playoff champion who played those games at particular fields.

Summation
If you have made it this far, then congrats, and you are almost done. I used to be a huge fan of playoffs in football, believing that it somehow crowned a truer champion than the USA Today/Coaches poll or the BCS title game. Yet, the more I thought about it, I don't see it doing anything more than slapping a different coat of paint on the same broken-down car. Implementing a playoff would, ultimately, appease some fans for a short period of time until its flaws would begin to show, probably when a team like BYU runs the table or a 3-loss SEC team gets insanely hot and wins the NC. Personally, the only thing I dislike more than the uncertainty the polls created is the faux-certainty and legitimacy that the BCS gave to each season's MNC. At least with the polls, early-season struggles were at least partially remembered, and getting hot at the right time wasn't necessarily rewarded with any more or less outrageousness than in a playoff. And for those concerned that the polls would overlook worthy teams like Utah or Boise St., I'm all for a rule that would allow a mid-major ranked in the top-10 to knock out a lower-ranked opponent with a tie-in from one of the "big" bowls. Sure, that might cause the occasional outrage and isn't perfect, but then at least you give these teams a chance to showcase their abilities against a top-flight opponent. Best-case scenario, a top-flight team like Utah goes undefeated and miraculously bypasses all of the 1-loss teams and is named the MNC; worst-case scenario is they go undefeated and are still passed over by a 1-loss power team, which is the exact same result we've seen since the beginning of the BCS/time. I am all for abolishing the BCS; I just don't think we need to replace it with some convoluted playoff system.

Comments

Chunks the Hobo

December 3rd, 2008 at 3:44 PM ^

I was gratified first on another thread and now on this one to find I'm not the only guy on the planet, apparently, who would just like to go back to the old bowl system (and ideally, though I know impossibly, with their old non-corporate names). I think anything else just wrecks the sport to one degree or another.

But I'm just a hobo so noooobody listens. Bastards.

arod

December 3rd, 2008 at 9:46 PM ^

So you think there no such thing as a champion in any sport? I'm confused. In other sports, such as the NBA, NFL, and NHL, we accept that some teams get hot in the playoffs. We call these teams "good teams," because they play the best when it matters most. We don't say that the Miami Heat won a mythical championship 3 seasons ago. We say they were the champions. I understand that the team that wins a playoff may not be the team that had the best regular season, but that's part of the excitement of the playoffs.

I'm sorry, but I really don't understand the point of this post. While it would be ideal, perhaps, if we could have all the teams in college football play each other and then crown the team with the most wins the "champion," that's just not feasible. Simply because it is not feasible does not meant that we should give up trying to approximate the result the best we can.

Let me also point out that the hypothetical team with the most wins after all the teams (hypothetically) played each other may not be the "best" by some definition anyway. That team, call them the Patriots, may be such that certain teams or a team (call them the Giants) will beat them (consistently in multiple plays) simply because of the way they matchup. This can be seen game theoretically in the Col. Blotto game (google it). In most years even the Patriots will have their Giants, i.e. the "best team" will have a team that is "better" in that that team will beat them head to head. Yet, such is life. I think the NFL, etc, are still better systems because at least the criteria for success, i.e. making the playoffs, winning the Super Bowl, are better defined. Making the NC in college is very much a roll of the dice, whereas in the NFL the team has more (although not complete) control.

Another way of looking at it: We ought to eliminate subjective polling and so on as much as possible, because that muddles everything up. If you accept that, which I think most do, then you should favor a playoff, because at least a playoff lessens the influence of polling to some degree.

Oh, and I'm pretty sure March Madness isn't a Round Robin tournament. It's single elimination, just like a playoff.

bronxblue

December 4th, 2008 at 11:58 AM ^

thanks for the counter arguments - it points out some of the areas I failed to flesh out.

Per the March Madness being a round-robin, I didn't make clear that I was referring to multiple games a week/shortened time between games. I know March Madness is not a round-robin, but the point was that teams can play 2-3 games in a week because the recovery time between the individuals games is less demanding.

As for the argument that I don't like championships in any playoff-based sport, I understand that that is a major flaw in my argument. I don't disagree that playoffs work as a means of crowning a champion in a sport; I just don’t see how they are appreciably less arbitrary than the system the NCAA employed for over a hundred years prior to the BCS.

My biggest concern with a college football playoff centers with the determination of seeding. With an 8-team playoff, the difference between the #1-seed and the #5-seed, for example, may be nothing more than a single win or loss, with no common opponents or some other relatively objective metric to truly differentiate between the two teams (SOS, conference rankings, etc. are likely pretty close). Yet, depending on how a few computer numbers or votes work out, the #5 seed is facing an immensely tougher road to a championship. This isn’t basketball, baseball, or hockey, where teams play 80+ games and usually have a large number of common opponents to at least bring some objectivity to the seeding. Sure, playing in a weaker division helps, but at worst that usually results in a weak #3 seed battling it out with a stronger #6, with neither team having a great chance of winning it all. Plus, the teams play multiple games in a series, meaning the flukiness or the effect of a “hot” team tends to work its way out over a longer series, as compared to a one-game playoff where a hot-shooting guard or stonewall goalie can truly make a difference (and yes, there are examples undermining this argument [e.g. JSG for the Ducks], but no theory is perfect)

Even college basketball works better, since a field of 64 allows a greater stratification between the top teams. Sure, you’ll see some debates between whether a team should be a 6 or a 7 seed, or a 3 or a 4 (and yes, I remember NC St. winning it all), but the top 8 seeds are generally “rewarded” for their superior records by playing significantly lower-ranked teams for at least the first weekend, if not all the way until the Final Four.

You just don’t get that in football, especially in college. In the NFL at least, teams play somewhat objectively-determined records (the NFL front office bases the record on past season’s performance, unlike the NCAA where OOC is geared toward revenue plus whatever conference you play in), and rarely is one division so dramatically horrible that it makes an immense difference in how a team is seeded. Admittedly, there still are instances where good teams are trapped in really tough conferences and are seeded lower than they should be (e.g. Steelers when they won in 2005), and sometimes a 9-7 division winner is seeded above a better team, but the relative normality of an NFL schedule at least comforts me a little that the seeding generally rewards the best teams in the game.

Not so with college, as there simply are too many variables that are never corrected for in the seeding. Plus, presuming the playoff seeding tends to follow a similar system as that used by the BCS, we are still effectively relying on human voters to decide who plays for the NC and who doesn’t. That was always my biggest issue with the BCS – people acted like it somehow “legitimized” the bowl games because some statistical data was included in the rankings, completely ignoring the fact it still relies heavily on the same voter rankings that it was originally designed to bypass. So I guess my point is that, with an arbitrary system (the polls) already in place that has worked, by and large, for years, why introduce a new system (BCS/playoff) that would still have a number of the same flaws as the earlier one, while only adding a hint of faux “authenticity” because a Borda count was utilized? Honestly, people still complain almost every year a team is crowned the BCS champion, whether it be because a #3 should have replaced the #2, or three undefeated teams, etc. I don’t see how a playoff would eliminate these questions, and that being the case why not stick with something that wasn’t any worse. I’m not condemning the playoffs in other sports – I’m just questioning how it would produce a more palpable champion in CFB than before.

chally

December 4th, 2008 at 1:12 PM ^

I agree with the OP and Chunks the Hobo. Playoffs are a ridiculous and arbitrary way of crowing a champion that would ruin the only sport worth watching from the first coin flip to the last gun.

Seth

December 4th, 2008 at 4:03 PM ^

It seems these days the parties on the NCAA football post season are down to the Aristocracy and its BCS, and the plebians and their playoff. I'd given up hope for a rational solution.

Bronx is right -- allow me to help him make his argument better.

I. CFB IS A STATISTICAL NIGHTMARE

THERE ARE OVER 100 TEAMS, AND ONLY 12 GAMES
The name of the game is Sample Size. Over 162 baseball games played between 32 teams, eventually, you can almost be assured that your best teams will be at or near the top of the heap. College football, however, asks us to figure out the best of a much larger group, with only 12 games to go from. This is why little things like how loudly Urban Meyer can cry affect the outcome.

Any playoff system we deign to create, be it a field of two as the BCS does it, or a field of 8, or whatever, will invariably be arbitrary.

You can't expand the sample. If you want to know why we can't have any more games, I can't help you. Not that a 13th, 14th or 15th game would solve the statistical problem.

So the other answer, then, is to pare down the field.

II. CONFERENCES ARE NOT EQUAL
The way to pare down the field is to institute equal divisions. You can't use 12 games to determine the best among 113 teams. In fact, you can't really use 12 games to figure out the best among four teams. But you have a better shot.

Inevitably, you will have divisional disparities, but if the divisions are made fairly, these disparities will even out over time.

Not so with the conference system. The conferences are NOT designed to be equal divisions. Rather, they're supposed to be a codification of regional and institutional advantage. A Big Ten or SEC team has institutional advantages that allowed it to join the elite conferences in the first place. Furthermore, even though CFB has nationalized, there are still distinct regional differences in play.

So simply pitting the champions of the conferences against each other is stupid, even if it's helpful to at least use the winners of the conferences to pare down the statistical problem.

III. CFB SHOULD NOT PRETEND TO BE SOMETHING IT'S NOT
All of this shows that any system, be it BCS or a 4-, 6-, 8- or 12-team playoff, would be arbitrary and capricious. There just isn't time or a small enough field to determine an undisputed, rational, single national champion. It's an exercise in futility.

Rather than shoehorn this sport into providing a Sports Illustrated Special Offer national champion on the field, the sport should focus on what it's good at: fall Saturdays of facing the teams you know, i.e. CONFERENCE PLAY. Conference games are awesome. You know those teams. You know the history of those teams. You can drive there and stay with the guy you went to high school with who lives there. And at the end of the season, among 12 teams, you CAN, more or less, usually decide on a winner.

IV. NOT THAT YOU'LL GET A PLAYOFF ANYWAY
The problem that's never properly answered by the pro-playoff proletariat is that of making sure the NCAA and the bowls still get to eat. The BCS sucks balls but it makes oodles of money because it doesn't change the one thing the bowl system had going for it: you know where your team is gonna be. Bowls are, by definition, HUGE stadiums. You can't just shift 90,000 people around the country in a week. Most bowls still can't fill their seats with a month advance notice, and the lengthening season has made that all the worse. Many bowls now have to wait until after the conference finals -- IN DECEMBER -- to tender their invites. Each passing day is another that Michigan State fans are not booking rooms in their bowl city.

Playoffs would make that worse, since now pretty much EVERY BOWL would have to wait for the conference championships to make their invites. And though it sounds good on paper, how many people are showing up to see Utah v. USF in Pasadena, California, with less than a week's notice, even it's the national semifinals!

The bowls work because teams and fans and bowls and TV networks have time to know where they're playing, whom they're playing, and for what they'll be battling. The BCS has hurt this enough. A playoff system -- any playoff system -- would RUIN the bowls.

This is why the Tophat Aristocracy disregards any playoff noise from the proles; they know their system makes money, and they know yours will make less, purity of competition be damned.

IV. IT IS DAMNED
Which is why it's so damn easy for the NCAA to callously stick with the BCS amidst the carols of boos from media and fans. You can't declare a true national champion with 12-13 games amidst 113 teams. They know it. There is no pure champion.

V. EXCEPT FOR MY CONFERENCE
Exactly, subhead writing guy.

VI. FINAL ANSWER
I say let NCAA football be NCAA football, and focus on conference play. If we have 12 games a season and that's not going to change, fine, make 10 of those games mandatory conference games. At the end of the season, the conferences will face off in their traditional bowls in an exhibition, and a fun way of answering which conference is better, yada yada. If the Big Ten and Pac Ten already have a nice bowl thankyouverymuch, then the 2nd and 3rd place B10 and SEC teams can play each other for best conference braggin rights.

As for the national champion, it's going to be arbitrary anyway, so let's at least let the experts be the arbiters rather than chance and computers and Urban Meyer's wet diapers. Let's poll all the coaches, and poll all the sportswriters, and poll all the bloggers, and then when the final standings are printed, spend the offseason bitching about it.

DO
YA
KENNIT?

bronxblue

December 4th, 2008 at 5:21 PM ^

Thanks for making my general argument a bit more coherent. I’ve only taken one statistics course in my life (along with a slew of calculus, probability, etc. courses through the CSE curriculum), but the overwhelming theme was that you only receive relevant results if your sample size is sufficiently large. Too small or incomplete a sample size, all you are left with is massive standard deviation and no concrete basis for conclusions. Well, if the ultimate goal of any playoff is to crown a “true” champion, relatively free of the arbitrariness often identified as a foundation of polls, then it would stand to reason that you need to not only include x number of teams, but preferably they would need to all play each other and, perhaps, multiple times.

That’s why I don’t have a problem with conference championships in basketball and football, since each team plays all, or at least most, of the teams in its conference at least once. Then you have head-to-head results, standardized opponents, etc. (even though there are a million other variables to consider). But there are 100+ teams in CFB, and most of them never step onto the field with each other, in all fairness, it would be impossible to make it happen. So don’t traipse out a playoff system, seed it based on the same faulty logic that apparently led you to implement it, and then presume that the results from that single permutation embodies some absolute truth about which team is the best in the nation.

My whole point has always been that I don’t dislike playoffs in sports if it is sufficiently popular and everyone acknowledges its flaws; what drives me crazy is the notion that the results of a playoff are somehow measurably better than the system we’ve had in place before.

papabear16

December 4th, 2008 at 4:05 PM ^

Good post, really. Now, there is a flaw, because leagues always use playoffs to crown champions. I think a better question is whether college football really needs one, if the best we could do is what you describe. Then, I agree with your conclusion.

I'll also add what I always called my February Argument. One February, I was working on a case against a lawyer from Austin, and on her profile, I saw she graduated from Nebraska in '99. So, she was there in '97. I was at Michigan in '97. Now, we had something else to fight about in the lighter moments, and it was a fun fight. My point is that if all we can get is a pretty shitty champion, and that appears to be the case (because the 'who' and 'when' arguments apply more strongly to college football, I think, than any other sport), then I prefer the mythical champion, the traditional bowls, and the February bar arguments.

bronxblue

December 4th, 2008 at 5:25 PM ^

Same here. I don't mind the mythical arguments - I've gotten into debates with friends from other schools on the matter, even when it does not explicitly involve UM (e.g. I have family at OU, and we've argued about Auburn getting a shot at the NC game over them). Unless someone provides me with a truly better system than some relatively-small playoff field, why not stick with the traditional bowls and keep the same water-cooler debates we've always had.

WolverBean

December 4th, 2008 at 8:45 PM ^

We can take this one step further:

Who says there is any such thing as "best?"

After all, how often does a single team lead the nation in total offense, scoring offense, total defense, scoring defense, rushing defense, passing defense, yards per rushing attempt, sacks... you get the idea. Even if we let SMQ tell us which 5 categories correlate most closely with a winning record, and then consider only those, odds are long against any single team being #1 in all five of these categories at the end of the season.

On which note:
Another way of looking at it: We ought to eliminate subjective polling and so on as much as possible, because that muddles everything up. If you accept that, which I think most do, then you should favor a playoff, because at least a playoff lessens the influence of polling to some degree.

I read this, and at first, I agreed. But here's the problem. Let's take the 9 statistical categories SMQ has shown are most strongly correlated with having a winning record, use overall statistical rank across those 9 categories to do the seeding, and look at what the top eight seeds would have been last year:

1. Kansas
2. West By God
3. LSU
4. USC
5. Oklahoma
6. Clemson
7. Ohio State
8. Cincinnati

Do we mean to suggest last year's Cincinnati team was a "better" team than Georgia or Missouri, neither of whom appear on this list? We can argue the particulars of which statistics should or should not be included in the "overall statistical rank" approach, or about ways to weight conferences by statistics, but either way, what we're really arguing about is the inevitably subjective decision about what objective factors to use. Which is why in the end, I think perhaps it's best to simply accept that there's no such thing as a "best" team. If we still wish to define a "champion," that's all well and good; but for all the reasons listed in the original posting, we should stop thinking of any "champion," mythical or otherwise, as the "best."

arod

December 6th, 2008 at 12:05 AM ^

using a team's win-lose record as the "statistic" most strongly correlated with winning.

There is no perfect way to determine the best team, however, I would think that instead of having 2 teams picked via polling or a collection of statistics, a system whereby 8 teams are selected would be preferable, simply because the "best" team has a great likelihood of being amongst the 8 selected. Now you might say yeah, but the "best" team might not make the NC game because of the seeding. True. But, at least that team would have a shot. By implementing a playoff it seems that we would be increasing the likelihood of getting the "best" team to be crowned as the champion.

I'd like to point out that much of what people are saying in this thread is correct, no set of factor is going to be perfect and the notion of best is probably amorphous and so on. But we still play billards. Simply because things are hard or even impossible doesn't mean we ought to abandon trying to approximate them. I get the impression that people are thinking in terms of "arbitrary" and "non-arbitrary" when we should really be thinking in terms of degrees of abritrariness (only a Sith deals in absolutes).

bronxblue

December 10th, 2008 at 10:31 AM ^

Well said, and I certainly agree that a playoff based on records would not be a horrible idea. I guess I've softened a bit on my stance because people seem to want a playoff, and perhaps seeing it in action will either validate those desires or just unearth a whole new set of problems. But, again, the problem with wins and losses is that there are degrees. Last year LSU lost two games, both in overtime, and yet still played for the MNC. And watching them play, I completely believe that when healthy, they were the best team in the nation last year. Wins and losses are largely decided by who, where, and when you play, your conference, OOC, etc. It might be less arbitrary in some respects than a poll, but polls also take into account these factors better than pure wins and losses. And even if the playoff weighed these factors (like the BCS), you would still have any incomplete view. I'm not advocating against a playoff as much as I'm just questioning how much of an improvement it would be. That said, I agree with your "degree of arbitrariness" argument.

tricks574

December 5th, 2008 at 3:57 AM ^

That is the dumbest thing I've ever heard. Every system we could have in place in college football would be put towards determining 1 thing, the team that had the best season in college football. Now, we can't have a 116 team round robin tournament, so we need a way to separate the wheat from the chaff so to speak. The regular season in any sport is meant to do this, to get in a multitude of games and try to determine which teams deserve a shot to extend their season. The playoffs then pit the teams who have earned a spot against each other to earn a title.

I know with a playoff system someone will be left out, but I would rather have team number 9 bitching than team number 3.

bronxblue

December 5th, 2008 at 12:03 PM ^

I seriously doubt this is the dumbest thing you've ever heard - try reading every Drew Sharp column this year and then come back to me.

As to your argument that the regular season somehow defines who had the "best" season, and that a playoff merely pits the "best" teams against each other, I still hold true that college football is full of WAY too many variables for this to every really be true. Listen, I believe that the conference seasons do provide a reasonable metric for determining the best team in that conference, but to extrapolate much data beyond that doesn't hold water. How do you figure out who is the "best"? Name me some objective metric that definitively ranks what teams deserve to be included in the playoff, and why the #10 team is so much worse than the #8 team? People can't, and usually they fall back on the argument about polls, letting human voters who watched the games, followed these teams all season, etc. work out who is the best. Well, that sounds pretty arbitrary to me, and honestly no more arbitrary than the pre-BCS system of final season polls.

Playoffs work in most other sports because you have a long enough season, and enough common opponents and parity, that the best teams tend to rise to the top. You don't get that in CFB (and a lesser extent the NFL) because you are only working with a sample size of 12-13 games a season, with many of them a mix of patsies and conference games. Going undefeated in one conference (e.g. the MAC) is not the same as going undefeated in another (e.g. the SEC). So the CFB regular season simply doesn't provide you with a "multitude of games" sufficient enough to define who is the best - heck, it barely gives you enough to determine who is the best in the conference.

You say that you rather let the #9 team whine than the #3, but that doesn't address how to figure out who is #8 and who is #9. Sure, the more teams you include the less ground the outliers have to argue, but unless you want to have a 24 or 32-team playoff (and even then people will complain), I see a large public outcry from the #25 or #33 teams.

In the most recent BCS ranking, you have Texas Tech, PSU, and Boise St. 7-9. Now, is TT really better than PSU, who is one field goal away from a perfect season, while TT was dominated by OU and struggled in some other games? How about Boise St. - they went undefeated, including a win over a ranked team (Oregon). Why shouldn't they be ranked above PSU/TT? You'll say conference rankings, SOS, etc. as proof that PSU is better than Boise, but Boise and Alabama have virtually the same SOS, so the numbers clearly don't tell the whole story. And that's my problem with the proposed playoffs - they are based on nothing more than a clusterfrick of cherry-picked numbers and voter opinions, just like the old system.

tricks574

December 5th, 2008 at 12:25 PM ^

But no matter the system someone is going to have a gripe. You make a playoff and you give any undefeated team an autobid. No autobids for conf champs or anything else like that though. That way someone has to beat you to not win the title. I realize its "more of the same" in deciding who gets left out, but honestly, its much easier to seperate out the teams. You would leave out TTU because they got killed by OU, BSU would get in because they are undefeated, and PSU would get in because they didn't get absolutly murdered in their one loss.

Not likeing a playoff because it incorporates some of the same system is simply saying that if we cannot have absolute perfection we should not strive for any improvement at all. It doesn't make any sense, at all.

bronxblue

December 5th, 2008 at 2:48 PM ^

I guess I just don't see the improvement. You decided that TTU is left out because they lost to a top-3 team, but then allow Florida in even though they lost, at home, to a bowl-eligible-but-mediocre Ole Miss. See, that was pretty arbitrary. And a playoff would just duplicate that same argument over and over again. We can just agree to disagree here, since I don't see how a playoff is any real improvement in defining a champion in CFB (but I agree it is the best option in other sports).

Seth

December 5th, 2008 at 2:57 PM ^

Tricks does have a point. Yes, the bigger your playoff field, the more likely you are to get the best team in college football included. This is still not a guarantee. In fact, isn't it entirely probable that one or more Mid-Major Conference teams in a year will go undefeated, yet finish 9th or 10th behind a wash of major conference 1-loss teams?

And what about the other problems? If you're taking the top 8, that means you need three rounds of playoffs to get the winner. When will you have these? When will you do the selection? How will you give down-ticket bowls time to sell their seats? How will you get the conferences and the bowls and the NCAA to agree? What about the schools themselves, who prefer the extra weeks of practice and the chance to integrate recruits during a week of fun in SoCal or Orlando to spending Christmas through the first week of the semester traipsing around the country?

And not to incite the "you're so naive" brigade, but haven't you considered how the extra games would affect, you know, finals, and the first week of the next semester, considering an education is what we gave these kids to play here in the first place?

But if you want to forget all that, leave the realm of pragmatism -- in fact, disregard practicality entirely -- and just design a moon-shot playoff system for your alternate fairy universe, I'm down. Let's do it.

If you want to make the playoff field large enough so that you are assured of including the best team, I would say 12 teams is when you're getting to probability levels of 99 percent+, i.e. we're 99 percent sure the best team of 116 is one of these 12.

To even out play, we'd have to scrap the conferences and rivalries, and no longer allow teams to choose their own matchups, since they can use this to game the system. Instead, computers would determine schedules using a blind test based on a composite of each team's previous three-year records, and spit out a schedule of roughly equal competitiveness. It can be programmed to factor in proximity if you like to cut down on travel.

After a 12-game season, we'd have records that are relatively comparable. For tiebreakers, we'd use a composite of team defense ranking, team offense ranking, and point differential.

We'd have to get rid of all the small bowls, of course. Maybe all of the bowls; I mean, it's pretty unfair if the "neutral" site is always at the home stadiums of Miami, UCLA, Tulane and Arizona State, aye? So neutral sites should be chosen after the teams are selected each week, so that each team needs to travel approximately the same amount.

The top four teams would earn bye weeks, and the other 8 would play each other. Then the remaining 8. Then the last 4. Then the last two.

Champion declared.

If you'd rather use the polls, however, to keep current scheduling, you'd need to expand the playoffs further to 16 teams. Of course, this is arbitrary as well; there's not a lot of difference usually between 14 and 20, really, is there? And I would recommend, too, a round-robin tournament of 10 rounds to fix the seeding problems created by the arbitrary polls.

This isn't what you asked for, though. You said let's just accept the arbitrary polls, in some form or another, and Team Number 9 et al. can, as one Fingerle Lumber employee was wont to say, "Deal with it."

My question for you, then, is, why is this worth it? Why blow up the bowl system, force fans to pay jacked-up last-minute airfare to whatever the next town their team is playing in 5 days, and watch half the bowls fold, just to end up with another arbitrary champion?

And what happens when No. 8, 3-loss Florida who jumped four slots into the tournament thanks to a profound week of Urban Meyer media bitching, runs the table four games in a row that just happen to all be played in the State of Florida, and is declared the national champion, in a year when there was a 1-loss Michigan team that entered the tournament undefeated and was only was barely upset in the semifinals round by USC and a blind referee-induced fluke by in USC's home stadium, a game wherein, just saying, USC's best three players were injured and unable to perform in their final against Florida?

Personally, I prefer knowing if we're still perfect at the conclusion of the Ohio State game, I can throw roses onto the field, then spend Thanksgiving Break making plans for Pasadena. Half the teams who go to bowl games get to end the season a winner. And the real prize each season is your conference championship, where you actually know you were competing against schools of similar size and prestige, most of which you actually played, under the same rules and limitations and playing style.

If you say declaring a national champion, i.e. winning, is the point of any sport, I say to you, PLAY A FUCKING SPORT. Winning is A goal, not the only goal. I dare you; take a poll of men who've played college football in the last century. Maybe 1 percent of them ever won a national championship. What percent of them, do you think, would say the experience was worth it? More than 1 percent?

College football was great long before anyone really put a ton of care into what constitutes a National Champion. Notre Dame still claims a national championship given to them by an AP poll before they got crushed in their bowl game. What's interesting about this is that we argue over that season's national championship more today than they did back then. College football is great because it's got marching bands, and monochromatic student sections, and fight songs, and rivalries that are analogous to sibling relationships or major societal differences, and super athletic 20-year-olds hitting each other, or making us say "Wow," and cold Saturday mornings, and tailgates, and traditions, and every November Michigan plays Ohio State. Long as I can keep all of that, the National Championship can go to hell.

bronxblue

December 5th, 2008 at 3:03 PM ^

Thanks. This diary may break the record for longest posts on this site, and a piece of me would be extremely happy to see it.

I agree with your assessment, and it was fun to read. I'm not sure anyone else here will enjoy it, but ah well.

tricks574

December 5th, 2008 at 12:33 PM ^

Because you actually have to play the games. This is all the needs to be said. If you need a 12 paragraph dissertation to argue against a 1 sentence explanation, you might be stretching.

bronxblue

December 5th, 2008 at 2:51 PM ^

Against the particular configuration of teams that some group decided represents the top 8 teams in America. You can be a fan of a playoff in CFB if you want, but don't attack my arguments with a 1-sentence response that doesn't really provide anything new to the argument beyond the cliche "Games are decided on the field."