The College World Series: How Do You Solve a Problem Like a Playoff?

A few months ago, I offered a baseball-based aside on a college football playoff. While my sentiments regarding a college football playoff remain unchanged from what they were 23 months ago, I believe pertinent points regarding postseason tournaments generally might be made by taking a closer look at how college baseball crowns its champion.

You may link to the 2007 College World Series bracket here. If you click on the preceding link, you will see that the URL contains the phrase "doubleelim," which is short for "double-elimination." This term is critical to a proper understanding of the nature of the Omaha-based baseball tournament.

In the first bracket, Rice defeated Louisville and North Carolina defeated Mississippi State in the first two games. The Cardinals, who lost the first game, thereafter faced the Bulldogs, who lost the second game. This is known as "the losers bracket." When U. of L. defeated M.S.U., the Bulldogs were eliminated from further competition by virtue of their having lost twice . . . hence the term "double-elimination."

"Listen, I know there's a lot of jargon, but some of these are pretty self-explanatory."

Rice, which won the first game, subsequently faced North Carolina, which won the second game. When the Owls defeated the Tar Heels to remain unbeaten, Rice continued through the so-called "winners bracket" and U.N.C. dropped down into the losers bracket, where it faced Louisville, the team that had eliminated Mississippi State. Since both the Cardinals and the Tar Heels had lost once already, North Carolina's victory eliminated Louisville and allowed U.N.C. to advance.

It should be noted that Rice, which was undefeated, had only had to play two games to this point, whereas North Carolina, which had lost once, had been required to play three. This is because the teams proceeding through the winners bracket are rewarded for not having lost, while those advancing through the losers bracket are forced to overcome the fact of their having been defeated. This is based upon the sensible theory that, when determining a champion, one ought to take into account whether particular teams won or lost the games they played.

In their second meeting in Omaha, the Tar Heels beat the Owls. This necessitated a third game between the two teams, since each had lost once previously since arriving in Omaha. Had Rice beaten North Carolina in their second meeting, the Owls would have advanced to the finals and another game between them would not have been necessary, because the Tar Heels' second loss would have sent them packing . . . once again, underscoring the significance of the term "double-elimination."

Once-beaten North Carolina and once-beaten Rice played in an elimination game. The loser (Rice) suffered its second setback and the winner (North Carolina) advanced to the finals, having succeeded in becoming the only team in its bracket not to have lost twice and thereby been eliminated.

That's eliminate . . . not like the drink.

The same dynamic was in play in the second bracket. The first two games were won by Arizona State and Oregon State, respectively. The losers of those initial outings, U.C. Irvine and Cal State Fullerton, respectively, faced one another in the ensuing game. U.C. Irvine won, eliminating C.S.F. by virtue of its second loss.

Oregon State defeated Arizona State, so the Sun Devils dropped into the losers bracket to face U.C. Irvine. When A.S.U. lost to U.C. Irvine, the Sun Devils' second setback eliminated them from the tournament and U.C. Irvine advanced to face the Beavers. Oregon State handed U.C. Irvine its second loss in Omaha, so an additional game was unnecessary. The Beavers were undefeated, while every other team in O.S.U.'s bracket had lost twice and, under the double-elimination format, thereby been bounced from the playoffs.

I apologize for explaining this in such a simplistic manner, but some confusion has arisen regarding the nature of the College World Series. I recently wrote:

This statement of personal preference drew "a little correction for Kyle" from a fellow weblogger:
The author of this statement received kudos from other quarters for his having "corrected [me] about the structure of the CWS tournament."

With all due respect to the obviously impassioned individuals expressing those sentiments (one of whom, to his credit, subsequently acknowledged that he had been in the midst of "incoherent ranting" when he wrote his previous post and the other of whom later confessed in a joking manner to a certain intellectual impairment), I believe what we've got here is a failure to communicate.

As Doug Gillett noted, calling a guy your "hero"after he mocks your rival for suffering the indignity of having your school steal its women---an aspersion that insults your school as much as it insults your rival---is "just sad." (Doug, by the way, has re-branded his weblog and, in his latest posting, he called our attention to the greatest Google search thread ever, so, as his picture attests, Doug knows from cool.)

I didn't need correcting about the structure of the College World Series; I was arguing that the structure of the College World Series needed correcting. Perhaps I might have made my point more plainly, but I was operating from the assumption that everyone understood how the tournament in Omaha worked.

The statement purportedly in need of correction was this: "I don't care for the idea of ending a double-elimination tournament with a best-of-three series." I will grant that this sentiment might have benefited from clarification, but nothing in that statement is in the least inaccurate.

Perhaps I would have done better to have thrown stylistic shorthand to the wind and gone with this much more awkward formulation: "I don't care for the idea of a tournament consisting of two double-elimination regional brackets ending with a best-of-three series." I must confess, though, that I seldom have been criticized for using too few words to make my point, so I suppose I should be grateful to my critics for offering such novel arguments in derogation of me.

The problem, as I identified it, was that the tournament might produce the very situation that it in fact produced: "If one team comes through the winners bracket and the other team comes through the losers bracket, the team that was one loss away from being sent home isn't entitled to a clean slate and a level playing field."

Please note the presence of the word "if," which denotes a contingency that might or might not occur. Obviously, if both teams progressed through the winners brackets in their respective regionals and arrived at the finals with unblemished ledgers, a best-of-three series would be required. Neither team would yet have lost a game and the double-elimination format which governed throughout the playoffs to that point would dictate that one team beat the other twice to eliminate the loser.

That, however, is not what happened here. The champion in the second bracket (Oregon State) emerged unscathed, whereas the champion in the first bracket (North Carolina) sustained one loss. It is my position that employing a separate best-of-three series for the finals was fundamentally unjust in such a circumstance. It is my position that the first and second brackets should simply have flowed together into a single unified bracket (much as the regional brackets blend in together to form the Final Four in the N.C.A.A. men's basketball tournament).

So as not to antagonize the Tar Heel faithful unnecessarily, this mention of the Final Four is accompanied by a reference to Dean Smith.

Under such a system, the tournament is a consistent double-elimination format throughout. When Oregon State defeated North Carolina in the first game of the finals, the Tar Heels sustained their second loss since arriving in Omaha and that should have been that. To require undefeated Oregon State to defeat once-beaten North Carolina twice is to give the Tar Heels a fourth out or, at a minimum, a mulligan. There is neither logic nor justice to such a system.

Evidently, Corn Nation, SportsBlogs Nation's college baseball hub, agrees with me about "the inequity of how the CWS is conducted - specifically the fact that the double-elimination tournament ends with a best-of-three series."

Naturally, at a fundamental level, I share Sunday Morning Quarterback's view that "there's no way I can stand for a national championship by a team that had a losing record in its own conference." For the record, there's no way I can stand for a national championship by a team that didn't win its conference championship. As I have stated before:

However, my problem with playoffs (as opposed to college football's poll system) is that postseason tournaments necessarily rely upon a smaller, and oftentimes unrepresentative, subset of games while reducing (for no particularly good reason I am able to discern) previous outings, sacrificing a conclusion based on the totality of the known facts upon the altar of the "just win" chimera:
Thus, the postseason format matters a great deal and, the more information a particular postseason format provides, the better. If, as ought to be the case, the College World Series were to employ a double-elimination format throughout the playoffs, the problem SMQ identifies would not be ameliorated, but at least the championship series would be settled upon the basis of more than just two or three games.

As it stands, the importance of the regular season is further diminished and attenuated by removing the notion that "every game counts" even from the tournament itself. Not every contest played in Omaha was of consequence, as North Carolina, alone among the contestants, got a free pass for its first loss.

Had the Tar Heels won two, and the Beavers won one, of the final three games, U.N.C. would have captured the national championship trophy despite having lost as many times in Omaha (twice) as every other College World Series team. Are we really to believe that the sequence of the setbacks suffered by teams with equal numbers of losses ought to be the lone criterion for crowning a champion? That seems far sillier than letting informed observers consider the available data and make a decision by majority rule.

This argument in favor of settling important questions by voting was brought to you by the Constitution of the United States of America.

Perhaps past performance is not predictive---it certainly wasn't in Oregon State's case---but, if we are going to confine our sample size artificially by reducing to ultimate insignificance the import of the regular season (a sin of which every sport except Division I-A college football is guilty to some degree), at least we ought not to make our shrunken sample any smaller than absolutely necessary.

The College World Series consists of two double-elimination regionals culminating with a three-game series for the title. However, if, as was the case this year, the team winning one regional came through the winners bracket and the team emerging from the other regional got there through the losers bracket, the latter finalist is being given the advantage of having its prior postseason loss absolved.

While I would have preferred it had they adopted a different tone, my critics are correct that this is the system that we have. My point was, and remains, that, while the structure they have described is the case, it should not be.

Go 'Dawgs!