NCAA EXPANSION AND THE SELECTION OF WORTHY COLLEGE BASKETBALL CHAMPIONS

Since 1939, the NCAA has conducted a men’s basketball national championship. The format has been a single-elimination tournament—a form of Russian roulette—and the tournament has expanded during two distinct eras of its history.

The tournament started with 8 teams in 1939, doubled to 16 teams in 1951, and then expanded to  22 teams in 1953. Then growth stopped. For more than two decades, the tournament remained stable in the 22 to 25-team range. Only conference champions got invited, major conference champions getting a bye to the regional semifinals.

With burgeoning television interest in the tournament, however, the tournament went through a 12-year period inflation from 1974 to 1985. During those 12 years, the bracket exploded from 25 teams (1974) to 64 teams (1985). It has remained steady since.

If we assume that one goal of a national championship is to crown the best team in the nation, we should ask ourselves a question. How good a job does the NCAA men's basketball tournament do at fulfilling this goal? Does the tournament tend to crown the best team? Or is the tournament more of an exciting crap shoot that is won by the luckiest team … or the hottest team?

To analyze this question, first we have to determine which teams were the best in the country going into each tournament.

The best indicator we have, of course, is the Associated Press men’s basketball poll. The AP poll has been around for 62 years. At the end of each regular season since 1949, the sportswriters have issued a final poll, anointing a team that they believed was the best team in the nation based on an entire season’s performance.

Assuming that the AP is a good indicator of overall performance, how good a job does the NCAA tournament do at crowning the best team?

Well, in 62 years, the AP number 1 team has won the championship 20 times (32%), and a team ranked in the AP’s top-3 has won 42 times, more than two-thirds of the time. Considering the single elimination format, that’s not a bad track record.

But the success of the NCAA tournament at selecting the season’s best as champion has fallen on hard times. In the last 26 years, the number 1 team has won only three times—Duke in 1992 and 2001 and UCLA in 1995. The success of the tournament at crowning the best team, it seems, is directly tied to the size of the tournament. The larger the bracket has grown, the more arbitrary the results have become.

To see the effect of bracket growth on the success rates of the best teams, we can divide the tournament’s history into two eras of 31 years … covering the entire 62 years that the AP has been around. Conveniently, these two eras divide at 1979-1980, the middle years of the 12-year expansion from 25 to 64 teams. In those two years, the tournament invited 40 and 48 teams—the perfect dividing line between small tournament and large tournament.

In the first 31-year era, the tournament installed the number 1 team as national champion a whopping 16 times, more than half the time, and installed a top-3 team 26 times! The small tournament, in other words, did a fantastic job of crowning a team that was thought of as elite during that season.

In the second 31-year era, however, the number 1 team won the title only 4 times, a dismal 13% rate, and an AP top-3 team won just 16 times. During that second era, in fact, 6 champions finished the regular season out of the AP top-10, and 2 of them, Villanova in 1985 and Kansas in 1988, were not ranked at all!

Part of this, of course, is attributable to the “P” word. There is simply more talent out there, spread among more and more teams. A single elimination tournament gives that talent a second bite at the apple—a chance to do in the tournament what it could not do during the regular season. The result is stunning upset after stunning upset, which, of course, just makes for better television.

Please understand, I am not advocating retrenchment to 25 teams or saying that this trend is bad. My team, Duke, was ranked number 6 when it won its first title in 1991, and I’m not giving the trophy back! I am just pointing out that, as the tournament has expanded, it has becomes a less reliable means of crowning, as kings, the very best of the battle-tested knights who enter the fray.

Perhaps this should be of some solace to teams like Kansas this year. Kansas finished a bloody 34 game-campaign deservedly ranked number 1 in the nation and then tripped over an Egyptian name. Kansas should take great pride in its conference titles and in its season-ending number 1 ranking. In a very real sense, those are far better indicators of team success.

Should be proud of a 32-7 national title in 1991, but ashamed of a 32-4 Sweet Sixteen exit in 2006? I don’t think so.