At the end of a major upset, between calls of the improbable and the miraculous, you will often hear sportscasters say, in an apologetic tone, “well, tonight they were just the better team.” That statement, which is self-evident if you subscribe to the idea that the better team is the one that wins the game, is debated over the water cooler and internet message boards for days afterwards.
If the NFL made their champions play a three-game series, could the Giants have won again? Would David Tyree have made another miraculous catch? Or would the 2007 Patriots be the greatest team of all time? More importantly, which outcome would we have preferred? Why do we care? Do we want to see the team with the best resume from the regular season win it all? Or do we want to see Butler with a chance to win it at the buzzer? With two diametrically opposed systems for determining champions, which is better?
We celebrate championships with parades and legends, rings and wistful stares. Yet we’re never sure if the best team actually won it all. Do these systems work as they’re supposed to? That is what these next several articles are going to examine. Over the next few weeks, I will look at the playoffs of four major American sports, each with its own way of crowing a champion, and look at two things. How predictable is each system (i.e. how often does the team that is “supposed” to win do so?), and whether predictability matters.
There are two basic systems American sports use to determine a champion. One system, favored by the NFL, NCAA basketball, and NCAA football, is the one-game series. The Super Bowl. The (now thankfully defunct) BCS National Championship. March Madness. One game, with all the randomness, luck, and glory that entails. Other sports, such as the MLB, NBA, and NHL, have a slog of a playoffs. The NBA and NHL playoffs take over a month to play, as does the MLB. Five or seven-game series, over multiple rounds, attempt to eliminate the luck and variables that lead to inferior teams like the New York Giants stunning the world.
The NCAA basketball tournament is predicated on winning a series of single-game playoffs on the way to a championship. This is partly due to the demands of the sport, and partly to increase drama. The NFL playoffs, for instance, would add an additional half-season to the sport if it were merely to operate as three-game series. The NCAA tournament, which is three weeks long, would be similarly expanded if each round included a three-game series. There is also the limits that can presumably be placed on young men who are technically amateurs and student-athletes, however ridiculous that designation is. So, the limits of time dictate the format. That format does, however, have the reputation of providing for a more exciting playoffs. The perception of the NCAA tournament is that there are more upsets and inexplicable victories. It is not called “March Madness” for nothing, after all. There is some basic truth to this. The more games played between two teams, the more likely it is for overall talent or preparedness to win out. Fluke performances retreat to the mean; bad days by star players are redeemed by good days. In the most recent NBA playoffs, Steph Curry (perhaps the best shooter in basketball history) had his worst day shooting the ball in his career in Game Two. The Warriors lost that game to the Cavaliers.(1) Had that been the only game the two teams played, Curry’s incredibly bad game would be legend. Instead, the teams played four more games, Curry had several good performances, and the Warriors are NBA champions.
In the history of the NCAA men’s basketball tournament, No. 1 seeds have won the required four games to advance out of their bracket 51 times since the NCAA tournament went from 48 to 64 teams in 1985. There have been 140 bracket winners overall, meaning that less than 37% of regional champions were those teams expected to do so. That would certainly give credence to the notion that the NCAA tournament is prone to upsets, if we only counted the four No. 1 seeds as expected to win. If we expand the “supposed” to win to No. 2 seeds, as the national media is wont to do, and assuming that those seeds are nationally ranked in the top ten, the unpredictability of the tournament falls apart. Since 1985, 47 teams outside the No. 1 and No. 2 seeds have made the Final Four. Seeds 3-16 make the Final Four 34% of the time; seeds 1-2 make it 66% of the time. No. 1 seeds have a higher chance of making the Final Four than seeds 3-16 taken together. A No. 1 or No. 2 seed is twice as likely to win their four games than the other seeds combined. Now, that obviously has something to do with talent disparity, but it is simply not true that “lesser” teams win. Seven times in 35 years has a team that can reasonably be considered an underdog won the entire tournament. Oddly enough, the most famous of these teams, Villanova, won as a No. 8 seed in 1985, culminating in a victory against No. 1 Georgetown called by loyal Nova fans “the most perfect game ever played.
I could devote plenty of time to the madness itself, which typically occurs in the first weekend. Some No. 2 seed gets lazy, a 15 seed can’t miss, and Duke (hilariously) loses in one of the first two rounds every few years. The website diddukewin.com exists primarily for the delight of the majority of basketball fans for whenever Duke loses, especially to a school like Mercer in 2014. But that madness is simply a distraction. Duke has won five NCAA championships. By the second weekend of the tournament (the Sweet Sixteen and the Elite Eight), the teams that win are overwhelmingly blue blood teams from major conferences that are more talented and wealthier, and thus able to field better teams. The question of whether the NCAA tournament really produces the madness it advertises depends on how you frame the question. There are plenty of upsets on the first weekend of the tournament. Blue bloods that were looking ahead to the potential Sweet Sixteen matchup lose to Florida Gulf Coast in a flurry of alley-oops. Duke forgets to play defense against Mercer. But once the tournament moves beyond, inevitably a top team stands victorious. The NCAA tournament is a cruel beast, but it does reward those teams that were deemed a top-10 or so team. The NCAA tournament has the best of both worlds. It offers a glorious orgy of upsets and madness on the first weekend, while forcing teams down a gauntlet that usually leads to a team that was considered a contender cutting down the nets. The NFL playoff system seems like it is designed to test the power of luck and chance.
(1) Curry shot 5-23. Thats 21%. He shot over twice as well, 48%, for the entire season.