In part one of this post, I looked at the NCAA Basketball tournament and the NFL playoffs, two systems that, by their nature, open the door for a more random outcome. The single-game elimination standard of those two playoff systems have given them the reputation as more exciting and unpredictable than other playoff systems. One does not gain the name “March
The NFL is no different. Had this article been written two years ago, before my beloved Seahawks destroyed the Broncos in a surreal three hour scream of pure delight unmatched in human history (at least according to me), the NFL was a perfect case for parity and uncertainty. The same number of six seeds had won a Super Bowl since ‘02 as number one seeds. But after the last two Super Bowls, which featured pairings of number one seeds, those top teams have a 2:1 advantage over the underdogs. Still, seeds other than the top seed have won more often. But we may be entering a new era of dominant teams in the NFL.
Here, in part three, we take a look at the MLB playoffs. The MLB, which used to be the purest of playoff formats (excuse me while I sound like a grandpa from the Midwest for a moment here), where the regular season winners of each league met in the World Series, now has an interesting amalgam of formats. In the current system, there are three divisions in each league, and five teams from each league make the playoffs: the three division winners, and two “wild cards,” the next best two teams by record. Those two teams play a one-game playoff for the right to play in the Divisional Series, a five-game series against the top-seeded team, while the other two division winners play each other. The winners of each of those series play in the Championship Series, a seven-game playoff. After that is the World Series, another seven game playoff. Here, a baseball team which has played 162 games to reach the playoff will have to win at least eleven games, and twelve if they reach the playoffs as a wild card.
The point of long series is to reduce the effects of anomalies. In part one, we talked about the way that multiple games reduce statistical outliers. I used the example of Steph Curry’s shooting in these latest NBA Finals to make my point, but in baseball, the effects can be even more extreme. In baseball, a huge portion of a team’s overall chances for victory rest on the arm of their starting pitcher. Some teams have one incredible pitcher, some teams have several. If we played one-game series, depending on how much rest we gave teams between series, a team could win based almost solely on the dominance of their ace. Instead, teams are forced to use typically their four best starting pitchers, and give their ace two starts a series. If the series comes down to game seven, managers must make decisions on pitcher usage that will affect the next series. Baseball’s seven-game series reduce the effects of one legendary performance, compared to, say, the NFL. No one would remember Madison Bumgarner’s performance in Game Seven of 2014’s World Series had the Giants’ other pitchers not gotten them to Game Seven.
Seven times since 1995 has the overall league leader (the top seed from the AL and the NL) won the World Series. Seven victories for the top seed out of the twenty World Series played. That stands at over one-third. If all things were equal, the top seed should win one-fourth of the time, until the last three seasons, when that equal probability drops to one-fifth. Conversely, wild card teams have won six times. Teams that won their division but did not finish in the overall top seed won seven times. So, baseball’s efforts to prioritize winning a division do not seem to be working.
Granted, one of those six wild card victories came last year, when there were more wild card teams than top teams, but there was a run in the early 2000s when wild card teams won three years in a row. The lesson from this seems to be that you should try to either be the best team in your league, or play that best team in the short series. Wild Card winners play the top seed in the Division Series, which is five games. Those two fewer games may go a long way to deciding why so many Wild Card teams have won it all. They have the opportunity to play the best team in their division in a series where luck and anomaly can manifest themselves, while the two division winners do not get the chance to face the top team in that way. This may point to why the winner of the critical five-game series has taken home the lion’s share of World Series trophies since the Division Series began in 1995. Either way, the MLB playoffs invite their fair share of randomness, while rewarding top teams in a way that helps their cause.
In part four, I will look at the NBA playoffs, which have the reputation for being the least random of all the playoff systems. Look for it next Tuesday!