Jun 16

The Mathletics Behind Playoff Upsets, Part 1

Editor’s Note: This article was originally published on June 2014. The commentary generally considers that timeframe for its writing content. However, the date tables and any directly-rated accompanying information in the commentary was updated at the end of the calendar. This will continue to happen at the end of each year, in order to properly reflect the current statistical trends.

Welcome to June, sports fans, a time where the best of the NBA and NHL finish their respective battles for the championship. For those who love variety, June may be one of the best sports months of the year. While the MLB starts to see some separation among teams as the league hits its middle third of the season, the NBA and NHL seasons reaches its respective championship climaxes.

However, the two respective title hunts have great variety of events on their own, as evidenced by history and probability theory. The Stanley Cup Playoffs gained a cult following in the past few decades for its wild upsets and its “anything can happen” overtime battles. Meanwhile, the NBA Playoffs earned its practice in predictability, as the strongest franchises and best teams generally go unscathed until the Conference Finals at the earliest. It definitely makes for interesting narratives in multiple ways.

This year, we saw some of those differences. The Western Conference’s San Antonio Spurs dominated en route to an NBA Championship. Meanwhile, the third-place team from the Pacific Division made waves, as the Los Angeles Kings rekindled their road dominance from two years ago to win the Stanley Cup championship. The striking contrasts between the playoff histories bring up a compelling question: Are there any tangible reasons for why there are so many upsets in the Stanley Cup Playoffs and so few upsets in the NBA Playoffs? That question brings up a two-part mathematical investigation.

Playoff Upsets: The Big Four (MLB, NBA, NFL and NHL)
In so many ways, a league’s playoff history can tell so much about the sport itself on an American professional sports scale. The stories that come from the championship pursuit at its highest moments permeate through the respective leagues’ histories. It also can affect the league’s television ratings, competitive balance and small-market legitimacy. Let’s break down each league’s playoff history.


Limited Seeding History Shows MLB Zaniness
Baseball in many ways may be the quirkiest sport created in America. Play-by-play scoring aspects of the game allow the stat-heads to flock to the game, especially in the form of sabermetrics. Meanwhile, in a game where the hitters can be penalized for making good contact to the wrong area of the field and pitchers can be penalized for striking out too many batters, many games can be at the mercy of “luck” or random chance. This in turn could skew the validity of said sabermetrics and analytics, making jobs challenging for front offices.

Perhaps the zaniness fits perfectly with the league’s playoff system. After a 162-game regular season, only 10 teams will make the postseason this year. And this is only after the league recently adopted a one-game wild card playoff to add more teams to what was an eight-team format. All that work on a day-to-day basis, only for a third of the teams to get a puncher’s chance of winning the World Series.

Of course, the league is probably just trying to protect the teams that played so well over a six-month span. If the league was to expand beyond 10 teams in the postseason, it might be unfair to the top teams that earned their playoff spots. That at least would seem to be case. We look at the league’s playoff history since the league adopted the wild card format in 1994 (the first such postseason occurred one year later). Check out the distribution of pennant winners and World Series champions among the seeds since expansion.

Table 1: World Series Championships by Seed (since 1995)

Years 1 2 3 4 5
2012-2015 2 0 1 0 1
1995-2011 6 4 2 5
Total 8 4 3 5 1
Percent 38.1 19.0 14.3 23.8 4.8

Table 2: League Pennants by Seed (since 1995)

Years 1 2 3 4 5
2012-2015 3 0 3 1 1
1995-2011 11 8 5 10
Total 14 8 8 11 1
Percent 33.3 19.0 19.0 26.2 2.4

Clearly the wild card brings something wild to the playoffs. Wild card teams are 6-6 in the World Series, better than the two seeds (4-4) and three seeds (3-5). Thus, the wild card teams only trail the top seeds (8-6).

For those that may be thinking that this means that teams should be seeded simply by record, not based on winning divisions, it’s important to note that not all wild cards were necessarily part of a stacked division. The 2011 St. Louis Cardinals finished with the fourth-best record in the NL before winning the World Series. The 2002 Anaheim Angels and 2003 Florida Marlins both finished with the third-best record in the NL before winning the World Series. Only the 1997 Florida Marlins and 2004 Boston Red Sox finished with the second-best in their league before winning the World Series.

With that theory out of the way, it shows that something about baseball in itself can bring a factor of unpredictably in any single game. No wonder they need 162 games to determine who deserves to make the playoffs. One game doesn’t do much at all to tell you about a baseball team, given all the context and seemingly random opportunities that can affect a single game.

1 ping

  1. Evaluating Playoff Formats in the Big 4 | Curveball Sports

    […] roster turnover creating parity, the sport itself has high variability at the professional level. Only baseball has higher playoff parity, which speaks to the nature of hockey that the best team doesn’t always win. This data may […]

Leave a Reply

%d bloggers like this: