Bandwagoning, Gambling, and Information Cascades in the NBA

Anyone who follows professional sports is sure to be well acquainted with the term “bandwagoner.” Those who have always loved one team and could never conceive of changing sneer at those who so willingly change their sports allegiances. There’s something that seems almost dishonest, untrustworthy, and confusing about such behavior. But perhaps the key to understanding the huge jumps in team popularity from year to year lies in the vocabulary of information cascades.

Let us use the National Basketball Association as our example. We’re right in the thick of the NBA playoffs, with four teams remaining in the running for the Championship. The two most interesting — and most talked-about — teams, however, are the Golden State Warriors and the Cleveland Cavaliers. Each of these teams have had a different journey toward the playoffs in terms of public perception and support.

The Cavs opened the season back in October as favorite to win the NBA title — heavy favorites in fact. In fact, at one point, the Warriors weren’t even the second-favorite. This is odd, considering the fact that the Warriors beat the Cavs in pretty convincing fashion in the finals and neither team made significant roster changes in the offseason. This public opinion denied most of the hard-and-fast statistical comparisons of the teams. So why? I claim that this is entirely due to an information cascade based on the public appeal of sports megapersona LeBron James. James is perhaps the most famous (or infamous) figure in American sports today, and in the age of Twitter, I feel like that sheer fame alone was enough to spark an information cascade in favor of the Cavaliers. To the casual sports fan, enough of a Twitter onslaught over the summer about James and the Cavaliers would be more than enough to move their hand and change how they click in the polls.

Another interesting and more acute manner in which this effect manifested was after game 1 of the Western Conference Finals this year. The Warriors were playing the Oklahoma City Thunder, and the Warriors were heavy favorites to win the game, the series, and the eventual title. But the Thunder came away with victory in game 1, and overnight the Warriors’ odds in Vegas for winning the title changed from -160 to +110. Sure, a one-game advantage towards the Thunder was significant in the best-of-seven series, but surely there must be more of an explanation for the huge change in odds after just the first game? I argue that an information cascade also explained this result — the internet bandwagoners wishing to be sages, dying to say “I told you so,” thirsty for some excitement in these relatively bland playoffs.

One of the most interesting ways to explore this phenomenon is using FiveThirtyEight’s team-by-team forecast tool, which tracks predictions for each teams’ success based on polls and given data. If you look at teams like the Cavs and Warriors and how they fluctuate week-to-week  or over the course of the season, you can see the effect of these information cascades. Granted, as time passes, there are fewer games left and thus less uncertainty. However, even just looking at the month leading up to the playoffs, there are interesting changes in projections to consider. The dialogue between true statistical likelihood and mass movements is really interesting.

So maybe we should give the bandwagoners an easier time. They’re only doing what seems rational.

Extra Links:

Thunder vs. Warriors: Point Spread, Total & Prediction Game 2

http://www.oddsshark.com/nba/nba-futures

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s