Surprise, ESPN’s historical “statistics” can be misleading!

The Dallas Mavericks lost Games One and Three of the NBA Finals, and yet won the NBA title. The Boston Red Sox lost their first six games of the season, and yet they now appear to be a likely playoff participant (Coolstandings: 80%). Each turn of events has brought a smile to my face because each has cut against one of the more annoying practices at ESPN — the naked historical citation.

It drives me wacko when ESPN tries to overdramatize the significance of certain outcomes by providing historical statistics without also providing context for those citations. 

For example before and after both Game One and Game Three of any given playoff series, you can bet your newborn’s college fund that Stuart Scott will mention that the winner of Game One goes on to win the series 80% of the time, and that the winner of Game Three does something similar.  And, you can also be sure that in the first few weeks of a baseball or football season (but, curiously, no other sport), if a “favorite” team loses their initial “x” number of games, ESPN will waste time digging up statistics showing how few teams have reached the playoffs or won the championship after such a start. This, you see, artificially enhances the significance of early contests, whether early in a given season, or early in a given playoff series.  Whereas most fans would probably conclude that the losing team in each case still had ample opportunity to recover, ESPN provides historical numbers to try to dissuade such thinking.

The figures they use are accurate, I don’t contend they are not.  But the way ESPN presents the figures is misleading because they present them without contextual explanation. 

For instance, the reason most teams that win Game X or Game Y of a series go on to win the series is, often, they are simply the better team.  A fair number of NBA playoff series throughout the Association’s history have been lopsided affairs.  But the Miami-Dallas contest was a fairly even match. So yes, the likelihood of ultimate success diminished with each loss by either team, but given the even nature of the contest, the impediment caused by the loss was not as great as ESPN suggested.

And the “beginning of the season” statistics are even more misleading, especially when applied to a team with obvious talent.  I would venture to guess that nearly 70% of the teams in history that have started a Major League or NFL season with 3 consecutive losses have sucked (less so in baseball, I guess, but the baseball citation is clearly the more preposterous given the enormous number of remaining games).  Most of those teams had little or no chance to qualify for the playoffs or win a championship, whether they started 0-3 or 3-0.

But that should not persuade one to conclude that a distinctly stronger team in the same situation would face the same low odds. It wouldn’t. If most houses in town are made of straw, then one can conclude that most of the houses in town will not survive a windstorm. But that does not make it more likely that the few houses made of brick will suffer the same fate. 

Of course, making such a distinction neccesarily involves a certain amount of dispassionate, critical thinking on a person’s part, something ESPN tries to discourage.  Its bad for ratings.

Advertisements

One Response to “Surprise, ESPN’s historical “statistics” can be misleading!”

  1. Dre Says:

    Ty,
    Another great part of these facts is some built in stats that get left out. If I say “No title team has ever started 0-20” (Haven’t fact checked that but pretty sure it’s true) What I am leaving out is that winning a title is rare. By default on 3-5% of the league wins a title each year (at least since expansion to 20+ teams) So every year we pile on a bunch of hollow stats and obviously most places don’t list these.

    Also each year is independent of the last. While the odds may be impressive it’s hard to say if they mean anything or if they still mean the same thing (e.g. the number of teams that made more than 100 three pointers SPIKED in 1980 🙂 )

    In the end they are fun stats to throw out but like most numbers they are thrown out without context and used incorrectly.

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


%d bloggers like this: