#### Network News

X My Profile
View More Activity

# Do Streaks Really Matter?

By Stephen L. Carter

A lot of sportswriters this week have been analyzing the “implications” of the fact that the Redskins’ opponent Sunday, the Houston Texans, last weekend broke a six-game losing streak against the Colts, beating Indy for only the second time in franchise history.  The brouhaha reminded me a bit of the angst around Washington last year, when the Skins lost to Detroit for the first time in a gazillion years.

I mention this because, as any number of scholars have argued, there is a good deal less to “streakiness” than most people seem to imagine.  All of us have watched our favorite teams go through periods when they seem to win them all or lose them all.  What are we to make of this?  Does doing well one week imply that you are likely to do well the following week?  Does doing well two weeks in a row imply that you are likely to do well the third?

The answer most studies give is No.  To understand why, take the simple example of a game of luck rather than skill – flipping a coin.  Every time we flip, the odds of a heads are one out of two.  If we flip heads ten times in a row, that does not change the odds that the next flip will be a heads.  (Statisticians call this “independence”: the result of one coin flip has no effect on the next.)

How does this apply to sports teams generally – and to the Redskins in particular?  The phenomenon of winning streaks and losing streaks has been studied to death, using every statistical tool available, and the result seems to be this: streaks do exist, but they are random.  Thus, being on a streak does not give you any information about the outcome of the next game.

To see why this is so, imagine a team that is good enough to beat two-thirds of its opponents.  In the National Football League, this would lead to a record of 11-5.  How is this record compiled?  Is it likely that team will go win-win-loss, win-win-loss, win-win-loss?  Of course not, any more than, if it takes me an average of 20 minutes to drive to work (which it does), I am likely to take exactly that long every day.  One day it will be 18, another it will be 22.  Similarly, a football team that wins two-thirds of its games will likely win (and lose) in a random order, such as W-L-W-W-W-L-W-L-W-W-W-W-W-L-W-L.  An observer might wonder about the significance of the team ending with a loss after its impressive mid-season five-game win streak.  But the streak is likely an anomaly.  A team that will win two-thirds of its games has to win them in some order or other;  this is just the order in that particular year.

While streaks are fun to watch, the principal information they give is only whether the streaking team is a good one or a bad one.  Thus, the Redskins lost to Detroit last year for the same reason that the Redskins lost to 11 other foes last year: they had a bad team.  And a streak of one game (Houston beating Indianapolis, Washington beating Dallas) gives us no information at all.

Which, perhaps, is why football is so much fun to watch: whatever happened last week, anything can happen this week.

By Box Seats blogger  |  September 17, 2010; 7:00 AM ET
Categories:  Redskins , Stephen L. Carter  | Tags: Redskins-Texans, Stephen L. Carter
Save & Share:                    Previous: A timely message about female sports fans
Next: For whom the Belanger tolls

Posted by: earthling | September 17, 2010 8:51 AM | Report abuse

Excellent article. Dean Oliver addresses the issue effectively in "Basketball On Paper." A 20-win NBA team is almost certain to win three in a row at some point in a season. Fans and media often view this as a sign that the team is turning things around, or as something promising. Actually, it's just a fluke. Good stuff, Mr. Carter.

Posted by: TheSecretWeapon | September 17, 2010 10:29 AM | Report abuse

Accurate stuff, don't know about good stuff. Of course people like Thomas Gilovich have been making this point for 20-years now. Is Week 2 really the time to bring this up? With so much going on with the Redskins brand new schemes and personnel, and when the season is too early for 'meaningful' streaks, is the interpretation of streaks really a timely topic?

I think every Redskins post so far on Box Seats has been lousy :( (I won't read too much into that streak of lousiness, other than to take it as an indicator that additional posts are likely to be lousy.)

Posted by: SkinsFaninCanada | September 17, 2010 1:13 PM | Report abuse

We encourage users to analyze, comment on and even challenge washingtonpost.com's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.

characters remaining