CARSON, CA - JULY 23: Heath Pearce #3 of Chivas USA knows the secret. Do you?(Photo by Stephen Dunn/Getty Images)
In the past six games, Chivas USA have gone 2-3-1. While their performance on the pitch has been inconsistent, I found a stat that has been dead on when determining the result Chivas will take away from a match. Using a standard statistic tool called the Correlation Coefficient (aka Pearson r test) I found a score of .862 indicating a very strong positive correlation (a correlation value of 1 means there is a perfect correlation).
So what is this telling statistic? It's not goals, saves, and certainly not possession. No, it's clearances. Not a exciting, rah-rah stat by any means, but when the Goats clear the ball, they win. Or at least they don't lose. More after the jump.First, let's make sure we're talking the same language. From the Opta definitions page, clearances are defined as:
This is a defensive action where a player kicks the ball away from his own goal with no intended recipient of the ball.
The graph below is put together using data obtained from the Opta chalkboards provided in the Matchcenter following each Chivas USA match.
As you can see, losses (Philadelphia, Colorado and Chicago; 13,15,14 respectively) are the lower numbers.
Whereas clearance totals in the teams' wins and draws are twice to three times the clearance totals from losses.
These numbers make sense. Against Toronto and San Jose, Chivas was happy to take the lead and cede possession.
The counts are lower in the Portland match probably because Chivas took the lead in the late minutes left of the game.
Let's recognize this for what it truly is--an unusual trend but ultimately still just a trend.
It's good to remember that just because two things might occur together, that doesn't necessarily mean that one causes the other. Statisticians will say correlation does not mean causality. For example, ice cream sales and shark attack frequency are strongly correlated. Obviously This is not because sharks start attacking in response to ice cream, but because the two variables exhibit a common response to the warm season.